From 29d763e27a7434576b9b54cd3ee803ab899af7af Mon Sep 17 00:00:00 2001 From: Peter Carbonetto Date: Mon, 11 Mar 2019 13:45:17 -0500 Subject: [PATCH] Re-built pkgdown. --- docs/LICENSE.html | 22 +- docs/articles/eQTL_outline.html | 56 +- docs/articles/flash_mash.html | 84 +- docs/articles/index.html | 34 +- docs/articles/intro_correlations.html | 40 +- docs/articles/intro_mash.html | 110 ++- docs/articles/intro_mash_dd.html | 26 +- .../figure-html/unnamed-chunk-11-1.png | Bin 54499 -> 54563 bytes .../figure-html/unnamed-chunk-12-1.png | Bin 26710 -> 26714 bytes docs/articles/intro_mashcommonbaseline.html | 47 +- docs/articles/intro_mashnobaseline.html | 39 +- docs/articles/mash_sampling.html | 32 +- .../figure-html/unnamed-chunk-5-1.png | Bin 121683 -> 124501 bytes .../figure-html/unnamed-chunk-6-1.png | Bin 125623 -> 125679 bytes docs/articles/simulate_noncanon.html | 26 +- docs/authors.html | 48 +- docs/index.html | 76 +- docs/pkgdown.css | 50 +- docs/pkgdown.js | 160 ++-- docs/pkgdown.yml | 14 - docs/reference/calc_lik_matrix.html | 30 +- .../reference/calc_lik_matrix_common_cov.html | 30 +- docs/reference/calc_lik_vector.html | 30 +- docs/reference/calc_relative_lik_matrix.html | 30 +- ...compute_alt_loglik_from_matrix_and_pi.html | 30 +- .../compute_loglik_from_matrix_and_pi.html | 30 +- .../compute_null_loglik_from_matrix.html | 30 +- .../reference/compute_posterior_matrices.html | 30 +- ...mpute_posterior_matrices_common_cov_R.html | 30 +- .../compute_posterior_matrices_general_R.html | 30 +- docs/reference/compute_posterior_weights.html | 30 +- .../compute_vloglik_from_matrix_and_pi.html | 30 +- docs/reference/contrast_matrix.html | 30 +- docs/reference/cov_all_zeros.html | 30 +- docs/reference/cov_canonical.html | 30 +- docs/reference/cov_ed.html | 32 +- docs/reference/cov_equal_effects.html | 30 +- docs/reference/cov_first_singleton.html | 30 +- docs/reference/cov_from_factors.html | 30 +- docs/reference/cov_pca.html | 30 +- docs/reference/cov_simple_het.html | 30 +- docs/reference/cov_udi.html | 32 +- docs/reference/cov_udi_single.html | 30 +- docs/reference/ed_wrapper.html | 30 +- docs/reference/estimate_null_correlation.html | 59 +- .../estimate_null_correlation_simple.html | 42 +- docs/reference/expand_cov.html | 30 +- docs/reference/extreme_deconvolution.html | 320 ++++++++ docs/reference/get_estimated_pi.html | 30 +- docs/reference/get_log10bf.html | 30 +- .../get_n_significant_conditions.html | 30 +- docs/reference/get_pairwise_sharing.html | 30 +- .../get_pairwise_sharing_from_samples.html | 30 +- docs/reference/get_samples.html | 40 +- docs/reference/get_significant_results.html | 30 +- docs/reference/index.html | 755 +++++++++--------- docs/reference/initialize_pi.html | 30 +- docs/reference/is_common_cov_Shat.html | 30 +- docs/reference/is_common_cov_Shat_alpha.html | 30 +- docs/reference/make_names.html | 30 +- docs/reference/mash.html | 45 +- docs/reference/mash_1by1.html | 32 +- docs/reference/mash_compute_loglik.html | 30 +- .../mash_compute_posterior_matrices.html | 30 +- docs/reference/mash_compute_vloglik.html | 30 +- docs/reference/mash_plot_meta.html | 30 +- docs/reference/mash_set_data.html | 34 +- docs/reference/mash_set_data_contrast.html | 30 +- docs/reference/mash_update_data.html | 38 +- docs/reference/optimize_pi.html | 32 +- docs/reference/posterior_cov.html | 30 +- docs/reference/posterior_mean.html | 30 +- docs/reference/posterior_mean_matrix.html | 30 +- docs/reference/scale_cov.html | 30 +- docs/reference/sim_contrast1.html | 30 +- docs/reference/sim_contrast2.html | 30 +- docs/reference/simple_sims.html | 30 +- docs/reference/simple_sims2.html | 30 +- docs/reference/udi_model_matrix.html | 30 +- 79 files changed, 1457 insertions(+), 2278 deletions(-) delete mode 100644 docs/pkgdown.yml create mode 100644 docs/reference/extreme_deconvolution.html diff --git a/docs/LICENSE.html b/docs/LICENSE.html index 51cab6f..3fc9bf5 100644 --- a/docs/LICENSE.html +++ b/docs/LICENSE.html @@ -86,25 +86,23 @@

License

-
BSD 3-Clause License
-
-Copyright (c) 2017, Matthew Stephens
-All rights reserved.
+
Copyright (c) 2017, Gao Wang, Sarah Urbut & Matthew Stephens.
 
 Redistribution and use in source and binary forms, with or without
 modification, are permitted provided that the following conditions are
 met:
 
-* Redistributions of source code must retain the above copyright
-  notice, this list of conditions and the following disclaimer.
+    Redistributions of source code must retain the above copyright
+    notice, this list of conditions and the following disclaimer.
 
-* Redistributions in binary form must reproduce the above copyright
-  notice, this list of conditions and the following disclaimer in the
-  documentation and/or other materials provided with the distribution.
+    Redistributions in binary form must reproduce the above copyright
+    notice, this list of conditions and the following disclaimer in
+    the documentation and/or other materials provided with the
+    distribution.
 
-* Neither the name of the copyright holder nor the names of its
-  contributors may be used to endorse or promote products derived from
-  this software without specific prior written permission.
+    Neither the name of the University of Chicago nor the names of its
+    contributors may be used to endorse or promote products derived
+    from this software without specific prior written permission.
 
 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
 "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
diff --git a/docs/articles/eQTL_outline.html b/docs/articles/eQTL_outline.html
index 8a6c681..8a6498b 100644
--- a/docs/articles/eQTL_outline.html
+++ b/docs/articles/eQTL_outline.html
@@ -8,8 +8,8 @@
 eQTL analysis outline • mashr
 
 
-
-
+
+
 
 
 
 
 
-    

+    

       

-  

+  

     

       
eQTL analysis outline

                         
Matthew Stephens

             
-            
2018-11-08

-      
-      Source: vignettes/eQTL_outline.Rmd
-      
-
-    

+            
2019-03-11

+          

 
     
     
+

 

 

 Introduction

@@ -141,9 +134,9 @@ 

 
U.c = cov_canonical(data.random)
 m = mash(data.random, Ulist = c(U.ed,U.c), outputlevel = 1)

 
#  - Computing 5000 x 241 likelihood matrix.
-#  - Likelihood calculations took 0.16 seconds.
+#  - Likelihood calculations took 0.14 seconds.
 #  - Fitting model with 241 mixture components.
-#  - Model fitting took 1.30 seconds.

+#  - Model fitting took 2.10 seconds.

 
 

 

@@ -151,24 +144,25 @@ 

 
Now we can compute posterior summaries etc for any subset of tests using the above mash fit. Here we do this for the strong tests. We do this using the same mash function as above, but we specify to use the fit from the previous run of mash by specifying
g=get_fitted_g(m), fixg=TRUE. (In mash the parameter g is used to denote the mixture model which we learned above.)

 
m2 = mash(data.strong, g=get_fitted_g(m), fixg=TRUE)

 
#  - Computing 1428 x 241 likelihood matrix.
-#  - Likelihood calculations took 0.04 seconds.
+#  - Likelihood calculations took 0.06 seconds.
 #  - Computing posterior matrices.
 #  - Computation allocated took 0.01 seconds.

 
head(get_lfsr(m2))

 
#               condition_1  condition_2  condition_3  condition_4
-# effect_13096 1.381058e-05 7.503454e-01 6.874052e-01 6.678948e-01
-# effect_29826 5.177092e-05 7.346448e-01 6.923586e-01 7.622170e-01
-# effect_14042 5.669400e-02 1.529967e-02 1.311803e-02 4.596554e-02
-# effect_12524 5.386482e-01 6.905166e-01 4.978431e-01 2.552465e-05
-# effect_15456 8.662398e-05 5.733642e-01 4.369942e-01 6.176371e-01
-# effect_35844 3.413396e-10 2.143976e-10 9.175308e-09 6.573930e-11
+# effect_13096 1.351029e-05 7.304550e-01 6.633768e-01 6.419718e-01
+# effect_29826 4.749262e-05 6.730004e-01 6.283735e-01 7.309554e-01
+# effect_14042 6.116539e-02 1.556075e-02 1.341717e-02 4.594320e-02
+# effect_12524 5.552693e-01 7.011304e-01 5.154365e-01 2.625457e-05
+# effect_15456 8.487950e-05 5.662427e-01 4.228120e-01 6.055155e-01
+# effect_35844 6.452969e-10 3.804030e-10 1.797445e-08 1.206916e-10
 #               condition_5
-# effect_13096 8.183524e-01
-# effect_29826 6.874560e-01
-# effect_14042 6.019355e-06
-# effect_12524 5.360416e-01
-# effect_15456 5.033481e-01
-# effect_35844 3.296696e-12

+# effect_13096 8.025795e-01
+# effect_29826 6.232960e-01
+# effect_14042 6.049099e-06
+# effect_12524 5.527004e-01
+# effect_15456 4.885029e-01
+# effect_35844 5.432876e-12
+ @@ -209,7 +203,5 @@

- - diff --git a/docs/articles/flash_mash.html b/docs/articles/flash_mash.html index cfb1143..02cc9a6 100644 --- a/docs/articles/flash_mash.html +++ b/docs/articles/flash_mash.html @@ -8,8 +8,8 @@ Using flashr for mashr prior specification • mashr - - + + -
+
-
+

Using flashr for mashr prior specification

Gao Wang and others

-

2018-11-08

- - Source: vignettes/flash_mash.Rmd - - -
+

2019-03-11

+
+

Introduction

@@ -112,7 +105,6 @@

  • (For the time being we remove the additional decomposition on FLASH loading because there are issues needs to be figured out. @yuxin)
    • library(flashr)
-library(mixsqp)
 library(mashr)
     
 my_init_fn <- function(Y, K = 1) {
@@ -185,16 +177,39 @@ 
    
 
    U.f = cov_flash(data.strong, non_canonical = TRUE)
    
 
    # Fitting factor/loading 1 (stop when max factor change is < 1.00e-03):
    
 
    #   Iteration  Max Chg (f)      Objective   Obj Diff
    
+
    # Due to absence of package REBayes, switching to EM algorithm
+# Due to absence of package REBayes, switching to EM algorithm
    
 
    #           1     8.48e-03      -15489.36        Inf
    
+
    # Due to absence of package REBayes, switching to EM algorithm
+# Due to absence of package REBayes, switching to EM algorithm
    
 
    #           2     4.90e-03      -15415.40   7.40e+01
    
+
    # Due to absence of package REBayes, switching to EM algorithm
+# Due to absence of package REBayes, switching to EM algorithm
    
 
    #           3     2.24e-03      -15412.98   2.42e+00
    
+
    # Due to absence of package REBayes, switching to EM algorithm
+# Due to absence of package REBayes, switching to EM algorithm
    
 
    #           4     9.51e-04      -15412.93   5.18e-02
    
 
    # Fitting factor/loading 2 (stop when max factor change is < 1.00e-03):
    
 
    #   Iteration  Max Chg (f)      Objective   Obj Diff
    
-
    #           1     3.33e-01      -15716.96        Inf
    
-
    #           2     7.23e-04      -15429.56   2.87e+02
    
+
    # Due to absence of package REBayes, switching to EM algorithm
+# Due to absence of package REBayes, switching to EM algorithm
    
+
    #           1     3.32e-01      -15716.96        Inf
    
+
    # Due to absence of package REBayes, switching to EM algorithm
+# Due to absence of package REBayes, switching to EM algorithm
    
+
    #           2     1.48e-03      -15429.70   2.87e+02
    
+
    # Due to absence of package REBayes, switching to EM algorithm
+# Due to absence of package REBayes, switching to EM algorithm
    
+
    #           3     1.20e-03      -15416.64   1.31e+01
    
+
    # Due to absence of package REBayes, switching to EM algorithm
+# Due to absence of package REBayes, switching to EM algorithm
    
+
    #           4     1.72e-03      -15416.04   6.06e-01
    
+
    # Due to absence of package REBayes, switching to EM algorithm
+# Due to absence of package REBayes, switching to EM algorithm
    
+
    #           5     4.01e-05      -15415.92   1.15e-01
    
 
    # Backfitting 1 factor/loading(s) (stop when max factor change is < 1.00e-03):
    
 
    #   Iteration  Max Chg (f)      Objective   Obj Diff
    
+
    # Due to absence of package REBayes, switching to EM algorithm
+# Due to absence of package REBayes, switching to EM algorithm
    
 
    #           1     3.91e-04      -15412.95        Inf
    @@ -210,9 +225,9 @@

    Now we fit mash to the random tests using both data-driven and canonical covariances.

    m = mash(data.random, Ulist = c(U.ed,U.c), outputlevel = 1)
    #  - Computing 5000 x 256 likelihood matrix.
-#  - Likelihood calculations took 0.18 seconds.
+#  - Likelihood calculations took 0.16 seconds.
 #  - Fitting model with 256 mixture components.
-#  - Model fitting took 1.45 seconds.
    +# - Model fitting took 2.34 seconds.

    @@ -220,24 +235,25 @@

    Now we can compute posterior summaries etc for any subset of tests using the above mash fit. Here we do this for the strong tests.

    m2 = mash(data.strong, g=get_fitted_g(m), fixg=TRUE)
    #  - Computing 1428 x 256 likelihood matrix.
-#  - Likelihood calculations took 0.04 seconds.
+#  - Likelihood calculations took 0.06 seconds.
 #  - Computing posterior matrices.
 #  - Computation allocated took 0.01 seconds.
    head(get_lfsr(m2))
    #               condition_1  condition_2  condition_3  condition_4
-# effect_13096 1.370473e-05 7.437300e-01 6.793988e-01 6.592581e-01
-# effect_29826 5.040489e-05 7.150092e-01 6.719218e-01 7.521132e-01
-# effect_14042 5.807916e-02 1.537341e-02 1.320160e-02 4.592088e-02
-# effect_12524 5.434980e-01 6.936128e-01 5.029747e-01 2.573779e-05
-# effect_15456 8.611973e-05 5.712820e-01 4.327731e-01 6.141828e-01
-# effect_35844 4.435606e-10 2.638274e-10 1.157848e-08 8.065726e-11
+# effect_13096 1.337460e-05 7.226243e-01 6.539333e-01 6.318229e-01
+# effect_29826 4.626489e-05 6.555198e-01 6.100098e-01 7.208470e-01
+# effect_14042 6.267154e-02 1.562942e-02 1.349911e-02 4.592698e-02
+# effect_12524 5.612236e-01 7.049402e-01 5.217429e-01 2.651821e-05
+# effect_15456 8.418593e-05 5.634123e-01 4.175647e-01 6.012912e-01
+# effect_35844 7.394985e-10 4.355099e-10 2.080568e-08 1.387775e-10
 #               condition_5
-# effect_13096 8.130021e-01
-# effect_29826 6.669676e-01
-# effect_14042 6.026087e-06
-# effect_12524 5.408984e-01
-# effect_15456 4.989776e-01
-# effect_35844 3.936074e-12
    +# effect_13096 7.963808e-01 +# effect_29826 6.048723e-01 +# effect_14042 6.050289e-06 +# effect_12524 5.586740e-01 +# effect_15456 4.831697e-01 +# effect_35844 6.112666e-12 +

    @@ -270,7 +286,5 @@

    - - diff --git a/docs/articles/index.html b/docs/articles/index.html index 73c600b..46a8fa5 100644 --- a/docs/articles/index.html +++ b/docs/articles/index.html @@ -21,19 +21,13 @@ - - - + - - - + + - - - @@ -46,7 +40,7 @@ -
    +
    @@ -91,12 +80,12 @@ -
    -
    -
    -

    Articles

    -
    +
    +

    Articles version 0.2.20.0583

    +
    +
    +

    Vignettes

    The following vignettes provide detailed illustrations of the mashr package, as well as guidance on correct application of the statistical methods.

    @@ -127,8 +116,5 @@

    Vignettes

    - - - diff --git a/docs/articles/intro_correlations.html b/docs/articles/intro_correlations.html index 671d85f..ba74f56 100644 --- a/docs/articles/intro_correlations.html +++ b/docs/articles/intro_correlations.html @@ -8,8 +8,8 @@ Accounting for correlations among measurements • mashr - - + + -
    +
    -
    +

    Accounting for correlations among measurements

    Matthew Stephens

    -

    2018-11-08

    - - Source: vignettes/intro_correlations.Rmd - - -
    +

    2019-03-11

    +
    +

    Introduction

    @@ -100,24 +93,25 @@ 

    U.c = cov_canonical(data.V) 
 m.c = mash(data.V, U.c) # fits with correlations because data.V includes correlation information 
    #  - Computing 2000 x 151 likelihood matrix.
-#  - Likelihood calculations took 0.03 seconds.
+#  - Likelihood calculations took 0.05 seconds.
 #  - Fitting model with 151 mixture components.
-#  - Model fitting took 0.31 seconds.
+#  - Model fitting took 0.43 seconds.
 #  - Computing posterior matrices.
-#  - Computation allocated took 0.04 seconds.
    +# - Computation allocated took 0.01 seconds. 
    print(get_loglik(m.c),digits=10) # log-likelihood of the fit with correlations set to V
    # [1] -16121.1117

    We can also compare with the original analysis. (Note that the canonical covariances do not depend on the correlations, so we can use the same U.c here for both analyses. If we used data-driven covariances we might prefer to estimate these separately for each analysis as the correlations would affect them.)

    m.c.orig = mash(data, U.c) # fits without correlations because data object was set up without correlations
    #  - Computing 2000 x 151 likelihood matrix.
-#  - Likelihood calculations took 0.04 seconds.
+#  - Likelihood calculations took 0.05 seconds.
 #  - Fitting model with 151 mixture components.
-#  - Model fitting took 0.25 seconds.
+#  - Model fitting took 0.41 seconds.
 #  - Computing posterior matrices.
-#  - Computation allocated took 0.02 seconds.
    +# - Computation allocated took 0.04 seconds. 
    print(get_loglik(m.c.orig),digits=10)
    -
    # [1] -16120.3214
    +
    # [1] -16120.32135

    The log-likelihoods with and without correlations are similar here, which is expected since there are no actual correlations in the data.

    +

    @@ -146,7 +140,5 @@

    - - diff --git a/docs/articles/intro_mash.html b/docs/articles/intro_mash.html index 1546b84..a8bf9ee 100644 --- a/docs/articles/intro_mash.html +++ b/docs/articles/intro_mash.html @@ -8,8 +8,8 @@ Introduction to mashr • mashr - - + + -
    +
    -
    +

    Introduction to mashr

    Matthew Stephens

    -

    2018-11-08

    - - Source: vignettes/intro_mash.Rmd - - -
    +

    2019-03-11

    +
    +

    Goal

    @@ -133,11 +126,11 @@

    Having set up the data and covariance matrices you are ready to fit the model using the mash function:

    m.c = mash(data, U.c)
    #  - Computing 2000 x 151 likelihood matrix.
-#  - Likelihood calculations took 0.03 seconds.
+#  - Likelihood calculations took 0.04 seconds.
 #  - Fitting model with 151 mixture components.
-#  - Model fitting took 0.28 seconds.
+#  - Model fitting took 0.41 seconds.
 #  - Computing posterior matrices.
-#  - Computation allocated took 0.02 seconds.
    +# - Computation allocated took 0.01 seconds.

    This can take a little time. What this does is to fit a mixture model to the data, estimating the mixture proportions. Specifically the model is that the true effects follow a mixture of multivariate normal distributions: \(B \sim \sum_k \sum_l \pi_{kl} N(0, \omega_l U_k)\) where the \(\omega_l\) are scaling factors set by the “grid” parameter in mash and the \(U_k\) are the covariance matrices (here specified by U.c).

    Remember the Crucial Rule! This step must be peformed using all the tests (or a large random subset), because this is where mash learns that many tests are null and corrects for it.

    @@ -147,28 +140,28 @@

    You can extract estimates (posterior means and posterior standard deviations) and measures of significance (local false sign rates) using functions like get_pm (posterior mean), get_psd (posteriore standard deviation) and get_lfsr (local false sign rate):

    head(get_lfsr(m.c))
    #          condition_1 condition_2 condition_3 condition_4 condition_5
-# effect_1   0.7561485   0.7705861   0.8052827   0.8111157   0.8189465
-# effect_2   0.7245458   0.6905897   0.7760443   0.6922721   0.7012676
-# effect_3   0.7387366   0.7514411   0.8104817   0.8390483   0.8501608
-# effect_4   0.7855697   0.8454482   0.8360483   0.8660092   0.8503405
-# effect_5   0.8044543   0.8371045   0.8808436   0.8785908   0.8758300
-# effect_6   0.7207011   0.6833614   0.7894066   0.7002700   0.7569387
    +# effect_1 0.7561456 0.7705826 0.8052788 0.8111108 0.8189434 +# effect_2 0.7245451 0.6905870 0.7760437 0.6922681 0.7012644 +# effect_3 0.7387334 0.7514373 0.8104771 0.8390450 0.8501580 +# effect_4 0.7855678 0.8454469 0.8360456 0.8660080 0.8503389 +# effect_5 0.8044529 0.8371028 0.8808424 0.8785894 0.8758287 +# effect_6 0.7206969 0.6833571 0.7894009 0.7002643 0.7569353 
    head(get_pm(m.c))
    #          condition_1 condition_2   condition_3  condition_4  condition_5
-# effect_1  0.07387834 -0.12489759 -0.0763244886  0.092632369 -0.045367788
-# effect_2 -0.02353831 -0.17604807 -0.0324369466 -0.201653402 -0.170552895
-# effect_3 -0.10218266  0.18092162 -0.0846330445  0.022524410 -0.002684844
-# effect_4 -0.06238567  0.02794216  0.0654480304  0.008114002  0.032179504
-# effect_5  0.03341379 -0.06059760  0.0001468816 -0.003887884 -0.004665950
-# effect_6 -0.09934657  0.21556659 -0.0791493932  0.213026594  0.079063145
    +# effect_1 0.07387961 -0.12490000 -0.0763267542 0.092636351 -0.045368803 +# effect_2 -0.02353673 -0.17604965 -0.0324344285 -0.201657010 -0.170555034 +# effect_3 -0.10218412 0.18092463 -0.0846362908 0.022525129 -0.002685045 +# effect_4 -0.06238666 0.02794230 0.0654500317 0.008113439 0.032179933 +# effect_5 0.03341429 -0.06059869 0.0001472689 -0.003887683 -0.004665803 +# effect_6 -0.09934878 0.21556970 -0.0791533646 0.213031571 0.079063932 
    head(get_psd(m.c))
    #          condition_1 condition_2 condition_3 condition_4 condition_5
-# effect_1   0.4592843   0.4210491   0.3544632   0.4062381   0.3365901
-# effect_2   0.4861855   0.4277957   0.4046967   0.4447866   0.4149872
-# effect_3   0.4679437   0.5082161   0.3799417   0.3269355   0.3283886
-# effect_4   0.4389144   0.2963991   0.3082056   0.2749155   0.2809661
-# effect_5   0.4033371   0.3286541   0.2608030   0.2597925   0.2656081
-# effect_6   0.5226677   0.4981608   0.4482759   0.4906756   0.3899237
    +# effect_1 0.4592864 0.4210518 0.3544677 0.4062430 0.3365941 +# effect_2 0.4861874 0.4277981 0.4047003 0.4447904 0.4149903 +# effect_3 0.4679459 0.5082177 0.3799465 0.3269399 0.3283928 +# effect_4 0.4389159 0.2964015 0.3082097 0.2749191 0.2809693 +# effect_5 0.4033387 0.3286560 0.2608065 0.2597959 0.2656111 +# effect_6 0.5226697 0.4981630 0.4482802 0.4906797 0.3899275

    Each of these are \(J \times R\) matrices.

    Use get_significant_results to find the indices of effects that are “significant”, which here means they have lfsr less than t in at least one condition, where t is a threshold you specify (default 0.05). The output is ordered from most significant to least significant.

    head(get_significant_results(m.c))
    @@ -213,11 +206,11 @@

    Use get_estimated_pi to extract the estimates of the mixture proportions for different types of covariance matrix:

    print(get_estimated_pi(m.c))
    #          null      identity   condition_1   condition_2   condition_3 
-#  3.711952e-01  2.208546e-01  1.365276e-01  2.098059e-02  1.302892e-08 
+#  3.711932e-01  2.208664e-01  1.365264e-01  2.097969e-02  0.000000e+00 
 #   condition_4   condition_5 equal_effects  simple_het_1  simple_het_2 
-#  6.212090e-10  4.961779e-03  2.264413e-01  1.000873e-04  2.719204e-09 
+#  0.000000e+00  4.961490e-03  2.264380e-01  9.035503e-05  0.000000e+00 
 #  simple_het_3 
-#  1.893879e-02
    +# 1.894453e-02 
    barplot(get_estimated_pi(m.c),las = 2)

    Here we can see most of the mass is on the null, identity, singletons_1 (which corresponds to effects that are specific to condition 1) and equal_effects. This reassuringly matches the way that these data were generated.

    @@ -234,13 +227,13 @@

    Session information.

    print(sessionInfo())
    -
    # R version 3.5.1 (2018-07-02)
+
    # R version 3.4.3 (2017-11-30)
 # Platform: x86_64-apple-darwin15.6.0 (64-bit)
-# Running under: macOS  10.14.1
+# Running under: macOS High Sierra 10.13.6
 # 
 # Matrix products: default
-# BLAS: /Library/Frameworks/R.framework/Versions/3.5/Resources/lib/libRblas.0.dylib
-# LAPACK: /Library/Frameworks/R.framework/Versions/3.5/Resources/lib/libRlapack.dylib
+# BLAS: /Library/Frameworks/R.framework/Versions/3.4/Resources/lib/libRblas.0.dylib
+# LAPACK: /Library/Frameworks/R.framework/Versions/3.4/Resources/lib/libRlapack.dylib
 # 
 # locale:
 # [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
@@ -249,24 +242,21 @@ 
    
 # [1] stats     graphics  grDevices utils     datasets  methods   base     
 # 
 # other attached packages:
-# [1] mashr_0.2.18.0476 ashr_2.2-23      
+# [1] mashr_0.2.20.0583 ashr_2.2-32      
 # 
 # loaded via a namespace (and not attached):
-#  [1] Rcpp_1.0.0        compiler_3.5.1    plyr_1.8.4       
-#  [4] iterators_1.0.10  tools_3.5.1       digest_0.6.18    
-#  [7] evaluate_0.12     memoise_1.1.0     lattice_0.20-35  
-# [10] rlang_0.3.0.1     Matrix_1.2-14     foreach_1.4.4    
-# [13] rstudioapi_0.8    commonmark_1.6    yaml_2.2.0       
-# [16] parallel_3.5.1    mvtnorm_1.0-8     pkgdown_1.1.0    
-# [19] stringr_1.3.1     roxygen2_6.1.0    xml2_1.2.0       
-# [22] knitr_1.20        desc_1.2.0        fs_1.2.6         
-# [25] REBayes_1.3       rprojroot_1.3-2   grid_3.5.1       
-# [28] R6_2.3.0          rmarkdown_1.10    rmeta_3.0        
-# [31] magrittr_1.5      backports_1.1.2   codetools_0.2-15 
-# [34] htmltools_0.3.6   MASS_7.3-50       assertthat_0.2.0 
-# [37] abind_1.4-5       stringi_1.2.4     Rmosek_8.0.69    
-# [40] doParallel_1.0.14 pscl_1.5.2        truncnorm_1.0-8  
-# [43] SQUAREM_2017.10-1 crayon_1.3.4
    
+#  [1] Rcpp_1.0.0        knitr_1.20        magrittr_1.5     
+#  [4] MASS_7.3-48       doParallel_1.0.11 pscl_1.5.2       
+#  [7] SQUAREM_2017.10-1 lattice_0.20-35   foreach_1.4.4    
+# [10] plyr_1.8.4        stringr_1.3.1     tools_3.4.3      
+# [13] parallel_3.4.3    grid_3.4.3        rmeta_3.0        
+# [16] htmltools_0.3.6   iterators_1.0.9   assertthat_0.2.0 
+# [19] abind_1.4-5       yaml_2.2.0        rprojroot_1.3-2  
+# [22] digest_0.6.17     mixsqp_0.1-97     Matrix_1.2-12    
+# [25] codetools_0.2-15  evaluate_0.11     rmarkdown_1.10   
+# [28] stringi_1.2.4     compiler_3.4.3    backports_1.1.2  
+# [31] mvtnorm_1.0-8     truncnorm_1.0-8
    +
    @@ -308,7 +298,5 @@

    - - diff --git a/docs/articles/intro_mash_dd.html b/docs/articles/intro_mash_dd.html index ed30f19..3febc65 100644 --- a/docs/articles/intro_mash_dd.html +++ b/docs/articles/intro_mash_dd.html @@ -8,8 +8,8 @@ Introduction to mash: data-driven covariances • mashr - - + + -
    +
    -
    +

    Introduction to mash: data-driven covariances

    Matthew Stephens

    -

    2018-11-08

    - - Source: vignettes/intro_mash_dd.Rmd - - -
    +

    2019-03-11

    +
    +

    Goal

    @@ -154,6 +147,7 @@

    Session information.

    print(sessionInfo())
    +
    @@ -185,7 +179,5 @@

    - - diff --git a/docs/articles/intro_mash_files/figure-html/unnamed-chunk-11-1.png b/docs/articles/intro_mash_files/figure-html/unnamed-chunk-11-1.png index 05b2ec3684eb842350ecf329e26f2d1b3763ee30..a2bb2e6dcd3126885b748e1daa76b025ff5b7e53 100644 GIT binary patch literal 54563 zcmeFZcR1Gl`#+8(4UO9>vr_g*W_Bq;2q7z#kiGX#r4mwBMwFSo$+)PDB9xhV*tY3ZwG(z4Q4=2o_9*0&4{WeqJ2ZA}bRWiJpA2nPhH z>)oYOr;sSMGiEx~sAkUpz&=Hi>@?X64gS~pC0@O~f|g-&f?X%a@2t!;Mtp7*<}2NP zt$gXKJMTmkMO8(S_i@(cL){YjJfZteceNg8mB?T7y*lonH2G4L@1;pePM_h7`(n)O zE4P_<# zk9HEC0VYPnoP#ZocB`wIJ_<@D3*4uR8S11ePtsn26&wnzufEIbP7`n$uTf{#d&JnQL;%t|^k zdr>#NdrvAVHX)M2rL6LOzS{L0uM zHm67a@UCN~G$V3e+|T0=>hKKdStN}&DIA~{YGqWQn%wYFjoZ1H=WpwsrA3ma@-#rX zi|gIQt4K6bYK@``x#`2T9ISWKbM? zqW{{6ns%&GY(XUdWy5>ahY;+iPx`*=#f|ywcl;#oQRUKX1*TKA7|y3X&@~{J^yCyG zl4L%1{EtrGrW^-h!iP-tY@56F%C~Ctb96DBLMnxW$6`t}8`t&MC%4Y7aBP_-^m`3! z&KKF%YJVxsz4CF&{JNc!I*kR%5Y^nbm(`D}=@)J0hcyq$c!qs`IhM7gC#Ge^eNZMn zeD{Xwy2df3fQ#u%zDs`joaHPYtlKr^4IXtK4IYgi@AZg+MeAdRPy^i7jCT4pE2fJy z+GsZ_m!n0-m@u<+`4dd$lnhk19|Q+fZk=6VIY4S~LFKzoSjOZgRV>ddar%P>jHAh$ zx+rc9!pECAL`#ZTb@gi)>ku}24J$=q)Oqr!r=;&Z4H3MyTCBPF`edl`h^on9HBmM4 zw>hon5BcRKImTHKZEn%Gy|v6$ooF`b^B~e{5wV;kYoK&eXnt{#n{_cmFH_ffRw<#q-yk38#A9o>8^MNcQuPK4Q)8BRa2n@AXBMBh+PaN8{i8aWAcy zdqZu%&|_Nuh<)>_GzOt;6angy#VNPi&K|#cF@vn{YP<|Z=c9X1#$Bb$j!;w|Re4uT zx@GG+Ei`+8Z}v+i>&NbI5-y{{{OznLlu!+M8gC2KQ$Ip9`d< zl@G>^0#4z+8zJF$0lc4y9vnXGz5iPSjU3Jh5E2n(o8kRTyzj(u0_rzW0?+>$CwOe) zn;m#RlTuueBH+F#WOnMWclz@pq?MU?KOg>g;Qk%Be^=bksPXSs{Wr$_n;ibmUjHVC zf0M(%$>HDR@K?$BzgqzN2$GUsORF>p5~p(NbyiH*AyZrleQ&CJP@5h9`$3U!PC<3= z@c*Ui-be7kUBe8SOsza-W~Jz(MSZSjoY%9DOJLWAzI*L#5B2*V3OphG@Zt`3+sqqyN|YM z>IcjZ)rRw=3YTxrl zHAuywnfdexlT1~vVg2pa6ct)~z4RO9BG?gq>}l*~#g1qf*}`s1ow1x(9xI(9;X0S`Fq|Se>oo!fKBFsNVkITX<%9Mq zzY;swRbUq05JY$GtWiTGrL^C{h7Ysj-{K9vXW#mkp;MIf7@OfJqQv0YGV}@AF6Ma+ zf{EMFc8%w>j=$nJQPr5ex1ambWL5<#ALV|#ugoQ}a}R^_@s&>F*k*S<@)!|^RI?nz zf7UtaVb(IKTO{u{<~|-$IAIsA=lJ%B=VIE;{z@NS!@3ajmBl8>AX>}f6Z&oAXotym z!&&NNG7t;3Cfl1>PgegGPi*#u^CMYNq*(Z?)K%J z#fuqiQf=BKI~zr^Y_^(QkX|xj?gTsBVV7}%6&jVv9(LsKjR@aJF%j=uBwZp`7E>7% zo=Az8Z*fpOK*P^^!(}AYd9~NcpyuhZZ&$Y$Jm{<|33KjzemGYsd}%YC20{-B*-ehCIK*5cgFO79d7&M_x1wZsKwzC_W^{nXq~m(H;~ zz(edbEHE}(U1nK^Co#dpg%;f!t#Y%ZW2ZeTir&>XCo1H9OIfT0M+)1Zly84J98ik1p z&nWC-k4;}KDpT5*+MuE&f=QT)8C>w#9;B{F5!y;}9YGT)4LHF}HIcN-2VF^a@@?_f zDMEX?QC(q#9X~~`q%65@U%tydl(CepMRIH4&`?s8rD+{AS$p}~pg&WAK5ks8)I1G0 z-hx;ZB1X&id?8H7QY#frM=0F;R&kd>qVmzi(r@eQy7u2L@j6VjuJz7(QoGHTJ7FSB zPPWO-u8)pPT#prOP>)tKyHJVxrdOVc}j4QYHtdbO0C?gq8<7YVq%#)Lf3MXyzFB=h>bd%c}Fl-%F(&TeeV(!!NO zWOvu*Y=z|h4D?2O(Mj=>dJOlv3(TkwNp4w}&DYR&iqTI8XZ|$aB;&*G(oI39m?hE@L(G z^|f^2CZ;7hYcxGN!nVJSM0o@_fN_uCcRcFb%My6rZrWk$Wlw6TRT7Hy#!a* zi@xsdT=|U^t965>_2fvwVds%p%MHj$f_JD3apzo0eLuxXf@FrN-cqObV9bo=#zCvZ z+?((3HBW4>bZ7DRMm**pw5NHz9w<8VXtmER>%C%~WPJ$9r#c1KL+kKr0qcunwffkF zaCe*X3aj#E4HLZ$qwEWK(OY2mK639v!u{T=^)23Np>;z+k~g&VPF#1d#|<6r8|aJcZWS1M1N=x} zx{rdHAlCZB-o!9S4Da~$+y_r4XN4nujF87G+8C=F4?&qrI@}omSIcs+V7hI$5r0V#jS# z!)K1p=C>$yGP8vQ(h6m+jCQ(ja5h(LcUOeon?5%dtrp)`>9fmVbGoQ0tRYH>$8&p9 zzvFsrQ`ru7GZ?iyI-Dc3_+)cR3x1MlrC_~6pwyrg^ha{jVMu-4N|2b3iaMIF#Pm`) z8$)(P3I#s~S5|V7xyoG%==KabDx39@a9LL&;cY!vOrn5Bkx2`)W%+WuDyiaoL62~u z2cy@TH*1ktG6f@#R4vB!BxE%CeNK)ot9~qsks&cwsrG#?W7S&5F`+S5B0CiBE%c*H zkCqg8%0s4&KRr8PGnG?o)PlubZUjD&NFBk}suUw4P}t`>dA-!p0*NR&lbNMPi^zT* zo+Vj@`d|i1qr1JXADx>~%`s>Q?lLrACQ{QfpUm`@%F+aEnCWk7M>`Kbo^KSLy(1p{ z!OHcN=2c`LD{#7n z7j%eGhregvFFLkJ#wa0X)buKb>{_TKd$i<24}|Qa;;#0Eo__R>-%~WOdtde4gUP*!v0f+HiZV7F`?gf4rZ&`1mWx2*Oj@ALY{^Sd=z+Y%D#7qVL@P z9+MmR<59AE-^+;Rp<2v#4<(1jMb`>J@mJ9^#L0vSB2F{9Mb|o;p(>NzFiU~KWYDTq zjFpj(=jYKaKC>T%h=PtR#jmFqm)V9T7CtcrOAIX4ab(@ulyeGcacjV2=Hl-|vi%fi z3C1`vDuccUMOZiM!#$2wuw6TJ^vvE-k0ta;7!(pBS8GOw9&}pgc+aP`$;RSXS_04a7*P|AUJ%mFL;Q%b>qH+ zSn=;;{yiW6Zye35p9spJtS`nSMcw`ycm5lcPv0H8Er!P9g93RCYN~qP*+nfsDPC+8 zMe2Rn8g5k@Iu?{WYe73lkzn`v1&gdmP)*o_{nRy> z@#yz`?@f`#T|poA<+Ewm1F(wT{q+Z|YGHJrdw8z26>QL z%z5dXteN@7nB>m6W;7#q&JQq^=pdk&`g=A3@&!hjWeau1b078_EKjs03h?oJl=t>UPMtzdnE)E~U1>RdY$@3KiYx{4OEpaegxb!^mg&!KAtk_J+Waf9HC=2;b$aZ$QQeq~}3F&>isdW!`=zDpX zcu3^fW13Y-XJLK|V=3$xZRe}0jc<8CXqw+zY%&hA)(0%NZd&p8;zho>%6v7#$y;)3 z?xX6b9>Z`w)BETiga3xhs;eH8=WWo7*gvAMt^a7H{SCnD<@3vo*a5~|SByu0lf))x z6!*^9L|Rtk%;jW5O%02A(1{Q6e7v1 z7LtJ1(mRc#ZDfX>Jk9j?Vh&no-6+qkW+Mo%aMCHY)LT>*U%s{yVW%DbD`cx!AC2m? zUze=dp0>*C45SyWSw*duFQ_MC+B0-&0iDJ)B(EeMPRCn1l^Z{*5%JD4I)_%3^*b>Y zCidgk`aD*9BX|vDVjL!PoGS_UR6RefkV{zLIq$FEs>!AEs!vYI(4VXD5xr9nZ&c%s zoR9D{OX$*fI}ac649USqOmEcFG)&HxGDRz{`QZh6Q{*+9UWtpEIhPo8B`e3$kQNyK zK+bR6E7P0r$y`bBNonra05q6Z(}~lo*)6U_B$~> z5lzIQl7Z1AO{Rg*CGwEmi3FBBpF)lF>%n9X1uT1t5pJzR^di>TUOVT+SlNkq%`B%s zsBi~A?g{a}Bi=1S9*enc8ag8{ma$Jg1_c4!AN3bpTW#{f+7(*366@mD1(lj?0L65%CUGm%-9qD~BVjsW6a2CS2&@#0 zQ}4*sW9kiPY#Mvx&;4YH42|BNamsJyqxFV8 zYo=i_oqKUpE6s@?_h_`*Kv^N^#^8go^rk}X&ZPbhO^m2h;8INg607}&{kQE^G^5Ga zR4b23onVGw?nCdRHlpjhLnLaE`pNx)8My_75|8Tq^D^0>@0Qrnz4mzT{6O z@V$1#q1mleUqcR!m{g_!dBQ})*KD)S64*?j5?Q6T@KSHbp#>QHorB7UNNA>Vp6b!5N#I#vfxnQ&Lqvj%K<-c$gLMmwW#|W5dIMyz|_G75y ze!A>5Jhh3k8|Bmnnr zMIpJ3QO?6Gcs2L-Y;jxfYXN+5UF=60-;(b>d*!3}+Hg(NQutb>xWMWXhZb%^j@b|^E$OH7!WygJ^4xu`h?@lk&rj9j4QOnZV zTuf-9n+Bd_{zdFof_;$){ZJ`-+fuJI`zhYa$1-Q%+H1v^q~NR804ahtg}p){>i>?T z6iV!(!j~fKU7h}fQ9}B$J4wo?dx1!|8F)Y~6RC7s63rhY(8Vo}wvI_Z&F-Y};ZAtJbgRg%8pB72~Xfq=W)wEpV@MIE4zpk+!R*_B5Vbc>vwbZG9m? z#KZ{Es|*i`%}GHLmq=VMVMgkmj@$LPapWrKE(h){5BM?GEO+R~I;_F?sWNSiYQ9G} z#<~!;PV*8`w~u5(!)Y|zQH2U1Smi=GH*#_`1Hp=Wcjv-(a_#F$(2p!zk`({Zr~o^q zmki|jL6+H;=apbOH2?sOxoct~EZvLAZ(mwCpjo(@tOeXsj_JNXX zzG&a!QqNVyiB=C-^{SU_2${K)?BrJg0(cdbj=72fEgwEgEt0x5Lr}5OW%g2vX8s*y zVdUccn@F;aPaQ6|)GBje7^{klr z35vUFa!!d}Yabaiogx+p+_&ayWnIfk$`YEm8h%l%7TA;UmUaJrp9{qOODQ%t3cF0x z+;jWT3nG}(@&>);b@wq*^k<2pkdPmaLz`(+MUB2@3|5-h`c{?-AhJ z19fm`FB+}Sp8#UIYj??Yk9W|Vn(R?U`(*ZhxHZRJ&drM`BqK^_cn7D@`HF~W3 zg52kPUM9Bi3jgjNLg$gX&|n!qPautR?dljUxMTXM$uZH?+`8AbqfxZ(BcYlyX&M>w z_03f&D)A>hTAV*WKONksiJHo-2Q6*wQh*4p=we>%HfNG@EGnn;ZH`ADmS%qgZlTey zK=D(SO8!Td8kt0t_cc%~#y-g=D9zGH0>Q50S$i%kYGzOZ-Wn=Lt2?utK}UH{>Z=c} zK9fH97!8sX;3%cjyZd3S3%N-a%TxzM$ZhsmRcuF&9GpX_BE&K_t-iDhU zI;F0yBB*kMC7`%0=XCk1>(}#pLO)n-Bw3a0kG^1S7rLMm++a7z%#T zW|I)X%hYzZuRJd-xwO>fXolSU>bu?hsX~mJOS5RTGb^wg!hl{*lDUW0f#>BIEI=o7 zkT%F>&X)T{ON8=!VYk-m^|8%2x52iNc8!HBJrQ@Ny`>a=J`nfsIw&01y6}?EI4o&0 z(iwJ%f`5Gx2bXzO6~Z7@MoG|z$XKF4lgNFaTKN68zl=Hk{9^2mSgBbIg9Qq<`FzS(Z8iWu=X(Hf9+y?hzem13DTfXiE*+AISxCy?S{ zi}BM))j)s-s>@4ceUL0dN6nPKMDQ|=etG$VYo;wtvv`-T&L=Tk)O`{l)#l1Ht;yWm zYt06q)RL}fW$m=Y*Coe#x(YtG)ph{@kfq7Ob@_w~qGszOm}-P^dt5fyOCAH&ao~o; zt&onaNhjdZ5P)l(HVP=90=t|wnMyTVvE#YI)$cw+si}J;pu9fXYA76cm0mi`T4nrf zA@B+)Qr{w4dRu|Py)q9!k%imipUCP1p@g1bj) zt;%@yDPJ%vKG`}lVZSQmv0w}dGjTbT*%GxrsyeP1yH|cwzB_4Ae70mCEQ>wVjoE~6n=)qg`xP!Bg<7x^)^o8lxw`xP1=p9c^eUuo2 zo2IfLDkqkrF!JM_dBy+?#Dj^<)#}&ghZLd`R=s9RXHm`W(&XzmwOq<6?(#r6GDX*s zu0;;sE4`KQ4>s#XPt>*)AYC5(XBkVHzjTTQcl-eHb7&#!0pIWq2)D~l zfwcj&)c;XUf~A*b{x#(kJ;oRirdO1#U3r}CSivUqWJzx@ zU|n#2^sCYsu4~7KM;gIAUdJA-;1YX;L|pn(1eWU_!w2wkOWZM#YK)}v_v1$U!*|y~ ztEr!5GRD0!AyLEe;{-yx-ss_P+j{E5-a4L)=cMs3K>`@-*x~7WxGR)c8|6*Vv#|N; z7_KtpEyV|&Ck@_rsqvEP*M;50{b6X?M*M;#@;tl;$;DD!&j5i>7|++Fk=QqUKh0AV zvi%T*9&6T3HnykQ@A=0#Z zTP|1{&Umim#j2}E@joQ+X;70Q=o8y)(E5wEQfJSG+BDtJmxSzz{i?kk_kJqQN8o1= z0GnQ;h0P531N;C>(r>3w{&o4#z0tOHY&lTpS+-}Cz{buaDAN-9@R|JhRdphg_o=Bf ztw-^OFGX33AXB-<7XL$b!XPP4btlsUKA#u_+(55Rdw3hzU4O5QWUti7mAj-rSAJy@ zIlNy?ITQI^N1q^x4OV1zUWX;0H|wT0W4$ufX#>o(d7YZI`OiD}utQ%@S{%86j}-UZ zd@ZB#KoM+9MB^Fu7wSd}^*J+1z!+YK;?by^ghlBmclFg4R!~?Ab>n@hyuJ1_J?Fr_ z)MCEbeND7vsae4Q!d%qET%f|wU%zUIyl?KoHoBiT{_)HyI`F>T)vEX(o`p43CvwdW zCo=+HXbGQMQY+xE<2rZ>_Q|UaxO^^wPYEoc5tj`6Keqgw(5Uv?n8laii!i3X8!DSD&2IkSmnC^5vF+g_+#JLU5 zt){@1M1}CR_97G|lWgEqm&H+mjq8Z(3UMN_ZX8QFr zM9HOv;Ocq=Nm9gfwG>@F7VE_TaaFNs_awzn}8kpU}Gt9sQiTDqA!K-#X|E@|8 zSfS=2V^o8i?g%Jncg2>zN(axJ;lPU|*(j;kMAoiVF4(R0mmlT^J{l|xfBpsz19J)S z=VE5<=|@HE?^1aJ>qP(YA(_8rZ%Ia0d*yRTL`4*yE%9w^QR3G!9|hdfNL25bhql4;=eTToV0xP{R8s)smGO0>isNI_nro+s2Q0FQsnv z5n{pAI0W=BRoiS%l^_XQ*1Ss+Z#b^1#hoEsL###MQ{y+=K4u)PQnIM? zf+<(DvfOpKjw%^6kUop)5{HK5^Z5(s-v)}We18Hky0l@vw4w4loyw1U$M{0w$oA8sgI&Jq*rGwV{KHw9T;3cY}?_W}y0fs0yOlXwv zJSNceMsvTGzKmRuW)QuoV7EgF$M#1?Yev)O(KW$`sGou+9S3!h!R>L(LsR|K8p#-PXDon^B}wQE(`bK!+B zKq@o3q|-5rs}(!j=J^;Jmf~Q(^|2H|TOmO_dj~l5j(FEglC=UEMPgHjuC|yVCJoGf%fBw+6svS}Xu*n4QnokdAmiIWblR;xaH7#=N%HoEkN**0R_ zl|c2K;*l&0SO;R?iw3h<1C&{BMDKSzW#|Wj>q8{)GXP216+47fZX`o#wybW}yueSY z={j|e@hJzH`7$__cqeH4Q%^_Ry`V`s`!k^;F;~0zZ%{^CwWVV5P$RgzkQL-JBqj+nQo4W%fNaIa zB__b*niIGUY}N++WF+JwW2*EVt_j z6c0au86`_-c+hod zQ=gAQZ!!Oc)p9`SzbHZDAU(DTnW-9xwy82tTmc&1B$6C+;0fY7|f;=x+lp9$( zHa(!)$&(Ffd4k>E6k1DJa1~waz6W-Z6Uh1(2arqsCu)Q1PnD#$mvWNb69D4V+@lMP zP=!{#axa)e+sWuf^7$?BYf>yX7@B*Etk9Qzl@p&cNWVI3Scj%WMaT!SCo5wUWS^*= z+u2&IxURN{qlAqJcrL{=Uc$2di1jj0kE)q4x(jr?)R|) zG&~m7zE%!hAw$T1$TMV1=yM(lNL&V|ESA)vWHV}|`+I4w5Zc`?fH1w;5yNf0p)nVI z&TcUK_d$2%KYJ5LZ6Wd9WAdd1cVPhM??0*e9uxXAzG8AR=D&(%9^cQYA`vdJv?n!W_OZA^8>g5tc@&NK_Xi7+L_&{ zMu_Ld(FT8}K;e_7u~wQ?X= zgpO+HGRRv-`ZrK=FFf_#=c}39th$TY;Utki&jLtU0P>c4747Av3v_pIk{T$xipeUm zfZJZhI7MpyQbhV(=p}62+gzIar+jtCNxen191^(M(nK2vxB#Oy(^iFgN4P3lbZ*#m z<{FOVHJpuMuwp+5xIQ{Rh{yPHy+anOC zSY4oAV%rcYz-u#DRSapt;QhVb^MDg-K@wtwjBS13*x6GXtFt?a1zXBq8)|g6pH3i{ zwg&$0Yrjo>@2?y23+Q+72z z{vWsB6>tbDi7&U(Ohy@uLh)sW)Rt!U5sJ0$%WawlTWj+H5?eE+v)lwi+6tF&>V_w$ z5N2s+DM_^Ke`MfmGw-+{vhq!qF=lsH($o^zyDa1#9Pj}p10wPxrpK-IMZPCXDRaZ0 zU&KRc(Yf_0n8~0$U5nR#{43{HAcMHbLo!+_P(0H7GUxqc0iCS9Kje8BaUKo{ZM+#R zb6J${76Z-=$$cTClaUCh)AQ&0)!oNO+V02bc@%NvwGRNvr2zH3{_b6S4n75o{+&ET z1M9K=NVc9zwxNe|cS$%Nvr>9uK!~3IP$dAZ$I^$y`AZz9wdb>-DT~%VQ}84X=#?ti z58g2}2yOG=%QV_{sX1mGGPBh_({I10STpNJ=OBeQ{GnGWA3q8#|L5-jafT$ZLz1~p zp3vez4~g~`)eYJLUmm2Xr%8PEDFXHk)qWYXD@dI${xJGyo$mxD(#{27f`yE}M5q|Y zJfwn}qWlLr-_~rcy|S&b0X${jV~F9BY0ZEG}o~OV{#T5@QBo>MKI&cB7qOj@1{4coZFYm zmr+xcf43ze-HT!*V2UWG-9m~iXv%oiyA7R5vU$qF;^6LCJ~&tKddv;>RBf@%Q11IO zL3_q^$8|4OPT&#rhw(^d{l`$J&auS0t9~&5a=8w?DTZ&cNzefo-QtFFIfrz(f`L!h zYA#(RUgnZ(kLxcQ4@xUXzm*;%r}1e)+h=gvDDiP&=a^ z%DcqW5RX1wDLOF;?M2}MbTKl|mBu7*S3MGOvio$xugc1EG3s*A<>co?bAT74${BUX zNNBU?ASXWs;9acM$A|PRZmkqm1T1OccdzLuB31W~-lFnOR>X;4da3Cp(H7KyAcOu`^pA{ z6{Tz)oJK^#$tKZ;$Zb#g?so%%0fASYxM*miUH_8*Uc9vR7ueM^Afj|FVCe ztz>T{B96Zufij(2vE!99l<7kNTI#7*!3~E9&82mkXT64lWy1pPU~_BG`M3Z<$ZOmf zo$d+d+z<5qf5hy?T?-+|7DqG`u(8(Dc!J5GHiS)0o^-vcE|g;k8Zw(?RF&!*C3m*m z)dSEJ&&L{LqP9aNU3KG|Wf~9A37;uMdv2QZAk`+~a&?%Q2Zu$_5?6nju1z&%PBoRz zaGob#s0V!&#-g=viPZ+MVQ%SLaJXszh)d9QlkAK1x&G|ri9h$JcC!5;b;%C5l_zne z;ao6E5;0ao_lqZ@b_-9Y=jATy6smg zgyI(4ADM+qmk+tT-%VY`*Ca-cnB`1Hp#LTpBWTfeF;>E3a5Ty)&}OLS(tjwhnUfIH zA1%LW^%KhAzAVLANYL+12;Rb`>)=u^w1@iHjR=}vyO30C=w8~0m)gZu*_ zL6M##*Q|3ML)Y=6_QTy!R@;RwvCjYH0@Gfw0xzEYEJM-oqP-<JUYrs#R`iy7DYe1b z33>T?;hn|UD_P>Ke_y_V1JsbLWa??4-!m$Rle2FfAb@_+iOxr1$fIe?S|ldeAgyLD?ALU*aje2jA^3O5Bi1 zny(9wHnI}Hho9t)C=Po^9cjtL2W%toXk0~B(Z}#E!5DdBwCI>i2f8g5 zyOa#lIoEAuJ9`;EIQ@T>96p3;;4hLt?yB)WL`b~;_r609sBffvTY_wLrWt)YT=x&Q zYp?!12BG!~5`NeER#DVu9hnx*GkA?T(}=Wi*^BKH8u$xdsN^_5g|kQCYkD*Y0Rg?; zHUK*kC_IszP)lcbWeQq3dF@8OLr;3N0{+a+{hj zareokBdo1VUol6<5JQ=VC|K=&SVU@HAx$?RD=IkdDFzlX2P6ZeTM=x!q7An63Er;# zK6`3_cZtle0D&j$g0n~#a?Am8W(CF_kU_J#7t$PA*aT9CgqgyNgK>D3a?m2q|bBmo9sF>y$bw&qekL? z>z?5q{x97#G};k;O&&u5Lb;{@uKXbe3)ckF(^_pu>35Cczk`Z6Za1kzS}IIB+X4?JcXM5U+w?a^bXk|p++ zW9G*`9u!SKqO(;l?};*zqe)Lw?F85~1nn8F8iodqMa{hFxZ$dD6rm4)oQL2MD&wv+ z+o-{gp&uwK1Cnn2&I`|$<2o2_%MEvx4NEJvdN6&;M?s5LJii|LqGuz9zJHXaEDdF# zrGyi|d>T}s92O>%DS*-=-HT@NWt_dN0FTLBIvE>g8*KrwLt3z*P%c)nu3pojaeW@I zy*-X*_r7`15){#8<^^pU>vzg4MjW8CM#ZX}$L^AX$k>wb3Pgo44+c|%?k%wW7?xh% zVZ%GovOn#I>1v1nDzfL6dTr~u6dVNA2`uHtqDf}ueO(2N=QC~XOuMLYVsXHcpGGa5 z^8+kCZG_o3BK{Sofar2qzGNoy0T#%?B^d5G$?X=&w(FJ;aW|Yk7@t4@C@qFc%=Bx1 znG%$DwCOvB*Y3)V_2d~htwBB0!`X0Umv170F2h`uVXY%>w=Dt?;qU-hdC}{T-$Xj1eiXTCr$(;QJ#>_y_Tsd@NBr5XX_UUBKZgJv?EDDE_evp3eUGjZ= z#*Lhc9Rl-6!wD7AaI!{u1vZsLaF9&z*L8{IR<(JcFbs_g-Zrjs9Je9G&2Q!*2wA_E zCHm^UNH9K`l&n-C#wW7okTBc|J(lD>jhM5wWBStPI3VQ<`-YSv`QPSSA_gE6a=ctV z=Y*8YApOW7P55Dm5fnL!rGmqqx8Rrdk)7(uNJ+L+O#res>lwu}wc=eyo8Ju^4u1-s z2lizoFA)u3i--A2#IuH3Xn5y^%D7qNKi#f%80ilIl7MouJ;NU{HX*S9x;ka3kXHN9 zUK@|`*M4VXem00yD1<8NLDO1rADNO@_zhQ+&dAZ2D^7NcrInC69>X5zL>jd&$C_xD zIolxKV$eg2r?wEX;|C$WA~37d@Y z0v4%>ClpO;|0k^)Y^7<}od!NUKo0e&1T`fBipU*3BZv(WX3lB}+SbPRw(H>a#muU!;xdt_G_-~!qXG7SIyg6Hc_zy$Y+dCb`9$sc%2zkBw z%5JvL*yB~lWKUjGpV@5rt5t0`^!C(p`j}0?*qclKbwJS&>a_cgiqIyvzqn0gj~og63O z(MP${40SBs1DdfeLGHy8v>~s{!r@pJg4N67_TyK@pFfQuIY<{#vsPUq=Cbe!G>1gM z7He$eEJ`6jd8zU2+2JLz22Fj#_Mair1xr>r4T*LCR&RgqfM*b`kdhU(cn^qSrUh z+>?IH>y3EId^iLi5MTDj|2l3!oX+dbi*CORZiI&0pJrK;XwJnM#^n&R{KzBt0o%aM1G9d%9apw1Vbm{CPc?va*ApP;x|8XQM}T zBX&JKf|lCL@6A)(PlTAZch_EwOOp>}mwSETfjxE*NKr+)+PyuY^{1qWe)&32oc@T& zWBAh%r+j|hlGjq)M_&ru>(`nmRpmPN`O`b1c%?lwmC#)Ef|!AJYz>aWy913m8;ej6 zeZIe!JmJEFeWOczHy_$y0g*6QeUhPJ-S_L<-q}OQap30gl7MQQ6k*=c718W3AWD1; z+h1@Zcx>Mn!UID^@T|=+WHHJ|?R>+zbeO?BjuN~w))bqiCj9OYHJoA(TDz(Z~6tzs$|4!+gld5&NIL=aTpB zRQ46H-o#TX=ZUL~ePHG9i7^OJseNLoA9mu`)u}^z{mAYT zux(qqk?eVHD@4pm9Cp5H#Q9XVd%w*GdZz3v{iuODKOM%15)+VA2cFo7|2m7;72u_h zp3k^qKXU!e;XPp>^Ig%683Gc)u6_?gt=_Txn{~sDF*@>LoX7G7R$_hZOrd0F+zjND z`*qEQ9nSthAfuy^rlBarCZoGL-A$e2vdeBAcA(Jh&i;T=vFld0hxUZ?Rt#j8fkkjT z56&9>Ll}K@@fmd5aiP6-%C}a!t&GB9@e8whN57}|bs-%0$w#*PWLw&pDeC5XBByh$ zA3#&!)mb_!QFLj|6np@vvIrN6d=hNYrftT_1zql1}{5l-Y)nWK7 zjP*>o)m$KiM0?Ew)3!RjnhVgh(1Qly@^RzxWG1P0gpk!Mpm=q&GJqeF_4ZwxWBGOP zy&<+(1B;LVXJVDK!GQwIa1Hs6U9ifS!Cv1S9Fmg*iYTV;=I*D%VCo-Nj+@M3dunhL zp2nX}%=zn(yyF@Jq98|Bx^}FA*UyHIhgZ1ewfEoVoh9JO$`)mnm zj|p4H4cCR%Zw(9m`Yg{XcaCJ&5IjU#pKgTk$e>sbJkkiX-*>n;@=`m&8g(6ZghSD2 z6hRZ|?J3!*t=o2mj!t4b-+2tWqN22q^Co}CQzC;qE0JcI!O*S0ubE@I=N+PEF@rYz z?IH57M_?k8`&zJ)2(>phv-!crD5-}Nnh6*6fOCKbb+oMw&sn_J7r;GcH5+6J8 zR&ew{)I=}uJ#b5Q>}#fracI|ShK1=A;sW)KEx@M-Vq^!p2zQZ;AJJ=ds(N&!1O;iW zHM0}W;t3abTN(FYbR9qOQpCaZ(~$M)-yX^P68G1n9MB3`z#e+!%d9}=hogiKhZ~|a zOPppii^AE}>!zN+p8eyr{3kfiClUG!?-@QicPQ-0G9pE7xUbPG!2v~F5CU7{KP4TI zd4A$@%(ZwKG8ZV{mQv1os{NYzyooQ6XE)#e5Uo?p%qZ@9?Tm>3x$>iXWeDCv>F|)5 zGTRZ3V!fx#|DN=0PkKO_0MYFeEWFJW>gm^cM~@gCeP{hzW?wQ8n<}88oT6mEcKC}~ z&HR7Z7~l3@HzAbc;qZCm;d_5&BVz7WqZu2cz>gkszog_!8KPsc!=_AJ z6IXtX5U2Wei1@udy6w}2e;Yt4G2*)fwbK5Ezh2Lq*j>S!B-<(~kdo!s0QkovKLWCT z=es$azg^GUYVTKq6^oXf!QVIg{NXr4xTN92X%?CCWlXBO1TXfOkC30M9=3 zx;ppw>!D6d?fPD%So`Gn3aCojo5Wfy|9aSO*Q+do7v3zyT(|i>%sG>OB?$DRs(=1l zl3EKRAiIecq~}ojZEz1Cf@h!qdRyza>k*>=h$EcpPBN(eeQ*a)JQIEOfm7kf<@)J~ zSd?FdK!~dchM`RD?UmM3ppx1`ccpwLgfSb68To(iI{+hmu<0Qj*h_r5vW1wzbM4l# zbG8Y|lSi#a7~JR1yTDc(>H~%cU)}BfMIdliM!Ju$plPN<<2?V7uXdM0wjRrI&1RbtVU@Vmd_Hao5^G@(! zZxwHX0XEldbtXjB7CbhvVw>L}|GS#bjJM1ZhzR)>HfOzdbl5f0$IU{ixOG{fu3IcP zSg8JrB-?^Z34e1Or`;RDCU)rwMKHiDV*|yrUK0q3w76&BNNxzhB7rfR<*TQ^t&`)W z@am1+3t~KYH%H!4Dh%c|I~B-h$*{rE&%ei%Tz}Y4 zQzx4ie_wfb*kE~SJ&9uZ4PgB7iJM4+rV?2>K!&TqvACQNP`<7s_>~!;PINj@Jg+Ts z;G9WJ>)RKtnR+_UCc$fBy0HrIa0IMOys;cAwU5{Db6&=q33zlind44RFy!hn(Appm zdb%Nyp(b^Wuf%ocFdbR3mBQs9D72V2eF}n!l><_J7+N%EW+pqbsNHXR2OZ;+0oTtF z0Yc|DfkPVi_WwE1gB+|$b*@N3AC4k zE}0kUS%%-gd;aU}a-O^yx}RS+x?%_G)>{Ve_5yz0v}g=?TLgD4lUhUKt+t`V^p%%R zbro<(>)itAn>qh<2V5WPZcj>tHiOU|}v8 z!CK?Xd^SfuzmTQviRjpG;0#AeEI`w!JWvq3{N>&Rz8H<$I{(*gB~Ex#zekGxelJ`0 z%Yk&~0B$m?!Wdt>>L-FMU~e9z>{5Pt_R>K{iFDexKxbl}>bJN;P4pa`7WFXvDP;YK z@#kiC4HN9Gmg-wyk+aQCI6=dDGCZL)oU9rb|n5qti!EdB>Gi0A$4mwr`DV9y-ZnavNUq0?dcdxnf!6*12^p$=m^dip8O1p zhZ`C*+w1~-<5gKEgo7KKEdE-r{d+1O%i*K~%a$LSVoG zX^@nVkdh8b6%a`kK}r#jPLYxrBn6d_?vNIwMjD=d={e8ue((GJ@BQ~(%XQW{a@H_2 z*L8jO-k;hu9Y<#Ti=|#0ckS_fiXHN(kq~z1DO8Q*zwW{~gJHt2b5*r4x?FD(S4Knj)P&ECw?p zHn^oVj|$xpraUoa!Nobpt6PvXc3Hhe&~5TRnn%x9cqzRg>F4Ly6)dP?KXDhz#MDs8 zWB3Kn)R56a(l>&J=N9%t@B#S+>R*P*iplV(ljd45G4A4)xIxqM5@xfeV%_<-v#gA> z5>~N5^Ih-w@GWQK%(V0@p=w^sSe<`eqV4%U-)g-TI4|yN-vjHQ_n4pIrJV|aDPlQF zdzSB)_AjiBm{4Zk0g<$yj`+<{kggbc~&4@f3DoorKVZ)7s3XlOr>04yoYvF+%M3`2y48wI&n4Qs89nm zB-EZKF{deaAd+wfhLt#Y!Qu9tMs0EQivs{ssPD77TxJ`%Fg3ZYKjn+rcwdmM;Q_;5 z1s!AkaE0|VeiNB;m1-iVx8Gv3rYpiM%ojr!3sP*1|DrhT-C%+r*kVH|0)F_<=+UAH#OX*_Y+nftF#X zE}o8acp{|>r>*V`Ck=U?+9R`+VVd@v11_=OEFL>im#I6_Q+9=i-gt}qu~JMQo|5Nz zmX}aruvvy&LGqNx58`UkOx+(u( zpKp4slf0mds>ja(st*wa?ZDojwcWv@04+}06JC{AX?sJcN#uAL=9dMBn?3hO2%H1| zr(AKv1rHZ1(8*75unm95k)_E+`lTLD%25@i@AkrT^+eBoB#csz3<7spAZqM;U5;*H zO7p=Pc;Y=A3jU{kA4bmL{#@M-`(c>1D%w*|K?^lsKt9r8#Bc{c#^9@TqxGEpX$tC2 zu5IZjQmT(~oc|npfBU@V97i9k`* zsB%}HKY%&CvXThG#bOJN>tI!ki_IP%($mdigCv4ukdh68@~>)Jdq9gmhwSF1aXKT{ zm{PmJ?$NuaPP#Fjy)Cgq`P=KUD*M0 z?$Y)YU?!e{2!FCc@n|3FzeBfwl@R?`(KIZ?TE^WgsKkJ|_q#{(#aAZP_2=1ZT+kG zxJtH0#)wehF#z#jBkpu>T8s^xn$#8g{HzjAa!`WVfEITzCwxuJ2=RIxAo}rWNAq(n zO4PT^IR1BUS`-nW&55-h0VW=Fy4TSgJ}xtD8NY2Yvz@v2A{v}s z*Py3Ew^O>XT;N>X00%S2={lactmz58|3Qa}UIn;uVgn<5!Q==Q-`JGzz=F_ly@zhRqB;dJ-Ez>f^EmlO>cTzX+0nlK-;Lpf2TF$#>*pxl_xitY4;40d z)Eb3+v!C6|e|wGp{D>VeT>gCZ&Q$4tyX$9Hq45@4SReTB-zJAjVTTb?_-1|Bf8N`F z|DEeO^yC)ZGsXX1v;)x6FkKc|B?Qs_4>WjET!2rcUxBvLrbX zxtZlo@f%-&P=Kl=zug6ApbbA8d5oTc5r*p(4c96-Fs|1=*qQ443Cfk{5Sy`ZfUHNV z*xK49M;M+6M)9^7PUUo+{M#DuZZH(;O0wgIOKqP294J;`U4n%H&zK~gxqeg_ONUNN zlmOriK^zfzCUwSPbo_kUIa)Ecy*XFm1gg=SxzT&B=Yt=d8*t!6eu2rV{L=0&NVfUm z@M!O;tZ?Z&DdD$Y0TUVb$`UPqC4|-K8Vy1XCXvbJDjRFj%sxS#ah=?)!Q-Tx080p3 z<1rS(sLw)sb3IJTGxT@j2vnXSJn}s8zwPq{^P36P-!-3H~Wo39s@bF5ckGZ zWsNWJU~j7(B-v@hi=d2)24Ptn$S*k{h2%r6E-lY3HrpBS1Q^s%kiD&&_VlA1YdPq= z#*)i<+!WX(0C8%Xs&#_|xDZdd${0B$FhRCQyJq4NDT)C!=4`EodduD+r-Q{0^5>8R)*m@iu-LX_YvR+LmwGanpK#~V!+ka4sFo1 z_y^6}TI+tGxfd50FoR^YyrDwmXI8xwYC36f90nlqJ*RnLwEQ(D!2V-Y=DiCAtxWhw zLo0cbaBsPvyhU;C8??Ad@J)`@^AuH4XYBBp*E2|a7Cls_58EQ)iBls^OR*NW?2&A{ z*sabNUG;Vmf6xom3`ua=s_b)EUNT#5KGtsJhf5_trrunYONa4{efkE*RpxS5bCDwd zqeBM1lPgQeO6oV4YRM42YKsCghBqjW5o z3IlH$St92!oPF-91FMYN;*0logF?!chaZQ#nyovWrsAkf8GlVr4uvlXWAGd=pF?V3 z^f{DmZD|_jB(tbVPvCH#SM(?Ga8i0EW4D;#)J)U}*UZC^8--*e{x|osyrS51EARF5 zDSySe4^ultPa0cqH{p>#=nUBt4`<`-Tn3cK9Y^>Yh|q+u3$>GenX{;+b;QA6wQQ^} zPoS4{i?7qxxO}*;jg!sR<<9BQ%JNCSS#fY4KP;#tgO+*G*0sO)X2kltK~QhFdoZq$ z8q*m3nzp9+&JDu?W0{RHkTW7ji46yi>re6RrhTRDrkxdnNw&EY9#?%rKF}B!9$hO$ ztf=@?N*}L}rhGtD@X#qa5$60utS?RA`&&8;$5U<*@%e|ECLI@4k9mIo=NA@gbq97Q zeG0m+k9LD4u;2)#p0+Qg_YFSB4_Ha?z1=D9cUqQYbk$HNLtx#+p$+I7!PYC7#s0!VoJd`RFQ>o|T&;BR z@_YTN6Bq^_TM?1f%)*xh1jdA_gd~}0LuEl60pDudt!O06#!y|YK-rXw#4wS7u)eT| zNu!2r`Kg>7d7xwG*Pn2QrbrZ}H=Q5^INYk!3Uz8~2dyiYr9mAv?ED z;WW5@${3HCPWz6uFHg2q9sH85W4+fQ1x=r47#_@BFy7g?p+W_jZ^%f{t9BNjSZok?F8ij-KMbV zPqP4o6A*jqh<0f-C4nrmi=z)ZLy6a0G4%YD(%Xcm>)v^XT1SQ`R%7M&*g4#}Dj}wF z&Ir!=FUAe<9u63tgH|c7*2ua7pdVX8*f=V;?)Qz zDjmF7KJxb0;CAzol;NJ|tNS0qPA-cVugR>Gi>-V6+Ti)rJl3nTW5AE0}9>wgibaiaC{SUn{pGLNjNJ$*;elsCY zK_tLJrYX5~W(W^ultxNlKhfkCC}RBD_xCH7CgmcYx@5D#)Bk1xJR!YHb52)KBLs`r z%;Qm7T`mycRW68ne6Kz?XhK^}W+z?bUf+#OzK7~hmrgE|8P%L_<$WdnRpqX<%?PCg zSg~;zCPtY1ka_O$9MQ4|}5eo}R+Rp6I)Ng>O zF^lrX-!+bz4BUz4^~g!XFnrI>9UFN^DxLPW2<{T2^Do#Z$}XZpc+4d}<~-Zn$cbUe zKi|#N#M`#*_M&(~=t{IOr{60AJCv>x*;qQ}Iaey^1?GVrTB#{-YeeeZ&L={ zL5^~nyRU~)zOzT|x*n$i2`ejU)xH28MF_1QziyA($<@}8h~qxlwxs#TI2hATL?088_;f#=I=c&1ey}FD57OI( zDL=sgXgAYvIu7+VDAUK?4C>BlILn#O6Tjm;S>zOw`E&cTyT(QMMJVYnN>ts?p~sRm zhM1}b2F-tK8`Uj0ahk})I^WgLAa;4t@`x`gtJCi}bcy}TxxC~Lh5Zj7a0_h5o+iAJ zc5+)947w{a74G{SWg{)I+#$mV2m$#aar@tylVX{{J;alftp`zyr!;5`tgiCE$s{P_);bPy(c|*g}g5FA~ z#8EwE`#_Ef_uz;| za&{$_USs1iv#0O&AYQP+`-RuV4>-T-7G>?MkUQso;%sGVxW~U7w8XR^uF*$$`f?{} zzu=JUEv-#QU6XloNBp`c7q`}6Wl@#=gjX9D zt|rymx>PP7s01J`Ae^OnV-ExwpR2Oe+9CKk- zK4Eu?(P5`J^xDJccV*`(C_=R;X1R6gW8-tqE4*4Ue^= zD7sUa7OW#J7jLp(b8q*Hh6jVDx{I!tydDq^QE4(nU(zMycb)Sm(qY_aJg=SeBg<7X zk}Qa};z?~PFjuT2dh26G*x^nzWep`T-f;nm5mA7 ze*BVb0gU)0(bGz-=X^th)`gQ2%H;K9s~p_1XhTXYpA%QBG|gTY9(k`AuUUCox0fU@ zMd`!JB6>m1>}Xe6a=s|9E{uPpgi7x@s(SB@@*@&cE{V=2qnU$9eNw|MPj<6IiZ>lT zqj|>%F%Ebvz`MXqJq76l4&s*ba;^(?@ZGP^c$9UGhii9dA_2|fIXMXzZ-b^;!A+qh|DNO z$Bm+$m1^k260AhOMzZYJ?^O^po*~s9s_`lRwin@}<)k`iovA5E7OMuO{-akXN2yK=*4?+g^j0I)|(Ycn!8)M$JF7A>2G)BLx3R1AJz+O19>$2GT8iqgt423wPf zW(qA{zhJOe!MZu?D=jEaBU|cbCUVn<#sK$w&sW<1{CXTac9LEpXdIeBQtkjMT29-P z82!@UjGO|z-S4up+I0ly`|=a}3^Kajj>&KY#L*HE9q;6?S)JlPML}+`=gGVDb10ezxkMl6yM zZ_|ILf<`Cn7W-7~U=7uZ8gHw1m2H;wGjFhlEh!T2UoF4v`It8_Xa- zzycTLwsnDDx6DqTLHKTX)+UC%BVu=&x-j5Xn>zh3hYMcoY-^2rCMLEqaCUi&MuSZud9=dJT!lGJzt~3EJ++A0*I|pHeYMLJq9-N>8ygbejv70g2%nI)a{)};A_bRrA z9*M%qw@L&rGGQsvduNMn;tp$Bw^ zIIM*i_P|tZW`4>OMc~Ir>{vFvJ6I=CQm!C8=kZUL^pYa%8P{UhAMmCt&VM!k3*dS_ z$AZEo1IPm@@vTyu?^kqAF04>~oC+8I)~SE|%BiVq2q?LF4nGoauXIvx`6v4-A}WFO zgCoVuI{*Xx0MMhv|H$t?Q_IEiTwQ*T!KcUKr}tpAUCd!$jeVuim!?d?0NaS)v9LY|qfXczVihv{TvNNdyaPej^>g9x{8Vkp3_oP#?yjnc(cmZWo4x>5umcZC;pZdG%In;0looLN6M|BBTd{=fB$QQQ8DE)SFC`?vIGFOTOXa6 zk!!b&sDDgE>CPvPqF3<()pFWkfBBW1zB+&Z`jEfK_=Fk?T**DOj-Ut9n-W-?aS)!hWC>AYb&L*D&ut)M$-rrT{g0XM*Fm@q-H#%h7ev zrg+5t{nGpp;aK6t7VqbBr%rAI`eO;{peN%yBwl_uuY-@>ra*C-3t|PrEb0DH|Eiz@ z5eYskt;>UHs&x5JFIY%sss4xqvA+;KSq-7S|H)PqD)i9nt7v~CGodS3u-35z6eJq9 zKL_7)AG2-K{CihE;|0k<$$qcDPs~x{+|F75-0K^liOjq zO(uM4MYd~aoevJ7lM;+_N=U+{X!4NKN;ypgK+Qn30egx6sB83<@hbgcm(!&kh@ks{ z?8<|&MfTnEI@sG$#e|YGQx*6l$le*sh{iU`GbSCHFimZ5432>k#jA zHI`TVQ_6|#RoEG4iAX42Um;r3>XgrHc|s$9?pJ*HwurG49Hfez0mMNF7X{B~8ZB^mq3+adSXxB_1`(Iew7rr*|ncDRY(2vJhg8lZ_lqH>igo3s~_GEkhdnkE9Q6^0~ z;{J@dvJX+i@5C|w$ZU?}=YAYd`0gMFN(st!!Vhp%A66GLkN> z*}W|q51R+P-)GgWVbT9lvtG8z3b0mzfIB&h0aKS51 z1S`Ykg%GZ6P`f>KOuxu9(A3ox1YQ~?7FV;ivY#Cy@)ZzFY5$x$Cht@Tty4yJ?$v`w zv2gs-RpW)8C9!>i7q_|@h}9lxhx;&GcWQ=ZA*!ki@3-=$;+HJ!ribiiFM7^G2BLv$ z5ly0oDO)sMP&0icM4@JY*h&L(n=V%|Ul)jf|4pj>=St!)2r#I084+Hb{l#cWz-#;G z);7SA&#!n30z@*T&=h6NTVFM?1ziKUzHbc|6@Zj^QcTIv1cG!zhu{08MR*GvUv=M- zI(q$@sOV@o`m_amAS4Ulm^f?m{5s95m~D@We+i{q*=w>5-v|R5!_l>uHL!Jj(74b- z7fU35bnCc%&v2iNL+)7Wt}kzX!_%R&HxuULsh@NoOZ$Pu?Y||DdoewbT#X?^v2Vn%u07#SXdiuNuvvo6MsO+b}! zQ-Ej{5x45nSgV~sttDxIyIcM2O(gcUAT=nSg?UeL;l}`{{~0R52l+=kisKJ9A~t^x zY7!rpZ#0*_m60=Wik)s{g);q8(D~AfoY=Ceo#CwVuf8MA_YB56dsxFCAx3Yhj6{99 z{fNEGX!#O>%M(xuU%-ZlB{#p_hi`N{>4HizWuf`r!}NuXFJ+RgFTWy|=u-5mesMN{ z=NUi;8FnKT>Y2|-w8X%OHe}E1WRx`nJW|N*YBBW%^aGTf__Txr>T$Q<$q1&jX*TsRy6Re4Qc6Z$ z{vDsz9V)@=8=QuzH{bQVmP&u#e3K7KVj2sR*+~0%)0<51*52iOL)L@*8%!U{wH+Dv zB&r@qzycEGNTGe)2VY3SlK|?Y(&uqjD9*b^C}7l$SGg$dxWd-hc(J*%N1_Sci}rHO z)GLI#-21Y+*{==VR%H)iKJb8T;|$6Ssw>`g`~z8A08S_Wp&0PrcbAqgOhed2*HUa^ z9YPhOW)=(F zS`?mzNi6_lI#&Pi4)0U4uxadkOT!)DA(H{Akn;4@Suoeow{Q7#pv||IEsO$#u_Yq| z2Zcz#GY`Q44ale-j!^_ZBIC8x*<191BRg{4tlJx1*GMiB5Sv(TfKl@QeNZ!|QxkHW zVJZohbU7Fhl~w+cN=`ZGY9=wo?hWetCd7Va!vBV^qf7W0ZFUC4Z~P=OWlx@*x~FL| z@sBF^O=6W?AZmzqXur7)_9BpK2r8cZLebKqW;U`95gT^$Z;}?iXD(5=s#yM8&Th*)F{BzLg{?}lLStK(n$9^@b={be&xpqd}KaIPtfwDE};7bj3 zo6dSc=P|wZ0s+7M1F*XJCml30I z*GXfRr%BRY8Y!@d$_Uxk`8H^}sU*>=js77Cr#JLW5xLN|2Aht(A8&N85e#FEns^-I zRs5^vccilJ%L9IxS$~qkFH$FEvzmp27}pRLmXz~Q>B8+&l^#PW4-2+MdA+}?T()5n z*cx(C>h&+6Ezs><7_k^u-LDaGkbATwm1~DLsIwKGGEm{mIiW7nKR4kb=N~#gQmbw? zj4S*6m38^%lh<;KdxFg!k9?q=M^hCEUU8sA9c(~k(x34KVQQTO8x)8}&QV=GsjjVk zICAUbjyJO*tKe<4Z|WAz)9Ij`y8R1$oKM49TBh_*I zoj5GcYyW<%ta3+rv0>y?pj;?c&rk^&^KD25=-I_70}>Qwk&L8Bq0JHNw~J*74*Hc} zBwzdO+DsJG8tYI5L^_|jo@w=FrR_r}EW~=BcAUgy>9n5LKmBUCvyD6G`We;|$g69M z!yn{8`LUchGnHs5I<(C8On4KqjeqaCWJqI8b469JVGNk*6{&_h+$ROkC z7&BkDS?L!g^n}4%XZ_nqy7D& zt^!;4(gQwS$5W9Y@C1jo>MJ43O~haA3Ua<5*AMJ_3<&L9U~@#;uxdomrM7#Y?LXck zS_L21B<#5Vyvzz1-2HYJ_79{#0=yTS<^JlzNI3Ayv0#*G1)OU06fQ~|qL^|YW@|wp z&&R`0XXsyp*(c2VmTS~15F1y)$S^wYcN?}#Jadl_al@>}!|+PtszX5~W!VVk2qgwe zA>UnB50-5}?|Hy+NR8{XcE|OWM|dL*3-roC(jbPf^%HEWe;_c93p&Qv`5MID*I};% z%42OtXyi4-eDLuq=N0$^a_08PsDT0UoOg1xH2SSE-a_gGh9XE_Lhifz45ojU$J_^M9 zuH(nc!JIi*ao>F4VSTLgK+&CoL=nS>+LWCu0_VNHF<>S&?< zUyx!OH@mn9-b;e=TMveLJU?e(2$DhbR1tnW7wcFdd!B-2oU!g*1ib&V;{M5AnfZ&Z328{8Q$+}`E)6nWG>Tx{Lxnpb_G+fcw&@il@fqUsB&Rx4l<#jG}#5M0uOd#{8 z03FT3eE~Y+jr@I{_;E7zG_3Pl6J1Nj74V3c;)hoPqnak}8mbIgWn_~PFBe5(d zMe2@e@AG}7X>oiydGS6OzL0rubKKP9;mk3WqeoHp@HRq~k$tH{eLX$mL@;BR2Q+fP zU3PRPAb<_F8ooe;W*1JPm4i&VP>zGL`>P5zW7U<$bD6Sm6LRB4LtlD=RO(f5c)XxH|W8T zTWZQhzoP8K{jfdt+L<3O*BY(>M&c?81>R$@+UtE1G!uflj4K_KiY&N9ImgoEF801% znCXT@w-3a2k1<1MGAk1)rq*s=g4kvTkt%d0S@H*a%koKVv=eCHhR*?IZOEqE#Ja~ z;7)U475C+(Ewrb7Vgx=|B;5q3QqE0#r z+cgl(Y_9CZUAc7*cbs)&9SJ{OUQHQwNE-t*)x7`|7V*R475tuMg8IdePMV%he>C-@ zUWHARG|Req_w`jofq#zvGhQ&h^GxZSW5jA9ZnF5h>{0%`uk9=SH(PW6p9 zZdI?r=X?n)Z%#ed*As~I0CyQ$T6~<(s_{799achoPlAoW5)hj19<55?CAcO;V1LPL z-9kCS`J*M|9MJZflKW^-*ukR&>`jEX$MTT2NiHv0DRDtrUM{LzesU(iGP7xKS(jh{ zQ?pjL0ISeDoD9+&Z~bHgRYp!Z0%3AcX94EH*4uTRzb1Vyj__r0?kFq=}LT{8w71#nmM*@~Po^1a!aIdLNNLH==!db7*wS(E){CetsY zI-JT2i;SvPqHX+(f!8zru9ZJm(}C&^-@cI$EiU*T|E#+?iwx>iPK?0TA|rGxh|~oH z0PIl8>psf#(!n};CPAtls{C3Q{=E}0d4Y#o#-{Zxj&LeCnZQ~0(M>Zuu6(^yfntFs zcu}cl!e8rxnfaqty^3^0hD$#m*g{Z3Gz04H3lFG13*_cy=LjT5#YBHKyJc#-6n5Kh zwTA)pDrM{7(o^1MdzkA^9K4h|uJkq4xo2cqgmYq%)l`Lum$)=7$aKfZ(#YQcS5_RA zRzjYIvy13Tqu%7Ny?QH;BDQVR^1jMS@6&w#^Y7)?!9e6p(6EDRE2STM!dE%v5_e` zv%RD;#1t$qAPhiFxdO8rGwks3CRp>V+&6IIV=s}_PR`$NiZ8s-X|vQ?+J_{Rt4~gz zDg74uGlD7>FxUG2Dk2@2?-0mqOY1wM@o@|od`(Z`#HHOR^K0Pdx9Q7Vr$Ap33Ysu2 z3tjfkwK0%7h`YFa^YSz|Xc_$Ojyy!?apDnpf=zxM_8_T%^vV314SLIp{#G>jX!phh z9$k7*wrXKCtebyZd$suHpV=klnm65foi8<7hyqG|!*6t4j4w+v_4YE*tc!86WdcP; zP(|NG%W=VR&e=671{BQA2}nzefsMTSnGskp_SW&o9M-e(I{`4bDXqtZ#Wchig&?4F zNdn$VJ8)Z_Tej)@>ML>2L$b0@1WHugKrN66kmyKp8G>DsG7N%@2u7%kq)oF!E8o7o*anRwq91+^ zjZPaRHuD1BNP{+zUy)YBr*F05!)V~S>KYX3B<7*qs-EknGYtQ_)1M+RU#zXn6F=+ z>3=(Z z+;m>Xd(&Dy`}Gfm5YzM$;B|O7kR@K-A+UP@JHQvM$I~zoZ>j7QtU3ZBT?4=2iL6{f ziJinU@qQsda-}jLiytMPrh?L40Lk*PZZ0qVNLE=s-rFAv-o6rl%<9=Gyd-@4=vMY~ zA@4lJU&KZcex{}frMBYzo-elc-;uk*_>$Wq7h?1U`}?tQR5%Go1T$)fotb%MvaGf4 z5XXO+nV5gK05E+9Aevz1S6f_m#FULN)RW9Xh{|HaE%rW#WewJ9BIyLVnwgP~-%mDw zpS{3%2?}EId$X%>izy+)rFW?lLdpQj{*g8Ox7i_sOsG(pznmZCeuHAIVk<%y)U6SB z1$_1PuuM7@h;K7BBD&DDW`>eXV7I=+3F}S2l#b2n8SjI;3{Km8@5CiO$FXWD?xG!N z|7~hQV>-2-q7R^#S8{dKS@>GI0!gu5oDa7HlV%b3Idm)Y_LGMbmVvq^^Z;SJEoY>7 zcWc>ws*$wjH2?BiV3x>Z#Hm=rw9o;YjyUaS%V^l;=;mo`RC4~c_CohFNx~-9fP`7V zVqa%qOG3`iw&RdA#BL5_&RW7*Z69^J-and4xIpP5PgqCypQAXMDB7qdXdF#*{==Ir z;QjD~NyQV#!4f5(bvwXbA*T0Qw$aD3U>typBvTyQV2E?>)3-X4kwbsxcMiPgTLzNJ8!)33U~28Uo5e4RA@KCxF!!XZ#;~s z1`n{5(#etQV4mXy`@ty-JOLZG8*upS*d=1Vv?*;i-9fb;-%>i_Jj2*bt{?N ze~mHg;(A9EY8+wYm>%jl_81VioTY-d@0zulX;;6@WE`Kb_>H@A=b!1Yyz(0usiv}G z6t=uiRL`9!(4#Xbv3T;)Xok}L#Zu|u=&mE}NT+_CUrie3OTO8i+_7!{;NLp(ocXrD zv~>GS3g>+fasp#P238b^_m*dxK%V7_!4Htf7v%H|E5$rJR`z>MtJ4Pve5qx|AbEMp z8}qQ!bclw7X~F%5E}J#sx&o(_&`1h9bTgk_4}ZMT~Xr*~ZFYDK|_6&y~bDcP3mU8VJ z?s=-FT!VLF#A2qz`)ZaK*dRXh9@zJO1V*!}hWloMg|<&-Rf%ih?}PKldF4=_xdlM3 zs@|ORf=B(z#Ffm$HD&2@>%3;A2+cjjtoQYCgr3Yt@iBmpxN(l5H{@c7LYaZ$Q$e@+AR`wa`oQ#RG`^$Gj(Z=;@wpJ&UqM>%QKBN%jS@j(PTWmfW?V*6#RAXmS0DQ7>)C zNj;BnC%MGBj$aM}owP>vuAheK)xNXf%kBvRAtJ=VMGnHoh&cD=H?Twoch^w=nDpLxy zr(gTj&iCgC*iHs@L0vi9k{U{14lBLixODhVgSY}ebzeY;KJ=t~p}_jFa~kO?6YpLa z2e(_pES3(QjC1)`Mb7`$$?y>lk|YLF@>-%#2Z!)|6TNIm$cvm!C0nW$yCxw^gqoPAFd;*tVS($jzRI{|Z~0wgcz5<_XZ0)KLM zLfC{6X}Uz!bmXNG>GDb2;I1Ixnk(gBBE7RpoZ?^Lsg(t_>MWNa^h3CL9q?$HVh)w3 zJs1>Bt?W=2bF%ci49@zfaz6b;Sp@?-S_G0)Oqr1MCI>g7b z(uZyj_Am5-w1T2VkjIVCrIWYCu zbP*7v>K)-Pgn2zuzzoN2JNT2=X@C2^By7i!xy&KUSkF_Y@7UM0XV1brJ<5N+O+hz} zm7czSJ{tZuqx=6w)4FkF{Z`@ge!mB;ytzE#{Rc%+nX@I1Z(}CTSKq*n*!;QQAdBWS zHoG&dClFl{fX5*Psk(o%wE#c7)+5Yewiz#IE^WVmg;cud)e9wXZM@5+3-@?GX2dXu z!l>5pB6-WI!;f|eG42=jODx{ue&>A4iu;2%0ixH9ixaj|y-cxlF1XQ@XX1^?dA%-7cim6hEY$loA3pnSZKMGSk;`q(I(%nq73 z1S5-hJK~UQjkx~zPFZCM1O%)N;zQk1+<(}%5c4;zmQeo-8&>PC*%fpFdWr5T0^w-- zD@zx*lAoJpixF{hu(~Vo_g&aPJzNqluJ*zZNDfs#6P9GO#xwXHrZ+%@@-E#^D-5~K zG?MY&-r;QVx@!S9Uyu!9D7n36RmB&9>00P#2Gu~mQ&M+Aj{uQl{r$7bjE}wqQ8hU% zSAQGL=-+|AzK2Nn$`-Q!C&Fq#DS)rj{e$?2Gl8M#XpUKcv3f0|Pb159``c9_grO4- z-NOvSLdbe+r4q|a8sh96wn9H(a}mvhY97E9Vu9J@&aeeQNlPO#G4$b3R(Zbfh)xal z{p8g@A%f9uAX>?EUeXXh8TvdR;+)w_eiJ@Ko|%+xc0Te2hp}h?F*y?tuzpW&#z^_1a3L-c`H*`)CqH5(sg)?Y;MQUo%6iG?k^?C^NS!CaL&8a-79 z{To~~ja;3J$&(ySi37a2c%siav+s}$ta@~r!zeExx#Y&nhknri&Bz|@L>(G|q`(U{ z54fU#f~Md)PzM>x>JY4V7S8`zKuX%6XyzZO*8=h$RC)_qN4Kb%jK^TjT&N1pp*Bo| zxEMd?QQDnos&i#fg;2?B&wc)D?L~A9XUTPYaqo0%yGjuT~c2Vo{2o z)j!Aax3*(n3(uKFS9PTOA(1CDfTZ>eAQ@+^#II%j5Az!1E$2$qfBge~y85K~PR443 z)cWf_VqpYJs3n7cN~;1onm8U{Qsiq%!Gf_d2>?6}-Kq_^S`wf>Q8D2${R+nsGl1!F zx*M5kFDV9kg`c)nfFUB+4Uuf8v7F@WTZRWeQs{x(4vLUWf>+$Naou>-`mg*%-TB6vA=@0hWD9Au(gvzlz;5r^#Rf^cp$R&H{Ne_P6;1n`5$TS(&sX>%P+pm#`)KLrAESHKT&jz_u z(9DdU0qJR9ZqHNlWE|?Kj>jGR{C3KhG_S*fCP5k>a^}lM(@4K(NC9FASn5Zd zM2EOZ9B3}Uxp!)$4#|%dN5ZVUSFD{RWs%v=LJYR_J$GP1FYf{l>=sUze*bu6OW@J{ zct*Qf|J-w`QWdXQhhDmUk%qJXl~pSz6s2^wK)9^7`S2mdq0-zkJYAsw9uueQV)ke1 zh!|^g<11K1T+qZVpHzD;@wpD_43vI=b4IM@eSVtd^bkDi9l~-ZVIgR!U2k zK}ynib5ArdWd!IRL!Fwqgba>fwR;X|9!Gd6 zG(4mlZ1T|YP?fiPHd0%3Vg79>P8nTGk%|xu6*v+eqkG0nR@>JUMHI6Pz`iU?{V@vR z$1PFI%uxjw_DleN8l*R3|ruL`J<@!bXF=mm&~ z7n53mP5V9*bLqWFouBuhqyN7ozu(8R)aoWbEJN5P)brBnW~-ZNb(E;ZavyBie;iwz zy^q0O5I)lwwY>a?D?b!BdlJIinDeIt(j_rP(>3#9B-lxk{)A0Zhbzcl1BS~2g7ciA zRl27@N0kh#&#AR(zr<-)${{Lf^>**#fWMR45lkIV`;CQ`wBFyn>FfDr3!`;}<>~19 zhG0{R9FU))Uivm=*x~)yYNHV6L@`Jv$9+Rei)!tE>ESFYkEkX{R*&?Yo*>1&`;P!P zoK@Bs#{K6G2CHP@Li>B;M@SU(ju7RGE@G-=uxhiz%ABYAv}=rAB2)Oi2>J2nj!dpe$ezZQErKS&W6Xc=$K~ zyqbn?>V0bgB;x0Evw1I22hY^l0C7Qt>sMjSceUsQ1}iy(@nER#v~2-e=k$EQ*v_-f#DrT z(BeZ|7gh~?oCf4*d%;pg&MJjz!n5nmSIACX;DPkCFq~`3Ijh_;Pn$hJ+Ls1s^_q_a zst1m9JGuPoN(+G-|DX;FqC zH00?s()%_Cuen!^ukZVRugMnJiWcbZ|LEa(`8MVq&tJx%~kgqTBb(&IwYVIs`_WoMmUfi243~qldRQxZW2`s)5)VjOt5MsWN6s_*D>4pC#;vsfh z39U7cAj*H6Y0VueKTucw0K7&uEDUa~qKRpxVr|E7E?Z`fo4Uzxg>!544B!f1;0Ry} z#311s=fh3r?`_sX70eg(gOj3EbhEq|(L-0C&KyM>#En*!obqIuDj3k)dTxpw|KY3-h{oRBU?hYqnP|8IYg{A!88VH#-0ExQyg z&pfVsV=iOWUcwLgviF8^#OErxKN#rWcRG#L5fP4sj~^O`XPCzp>P@C=z2ERb{*Cxp z0Vgg7QKE}&-QTnyF0gmG+z@{N?FQQB{*??j0=jA@Nw4+veVC$mjRPw1E_tkkpF2pD zRMYX}G*&SE5}YUP;hqAO^SS8I>ty@SE>=hou6}!>6{LNxlH&g$g_|ecZyPyHW`Kp+ zy!usG_bV)TBbkl8s8=K>i#C5$PxLQrIxS^fX7_3Y1VQUS2n{;vYbqo61Ibl3tJV*O ze}ls(h_peF7xiX0Xr?PM3_EwQ` zWaKeYr!o>sR@sr2Rd$jUIoT&$GD4A(Bzr~2gN$Px9z@3;(Ku#^b3z#*#e3bJ-uLtV z+aJ2`bKm#xx~}i_9p2q6ypkt$B2!)-3!fxI{c}f7&fdpyYP!h8YcFJ7x{#UN&!wxH z_isT;gXS7Y#lR@O9(?jKQ0Wkx2|PUBe4Qi_#Z4PPM4W((ItZ2{#Spg(Cfr58V@Nea=wT%0C!}}Ob)n|Wz^Ii1Gy-kL1~#*M6q6LHJ)aj?P=VD8x}C;$ z-wkrEz|BFSVMYR|bDtf{Xz6}{P&|HMtk%u7a81>Kyx~r#7OmXIq(n!L%R?gH=})H8 zC$tloC?25W)2Zw+R5jJ7Ei^4Qk(v0Eq8%d(QManas0y4sIY9-*(CGPJ2f8HO4bJM@p)Q2QFiUX$F!d5ECK&jOO^f>tx|84YFF*vL*gJU>;Oh`3LoH zEul4m`#C+zMl%LJLAoKEQ;b&{u~mNI^7NDY;s?rM^KBG=P_hwuYZ za7vfY`wG&-t8^%O#di~4Mymh26bdsuA5Ga!IBSN=)!dFUX@x~<;`dRKWJ^1Wax&=_ zUP!-WZ4rO~I&F%Y#gva)76Iy&^?1N%cJQNlL*!Pp#`NJ}Q5bMc?Lg^mKqrw6v*z

    Ww3g{t+$UAUL;o`@wlru)H6yT}BBxF;LwNdY47Yq#_010= z(25PJ?uX^G`SvtFN9bQ(vxrfkBTjausU$GFPAP|R`_j$PZCL3abR5=luLqx_w-6)c zD?_IZ@+6zrHnqz1Ac*o}IW1xary1z3V+vveZ^vZSAs|-)l32())7b!lT0T3Z(qi3_ z49&C33(uSTd9k-XRbNW6%2K=4%@2DJ%*EccioLA>?g;;yR-08P#Jf9zPc@Bbiw-=o z^=VL7XySDc^56~mg)PDy<6O^AUDIY+n=x<~rWm&bkP#S?g3k`1o>VBexz)Xup#6}C zT}roqcVL{?E(oJkUs-yXIq!+KLKT9qg|{6f-LStdWWQ)fnvk|{Z0xHh z2ZJ(gULMF6*R3yQ(Ks&r?zb?lE35tCu(1wKxLezFS zq@k?Xr8kjyt8|^^^pD)JXTv&Kxz;Z0AwVp-{dyWki;{7K!R1+rWV>DaRx-(ebSY~g zXz$2KUQKwv9@3&;rVIZYJtG`L72yDwPDjSUJfXZi*v%l~90aTO`yHhxtpen~cg5V@ ze~n`)C&8&IcWa?!88Jf>^e0V?@cpcI8Mr`+=hB5=JN@QZotbo>{w#a@eN!8rg(J|+ z9zjPJ)R$%Q{I}t+6+*@pP+Y|YK|;bQ_mLB>!$nDV(vT?goOq^68;OChlj+&=jiipJ36kAsE?@!i>TzCvp@FxtT16D=8kscYtqn`-4;JLf}^y z^~d#!Zrj52JUe|QOkV~O6)7;Lhl4GJ>6!AH5)9e*>W?EG_Q%)a36>J8#v!PC+VnBZ zcZ6$AUNWWZktu)ANixmCl)O+Bt@27?{2mmSbdHG_N3KN(fbu)54Dsv^e1U2hb(@3CQq?{nDjeHYrC>60d<+&%t1rfECI6Fb{DWlF?3bs`4=CqY>KI)&^ns zRmoirk##D8DlD;oH2xv>Gqh79xvj)tMRJHO8WbXXh3m)Q`V%wX^InxO*v4Yd7Y|K8 zU40qYt0>5+&ACaX)w3k;M3KkVcDcLq01?YMg4_;v-xj#Vl=ab<|d0nikZSd>QgfzsA9ymO^dcI9#FyLa;CX3K!) zo&T2oQ0+88E4dugeo&Z>@0BFVZ{YN>&T%?4S(HW$*u@Y__$VMtWNbw2{FEaE4fDYg zXz@@}2>7jP%uhqco``FWHG%ti#pr*RPYn$rdzZ$qKUlcWySJf=J5mvsZKUP#mk zul39tPRdsva5iUG^N zUF(5$UKM{cgoxz9!4k%L;zqx0`A zYlVYoR^;h!$py>#&R&!p8Wka7vVFLKm^ZtFAlyBWf219owRHZgl2-;l_hzMkA0&%Oo&Us zir#63|6UT(M+bv%QRFjFM&;beIIy0aPPf2UFdd_L*U|B{=Co zO?fMs{}F#H8b{QL!LA9x z=-ZX{vhi}lGU%#hEt63>-2RS^x%`lF{dJ{hJxWX<9Yy4*&%%Y-&bID8K zr2I->N~DVBQs!Cjg6-ei{^7@Ogen{=hsG?f&74X00h{g@wI2iA)keoX6IAv|^~cqF zvB#+VlbgA6!V;_?USD{#0jT@gl`Bs^eLJ@HX_Vn_Es|2*@|jv2-PsI`zx?XXL0WnG zXPE&<-h|26vvE0wX$Qq>GcLfq)jrHbu?w6@C6mdV+KH0bDXx7xTdUE~1L^~BKwO9! z?bZSexp{cyOwG;BP5YuFuVUQSFq^sNBwg|~b#+=ZGc#j|)e!SSnS9+8dAgGDl$ma&!5X*i}q!5NQ8JHKTlVHdW_7u(rvj6#=oNJkASxy7D#s zx3?HdslWPfrLDtt_G_|baRe+3iU&Hpd4|3T=S?mvY9IpsZaIBLufkH($z%+M;d5C#_F>Pu>p3ZrzH^FDi@ZNtS{UP)`0gbB~( z*?j6XXCXb;GmAI(Ue$}&CIK<=0r`5TJH$W7?-NuM%gsw-5n68Rl(NO<(hX$Fpn+$# zQPgU?JQQP9^1!`auqMK`Dd?=}y6YORR%&YM{l%v{ije$$K-+&eM^K?aH62CUj{2Ly zO9tHC;Tab>Q@Nf;603^CAZD54Z3ip?VnoHebdC)QV_pyIC};Yqg#5C!6_VkMTktVP zCO#SoD45w$g0zYQY7GCp@1%=cTdU3=|8}6IcXjyN6Uh+581~&28u1p9ryZdZ&40u* z5O)V2f=QcTJEeb`@m{4g{JUg#f}jk%F&1Gh$r5{GR1@@V05;SvO_)qXzm~5>5x$x+K_kua%gS0PfC%UsI11 z2ZzBftFK=o55vyo)2{XL@H?Ss#}X0OQc<6xn*y6}J;v(ZVJ1tIp7-?hfZE2}6YLkV zcVKc~T?=~ueVt;#SW4~b$y5(p&bhS2pyvHTN?yNA76a~z4-;S=`2x59-0SI)`On|9 z6KWMaK5~r=WrcM7hV@Uedd7xvZDlX!7eh<0ySgBw_6mpAwe+g#rqJD{UlQyn>1IJ? z>@ddJ;@Z;JyGwmxn&ESWgxI|+#IxDAApulv?4r+<0#6~uIphnuYPTnpFWGMjwgDB} zu%8{L??n>jG2?pWheanaTE1sDa^g{r8UaA};XGEaETs}O`tQ4xuyzx%kT4VEK(V<# z_cjJt@TG}23U~KT7%}Hb&_PYPFk@3bvtY*x_0vSYfK;MQ z`J^ni;og;-cCTHGMbbB;L}w_l;*Ra@iKzJVsh$|FJbiOg*g0_(72PKhVra0IagJ_C zx_;)4TYFCndJN{TgO4$HGLNPO)Q9o1y+1UWr_4}r3!Wny<>KbX`CKXC;_b0F^kab6 zUJyRPbo=ekBh*VoB2lA2%C_Omumb5X?zA1wlq&di@0}a!eINS0}!k*_&lEWT#zX_)fzTS-K%f>^e?!zhvE#Lyi{qclr!IgBBxg8)y-X8#_J>bz*)}ao&ZQlwZ`gYkfT!A- zbb(Gpl@PpbK61a^>4;8?G}zMD7GBrdUz@Z;SDcZ|G%}L?omGaU`&7@+qVhqcuZQnB zWkbE?UV)+qK(7{O0I_1SH2URvc$(f{61i?&xx$C}rnQjr&S?Ce%vst{>fX8jc*UPB R09H^@>1rBi;4aul{0Gd<&0YWi literal 54499 zcmeFa2T)Y$_bm#D7*P}i1OXKTK|qw86$JzVC1+HUNS4s#AYwoPQF0KJoE4A^jf$Wm zK|qk$NX|`YLPG<0AIF*Lc7E@F|MlLjSNGnUDyK?XJ?HdyzP7UsAZBMn<-!8vY~OP6eM#O6}nzBcnWMDJy%$N>*Oh&f4y_y8U$%GkG&xv)h&? zYVzmF$V7sIG!1Va(xj6rceG$X)~IeR5a5(9O?{mDnU=td!ZM%UULo6vOF~^o$4r-I z8lyfnikvN9f1!HuiYNa>EL{z{*!M8U;@)nlLcZ{w$Gh4Nb4V4g_+J@&lsXwLem2^& z>~){njORi;_PNJQsRefHSZStCL?|6OeM~M7!@hso!@#l5^{>=SPmR^V7pN;lD_Lf2>EQ9X1xo!`)Vkg^n@Pk(v7iV zewi9R_mUqb9q%2byn`&whfWQ&KG>?Me(OObO~{uID{{Nn_+PafCo12X%p4njc_Y@j zH}SPS_WDMG&D)y+3pVf5nxYe=?qaVStZ@q-AN_H)Mk(`Yf*fJ{fR&9`(V_nCh}h75 z(o|1+`eO1@_hBy>WOr}NtV&4Ec@ucVd-Sf*!l?nC2Tj7GpVi-f&&quEA?~e9O|hWf z{Jc12^}{m#n7O*!Yc}f7=SPPp1Y_zw+%6DSE8G;8yly9mPLF$zE!>UZ*+xi-4D&xw zR6M~ZU*5tHYNJWRX?@F%?kyO zZu{oxQe|Cz9HiRCQ#vt;zxtt(!85@#Kq$-4r;55@$5xK0)AuWPD(h_7XUZW%g`xj` zGsAxm)fu`Rx`TnnFU%R3N2?{~#R{Vv-d4Q}+xYm=*dJfgSjc61U+NBhJ`*80o4(a- zE-S#mgjU-7lrV+#iGzptcltLOow>9W>Trhqs9T_f^3t&J2*{{(b9)4h1aF`p?-Ye%F@hN&V z4`(Q$Ysb4sE<1AT>Mgw1L6x8j**JgP{lZg~C%rhSHJs3#7 zC-6O&WZ7fFHj=hxP{pf7{&4Lz1x|TGQ&VHyK8%w^%T8IO>KyIkW3r}?!-O=JOSBhW z91T|;R`@-yH^SoaXbje= zT}oy%i_9X(hdeRCQ*(>KE-V?D4B4d%=QLc&r+VF9X?~06n`PVn;4n>^OmoCFj&m9{ z+xBqOj^4d{jk_pF>dmWX--D>skL*d|*NRsAs+00Ch)K_jit=&#;vL#sZ4ZvJT)na{ zh<4yY5=U+HTe*I-X_tgJkJ9BHj~`r-1=3j)x0hZmy^^xpTrk5{x@0-~aKE5AEaOW? zPC>c-=WqPD7;YL4Pq#|^`N1E5@?^*MYFCFx$`SdVZIn^IG{B`F<8VU;gBW1emJBXi0hRlVU_-9|4Bb1c==bZ}V_7gv~ zFWD3m@y}#_=eKXK4j3~JI!656$jQG8lKe~&KttoZ>uUqkC88A|r=WOcMe;M{&LfA( z7+%H-KK*;0;IS#Mb&&kLo$kC08Se#Qt7E^u(;pYvUY$$wGtIvX_wT~}yW{@M8vlms zzd7#T;_z4X`nNdzTO9r^4*wR1Uy$*CEdcft+P+;Qt2!w}iV4-{W|6I5oXZvSJykhY zWF6OP5^3m?5x8`d@-MXNm!~g7wo&$h1#3ksvhd~X;CNcXMpy8%^P!vHGBq6*ziIoh z+`qm1&zD6`p?UH!nYLl%tt-5*+!rUT=c7!5AKDc63K>`EZdBn~l;%dIH$~@#Y$Af} zTmEhbgkJlSRXh~0{(*M)+gwL)qCD5SF&h(maR*pm7?!)}l5OnVd%ffDzn^JTeON~m z^MOf$1f^eswPN`{saZ%=KBA|a6>8O<{qWMW6CZerd-m^Q;C-wrg+I2`i^7P^R&O!+ z{^~lveSfu{zV*r!3#w|UG8n$E=S))fxFcU$|9D@D{azmOfok0#`O{km6TRwfz! zh_6cIgi0La$pp{aL5s@qZp*!>!Mn6759C?hl50~p7mYU$jAKK52qrz=Yts&sImNYF z`Zk(LD)=wlc@_It zhhXs;#sc$(!130!y3gUZ15fgvQErD&3t4tNV>C_6Ias7JilIK z_;6?It}`!i>9b*+>Mvn`brQzDCYdZvb@F?z%$_0yvxp1HJ!ah3TaL1e8N=L48yxx} zEE()M8jq@N*S9n5YSOic58pcQ>Dft(#YS;#MafLrq~c`#$&sc66@>ZvIar;VFVSa{UaStUeYZoe-xzO6=}cvnSnmcnW!^TFRd+>JOZcaRRRa>VKyOF9moM5hq z%4guwqZ@Zr+4)y?Lw+Nm0&BDKP20HPLJRvRo9?2V^k;qh8kJVP?HnV`>LbjOu8xLi zr(wxEG(T^d%`Kl{8mtX^^jy+Ij%kc5h$EggNd?1KwK|c-Ug?odntRqM0$eHG@$*n3TPKwIo}f2j1I^CrS(VqQllHa9Sv-C@d1*&aUv802oD z&lB(MJSDjAN~@|gkDHMBv0=z(`5I_Z2zm8bs#?W_(~791g@w}oJG*Xl6a(sE@XIeo))M;bP&fZ(PP<>%j-lo@fycYTamE`y6THZdtIczSpIj zr^Vz)!(7xNY_h1J+W3%9!gjxRw(a>CFmBLl>=>4?K4szXO;h`WiuiQfq4I)%yaxYkDQ{oS zKb@%+()c`?++6nhxYq~=Mzzd8mes6kxzB^0jW7Y#I51(8Fez5TOOb+-g zE=%0V9u_gv@&ta9T6gSr;o*|u1l!Njd2M0S-;NW-ugsfWuvls5a>`KH=ZnJwA09DJ zwa$x;SN7&*CfjXBNH4|tWLnAwYj2FFE@OOrWL>T^6K^$*H!?>k>?ht4U_q*cueAz^;-O8U?|>Ab1~;z0Wq5e6R^9LGE_D<)|Fb(DA6>goz_@ufFqcW3kS)4mEf&TL6gKDiHq~|(S;RYQ1-big+Z#Mx)Xl3eEW`QMNELzbN21! zo15#)d|vv+Cv*nDSNTY6t-$ZwJ)JD>{B)sZ>r0cm;MXqGT_s310o?j&WUts*;ztbi4B&zuujSsa zWsCRblD$&I(U8=G1fsr$Ya1PubjLjqU|JML;~PXaRt5*Bi!qI5hXXJ;1(%&8ZSz~SSO`(awcYnM>5Ym1W%2*_U2~`&2l>a0{6`@L z6@R~#l{3J$@q_KgSn~^hv(I4)qhG)0-~9SgWU)=%1d3~LW4yRL0NVomL)w$I^TYMb z<;;sD=~~B6=%N+=>2alevQ@%DQ^(EKEqf(s#<2sB_=4%U05=aYO~#5X%#>qmcbtr) z36Ypr2vXjhL~n4nVAwWS>PjH`2cdIlxZ@C{_{Kc9WFSyvldxd2IPon-kgsaxMlzai z$gbqaJ!pC4=!cXJTqXsQ)PCo;_9IQ%uN#gi~s$8Y{6ib*FZ{?fU{gpKip0qM;J=>p{zLk!#cnEcl)?YvWjzT5RYbAV=Q zq03~EKjd&>DR$;OKz>H)(KyF^&FSvqkiigXDJp6vxg|pcOG-CycNnMG^p$d4aL#+(iSQ^+kN9+-9zR zG4BcGn6PhH=!TP@=(Bpa1`649UD%n4S*aDycOUoaZ<5}Wh~?cJ&CJUEJfo0i7HRD5 z<`P-#>D66iGYGgP*lVftq<&oRB2Xj|+Wg8yxR{*6cqiQnvb1Fz>9whX>`tTu8$x+o z8Sfb1V~_2tEcn0)a22IrZfm&ks|(YMoY_dN%bMikq)U^3rG6>I5Dg7~rxRWpz1EA( z?^KBw`>@x0)_-0zWxPE{zYdUdbVJ%w3JqxwQoTwMu%mihva@3M!86AavKNGd_QZdN zmX-FbMPqh>ir21Xs7pw36H07Hnu66#vwEZ}vwr=7{`LDk;-;;5X-=B7_2j3l9=Uw4 z(iZQtxac78T-h9Tw_wNS?;G_UWxuH7h$*!y>N1oq?JX~szCbC})C=1irK#8a;GqDG zCR5L?YG3ljc9oi0|`hnNStZEgR$~G2TYo8#g zC~uH&lvlNStNr14=n!tkI^`+!YxM}qP!;O7W9s1wl8cu%@&Yy_w-_6+iyGNydft08S_GP;(T3bYrk&8)fS5g zBX9h0l(3zVP;-0W0q*>_uWCcs*wPZbSKaS_WD8N@78;5ZHpNXWs=T8hE^->72r^zx zwKFQ}x|QXb--nqO<4hNl7yWUUHXjvOzZ~ATFPEE;D#r68Zo#nB&i&?s>$2p4G8nJq>{BWJfNSp68$2V+SxfK!Lxq~=_nU4It zN2A?pm7Q!#g^I~L3)wZLy3N`5dy|wRc{>f;a2!snPG13EF<3TdV!fBkLs>(4_r8r? zjjQhtld5xS6dkSUdSg}3VAqTT#1*}DW8SY;=%&2TQKIi6L%Vh7KHpYh)WSa+qg8{e z2~u7I09u2dsfcmz@@MnHT_q4kr9&VIr8%W5do`xe;=3=DvG+AAv<)wEEBh?OVzLXn z#E944(+qi|Wk9JMubzLKqPWXSIjmzGFiOsw24N4NE#>d;VaaD{jN(tK6@NEe$0MDv z8f;{g=GnbEp|3&|x_&Qr{LvoGUkzP27sci5|c5MoE!wDjaTTF+KOJf9#SsyPf4x`6CpdZAW7jbNz7;&{L_1sJ)i3(~Y zD%PfpjWf_KzUtwhO+qBK^bD&EEw2YWG~Qh8Kqy2;0H;fpn4Ovhnh@#6ikhD>1%w9phe=;vY*C9=<) zj5yc~op#ecY6{OGFqa~w-Qzkgp{~Ib)>*sE^px^Lhvcalgu&@1``{yeK1n?a(~5I$ z*A;9}i+195gFe71`RhxoNP!q!%|qIP4<1vm<##g+WxE_$AVH!s zXmX%TsE7`SF$&unQ~*n-`?EuGsau2+^M&{E`Z7j3SaRvRb=i_Mv?RL8DF0T>@kj$T z6KAHo6A}8h4{_s38fvd3p$f1EjA*p1)xkc|Pky!9qQqxbzek!dyVBX@9ibe#`D|un z1m&k5B{lR!b^wr0FfKY3$Xe4MtH7v#YD5DCHkW0*&Shncm%2=x%Y3Uhx6^TvrCkxX z=_(xi{@Ntluv~~2x}@*WwB_Ud_lR-Kqij4DCk{XdfJ(kndgS;tcP}1#Bpc7U4-5{& z;o5Q`tdd6|iB|c00Gg&8O|4sdH-D9egrRaHUV@mQW=F@$>e>c{V`5$@eCzM7X8ALTFm6E#Pz z@ng6WmyM4uOf=lR%6-lqhDjDp&zU{b#4u=M;*>*o4#24h_B*gHHRRh4Jy{>FuL{J_ zWp1Np3R}0C+EDLT_90~*1e6o)Lu7=5+C)x@~Tr$ z%B1%3$~x93f#oDi7SJ1C%Td_xKram7T2(_f*4MNp+M)grK0e#{0M{LL=JwD>5nQt@ zv$`e=@nh2|Cfbv`etLSG-)XGb72?^-3HNtXg*0NHP9_+bc&S?;ObjFBYYDi@q|GV- zgAY*L=!N@SO&$Y3%v*_TfiyShzSwd{+sN(siOW$}M$5|Rm&RM=$#Qqqo*}yH9}UV3 zE>*kxcqh}cB>cZWI_llE*~ey>%WPU&s4$4r7;Qa}xzinINAARh>Lk&!5qfzr|E z^+KB?!svr1-W3F=1ePgcw?F zhj70_zVh-BZ{XHl^rq*>um0dvPc;oWmbP`=b{Y_>$;#{x-4zx{he+!RbZno%SM)b2 zs1-^-mOrwGe0euh)NQM8B5m>9%{h@oZ-I&Rbx=aDO}e(!eQLF!cvq_Rlom0xX<_S^oq{w5P`8J*T;;vVaZ9iv%H z>#;>jScbIx2BB_Kl#A7UOd&6Dq3daNFDAL%$ZP@T90FeOD-^{*9~LpEbSvSAeMGP! zaTdWO0Jl9nmpUzU?x2ts4(axLBFX zw4U9%Q#PR&10cJs(cO)#54K?vASdVc*QAmjkW!b)yrD`dG0EQB&8gfW=m1of#Uq4R zjuLe=H`JUkjOaTn1=8Trt?lGnsM_@Ta&GpPIVUgstLqk7>;562Z8SlVW3jBx)?n(= zt_mW_>on>5iKnq-O;EOI6v zI@}(h31h%{@$e3k@XTq%~hK0}5E`6yb3X|5EiFmiE$ zO7^%&U?Mpatd}sj2a9FSllkkZSeg!A{9uzMo>YX&)Xp<{z1XgA@rMkhw3bo1_cstD zO%YJCC9dz2VS%}F`+y>LXoTUz9`UIss&Y&clUW?&a}#to)i0q^Hm5NgHkI$gw)I1V z3AGG<8$Rx3YYA|w6T`MKd*6CC2MCw2|IpilfFK{VkpE3ffB4zrMnuOY%(TtS)5cK` z%$m_sX-XkBVFQ=u(C@b`7Zo*cN0v6*ou-aA<0d0eQIeKrnc$8=%p*_SY52h$Z z%DEfk0g43~Z_GyuPFPQUY4X`v08*y5b|$rUX&6;(m3>;4_|AC>Ab6D!;xV}RzI%#p z@wJpA_m(2_e1Iw+WZRr$V`mE$tF-cjALa2ZqZPJgkJdno|1g<{eqLE1ajPvI0jUTk z^9I9%ocjf_a5`UT2&Sgi`q)C z4x_bU``G1bAf>X5hmPZV2N;Z8k1!v~;_ntpRbdh=e-T>1T#%q~Qu1>=g}viO-FcIc z99E)xexT>5;b@K*bLLvSJ?)pQ$o=^QXk7aug^QAxjrNN4`>L%dV{poVW4GJps)2xn zYs@7jumykif$X&)wKwTixy}v5P&j`H??5OG7C|QTO<=Y<<9iSpmYJGa=oG6L?(5Aj zH`{`w6))h^Vq22hNf8z3I2XuNlxMn&?b_#^P{MnY1UWaPitT^AYq{*c>?~UkIgbHJ zm&ty4YNEih4x+;kYV{Tk zF`6;CRs*1NaP#nDatmMWqkVQl`7?zblL%K(k0(HBbAL@6U{Zp$`(U&;ivmp^LcOEf=frYGAQSf%$Y{rls05S72}VU#;*eb<(; zY?iYb8Z9Otp*`j~GDJImbr|WZm@kR)h%fgzE@%pRtN+ngX>W`<7>(CSZ#WDksFlsW z+m%$M$~e3@xymBpYLE1|ote0X9t>eb*je)th?#tA^!8Qr^YwfowS8ud#d9Q4dUQM8 zz*z-1`BeFbtvA0+w&y%jR}np~<|+w0+5{-E&r%J~Sj8LJLi!+llrZfL#Z5kZ!*yX zB#_Q&Di88UA({-z)yU7QDWa>~y zw(d`(#!}=v2Yhj$`g#Ik=;Yj_@unE2ja0&I+keyjUDg}_1{;^5kD;Me^|5`}2DSr6 z;CpF*3<8%KN~3iZx+8D@7%eHrZ&cHX?Q$d{QTx!3V&zyN#1Wx@3a$fg%F>9KI3n{a z+qyG43N+IY>*jQ%7jj1nRkoJb^Oj#=lF5WL;j9$Sql|K^$KoP#0z;%XN<52nQ%O{z z;ERvwFCuMWWk+0<11ZNS^PVDLceSVz!l%>mP`shK4Ois}l!$;U|3 zriGD$Qb?7Y+;w3}*y0!QtdUWTkd+ZEkQ+j0p4&E>{dgBJ2caq7m)cOu(-=?YHvywB zSxIbfel1t}89&gp`HMQuV;jme5n_VgS9nrR#EpV7QxWmoeT!!zE`H<`i!c(j_(_MD zF9QeQ@&0{dnG-5OM`&Jy%Hj(Cu*`cmfZ=`Qy}qic>>+UX&$n}}75;r}V_)bFf%0ku zJOjcD+T6fy+Aew7eP)q@yzm-9xu4bTt^a(_eq6{yC(;V~$?(oXi(vZ4-Um40IcNUU zKTr#xkqL=O^KPn>e?40BZI~L0Xk}=;D^suoaZf!^;w???Tq@3R3F|s-*H@wJ>{|C1 zU6+1_-2a-U?)_VPNM2$a&tbBT=^LCRA8JHHV{XQhDN3>nt`x(k#g9@*KQx1_5$>Em zLb{oI;nQB;2-58t4r`;?{vw8S#hrI;uXZliQ6$13u;M4cbNSzWSeYYRW}(p=0QcbI z_j{!!K$_XtZC#*z$ZaUp0rAEG7LR%J$l^=XKBU!|0(W7j13+>n40vTsaARNj!CN55 zX7%;uMF${~YZ34eFOCXE{2vRnCsiH~*GCGwOLGx_o5XYqo51#260;$08>{Q#7R(0* z<3pSp&pqcq3APJ6k3iC{KR+Y8SF*4#72cr{;`Gqn&UV)PvVr?}%GcK>b&C_jyNCre z{7YA$tqOh#XbR`Vx4Mz)))Vv`8e|(o&s1-Ks~}~z8|||xGLfEi17N^o-2(Fuxbaq# z;m_gx0LE5U21D2bTB$j+v~mJXf<;dtGqctUcHRrIore8R9WEaN56rcxtA4=xY_5S( zkM9OyWp3`PJlh9l;a-i=7r*2)=l>+1JqLNn6&jVne*e9}prV~_VKFCt&~^rRP?Zo7 zKAl&G)e)}ta`c(w&`L)(q{%-8&WYvy{k=`^ut4dV)I8Yt6`UoD-zq&D+&!g)?$;4;7N1x~Ea zIkmOi%B3#PH>8~p2Q9}J@oF~D*>eV{UT59bUv|Z1MyURmy?*ER@a4nc3L z8q;tG2O%4yymtT3#PD->-$Eoa3=o5J0O8GYq=98r1kLRB1(&~Ui}jfD9K+&Vyubr3 zDUNUO9t`I~hn$KB1TC`FQ3xVg$sTxUdxqDqrG60Z1(JT*e+`gKYYGZKk}=mwd}aRs@gg55Y<{ zkZOAH3r*5Zsjbp>6?1Z@49f&!Rk6z>A-5vA!juoWe82uRQ!~r@QW-mGSI$Kd6?DxW=buCpWH4bT@6<9yMg@&r-d;PlXThTs*I-kd$7bZX<0xf`T7!|ir#xoq(iYI^>z-lQnp}Ur_ zF=ZkB2$YR|=O;u5kb%=OWVzS2%b{CBa>tizpfX-5?Qj zjf5z76n~xgmNf_&I5Mu2ly#dvoqkN|<;TR($z;D~+x>w_HjC4g8T<&$j3V2pqM(+YS#3f67M z@In*G8pJYYwfU(lQDkv5H6zfiapH&E(eUoHmLA;JL*zCpHccbDZ9e+5@XsxhPUL}h zgk&Gi-3e|Z(axGVDBWL@*lHmfa&~H8*?+Bvm4IPBlqx~-xa2(v=ZQ8nx2I@0tELpV zSlA`i;y%MVtPRJ+NOK-*iWTNtt7nts&6@)fLS$Z*)6n%t08-%QzXd&48GRc0HAx;} zE5n*1yb+`nCE{>H2o0-q!M#q0v~^hAx^wPeaz!U1fC(`iwQaU8AuRX{g7Lwy=?Qw_=C`jPf(( z&V}txeIudl@^Y)ZcJ63lwkKS_dHoXXT9Pq9r@Xkd$edXLrp#JIG9OLw@lGDsm|swC zwQL(bc-Gu>G=AjoEa-c(f$7h)0FWpI=J5*3*YOT6*V;YV6~ITQ;WwNL798LEq522& z8kP-xn^^zVxoHysz0_O8nu$sFNkA^1!Xi#)sv1R%&|j*RQ(Bp>$`=*l8u9Aj4!SdB zqwB3NWbaSK_v2KiN7U5*m8$+#;Qseqf&ac#HOp;wAXNuK^g?Exg@rA4kOU~Ywd7m4fuu8$fDB! z?CpiRrN-~U(ROU$@($i)ih#HLZQ@9e~Gvic&kONf2Y2{7k?F|b8KV+J-Ch&oxn zTMVG)R3T=`f+nf@)dlbsfHfhf_wC;{xjduFGM8|<25E#G0DDs{dUFkJfh82#_KG!1 zZxT*b2;JIdtOcVw;KpQyM$GPW>I5nJh?codVeh_Mln)f6PbVh{l0c*s4F3)Esd*~k z!awIgGE?f(*}$EM_&t)vOEVDT+gc9pr&@%xr&?y{+9#8Sx%M5MGi@+S~N`@)TsQ=v4?ZjRnNFYx})0+y=Hb zVxG)%>UEvmYM7dvE_ltp#ET2BtpW6=frc;e|5*^STOBdW$|?GMa&ny~ui(aqK=(PZ#Y&(@APJLp)|Hah+9;og(3fcz&F;D=ZP$JZ-6XkeU@4Pl-uVOLG7r6^U`Os0 zNk>uFBYqGjeFBc-Q84|I0=&Z#ZUxnueQ9BHPc&e z!DC$E>Y(f=Hu8BA3EyBQ9=oe}mM$cv4#y{+B>}}J=n)a77BRX3+Uo`JM$^=|#QwVF zI)d<~?A=fcsjAwzyNw2fNp-*^(U#_-cm=NUxd3O$2>~Q)> zJ8iuc!OH{?RX)z>rOC(Um5#(vg+%qOH7KnbJSB}E(3og_L&RrF)auhNN2|r;O3r~7 z`vXYLF+8jy_Sc-IyWY5j=b3FhU=QYC8AT>|L?{+uio*nyPu+F!Dc={bJ}yQ}+yZhS z&BZbnBTePTcQGAhTf4rt}*nW5BGfBDdou~Vt4 z+ZupKzPE>k3Fb<#$rV5zkF3gBOVJ!3&9CQugtN6@EgyQSq*x#8c+|_sb$;j*%(kpQ zcWe}Os6A+5I$93RsKVQ2Se78%1047nqFCI2&EN=+T7G9BCOk9cpyr4Kbb-zBVqGwc z39O#O+pP>WfJ&b>u))?3MKkw+uwMc+3A@*(i+aQ_O(fV>32q+QwkuJ8hpi|TnNKlO)t3$lB@n6tl80M zyDs}2%F!H{Q0k7{KcuEt3f2wlvR}GL2{aoBR~JeIwtoTaCl4({B$qx?1?j!MaWQ$> zs#Uxa7=gcak$5QR$!KcMVGN_~)Dq_qLc51b{lEd0RR?CVmoSnU+GnML5z!(>)iR0K z{mxVM#~qhc^n`xf)CZp*LFk4iyrsO32QLxNOuV9cFJ)eAu~(X+8NJ{u)4CNQLs$Bi zs~GAd`RrC}d7K`BG3GP~=+bYR3mo-+Z9~5Xbx;Q+)*5ta4_`U)L`rE1vAOef5>B>S z?fpwslv&sbD#xUy9p!fG^L{YVd1ftQ`GiD(xhM{q3J|XMp8AW9PfP{My7m1ba*Ej3 zGOJFyP-%n*=|!CXtQH`dCfX;QTjk=PjwR`b^m0L9xxH2!P3->gl{t&t`hlf3aoikk zzKsYh-NKEe{=k^4Ah5_?E+CHYw!!{B@V*(Fs~`( z{&RJHKFrNhc(nRyW((qpZ*qzT4$!dJQw>Sir~UwFShg6Ev!s`hLo_V@>A?R>m-xS> zKZK&z4c9^23g#*LezXWk;6-gMo&R!XhuhkcbUz$}nH>`yAn}m-#pga7cqU~ZFrXY= zHw4HVdVcs1dzs@GPw)%P`LhuYx91vmC-)jUd^mvEOY4AOQtGY*l0S4q?Fkyn_Kn)d z2X&wDX`VhuvONSTcOLLX5Demw<0LFWO$h!kY-NlfWl(1aMPI5@K}91C!-?q=0dtUu zPj;t5#dH9Lpf2&8|BbnU4-SAxY5)i`gCgsC2%J`PARWlTJWXt&PdIUwyh7)9UXmjI z*WgVtk}FR>cRlJcfMf$rfZ%sj+$R2Fiyrr>@-f4K`PQ;>OjX##X=)8 zUv(26F({Yymqwt!1QJgTP!Yz6nXje1lv5#XJ+&@8Rxh_~B6BL(scQyOY17Pn1U3m| zc3pc6_bU^r4StFE%DQfFtJPXR|Aq(P6}j#*(Z=b4D#@j8^=<-? zXsvn6c<0-{lIPC|%;rNVK33Vr`X@(Z=Rb-0w2n1rz&91=cO3a5HYreex-wyHbu!Q5 z8#wk@_e$cfJZ2W=?JC44bmZy-Ty|jiQxeE})`!jUDlYRJgt?ki^Xn<{rGHaD$d~+i zOv4{)Kr2ftsHHi4SBp%49wqUOm8Y!mVKYcu*wy*64w`>qtPP%&dqH+C361LiS^WTMJBW22F*eVnV#6ZyCgT$01`R7nnpn0M9VDgfVqrPLPspG3?)5L~m#Z4U--)po#GG zk?2R|05YhQ>n8!lq`X}YLK4;=?;QF-29Yq=QN~i7>fW`7cIZ1!jMk zvNts~RvJ-N>RmR({CmstebP;qj!Y57#q<>0VX}wERmMRUyv5ZP!~KhZx0>!a8=GJ! zrnf;oxEfpmXZW~60x9@5zMvF4Q8~QYNZL>oQX=ya(6Q$YNrEbboc#f4O2^Ux?~)Q= z{JsXvy9e+$+onzmr)Nr&_x7?G#R!q+Wgmn~n1M;j z-HfLXTj32+y$8VVv|j!$QvLTq664$q{Im1($ax%S!q>qV!59nPcqJnkgJ401kEHZh zfg|@Y8K`;KNnrJu+V}n*HFwV0|8P4M<*JbwtCt|%yaqu1=Oq0ck{$d|aJ)<>O9!I- zws!h#t`9;ZvtF*z*+x7|AC_a;_sL0E7`;mZTwEfgvmZ4pbAyAwPk$^t52z%D6^UX< zJyK7P7qZcFn)=QURgIa!!2aA6^Z}iObK6+x@a0A0<;A~-KQihO85SWzR7}7=>;SF* z#nmrg;V=t2>Gg3{B~(^k#r)@$jb$`CL$za$xau_D*ec}pZY{=AF8I(v$c3N!u)Q}P z@#H6shDa1!b@0{(v&7~g`Yn3oY}T4Ra^y?TKV~)4qKX?1TF@}ENR}hyAoa+#T zG>pisIGl@O{dtC~-4I;Q1}>duh2;h*Cg9uAcYASmzi zo&5Hi$hzlubQdzn2L1RRJ63PRwkpr!tT9zG7|ai)CQ8c7JsY%qxge$kY(4}Bm<2KS z&kLe^*(5p|XT1h9vUPBX7@I#8b>N3b40V#rJ+y!rttEi@c9?y;t~ z{4U4W^N^eq?U%X(?rrS!_2TPmZ|_{@kmQp1g>3&h%pL?D>I;yVVo$6VchyJma0iTn zGDq-PYNYH-omk%taXTS(yk;av`zo7OCGt8-EfOj_`aC4DZxn_ zOsU{Y@+hwrAljYZ! z0lJS@ygnSz9$N%uAQEAePSM&7*9I}00hQ%`c4V1}VNzijv8lk}ZXtbl_x-T%EV7B6 z>M2#vvxQlO74wY~|ZNd@P zf?tISTD3=ARKO=pah|##3`$Bs99??L59Uj~H@DP)uZ>5q z1TYxU)H$3-4WIdaRi4N&l5@Y5VH2^hX5k5nu;mEbC1)OV*j4N39StS9{M)(B>8&rQ zX2Foh{i3c`bPt=I}7%KvA!js_St+FuQq)KR=Wz@I2|Om7yp- zSG{u!qo`L99j98&DY^AMJdU&dZ_|g5+9-^ao_8dBzmjt`>7Ia9`{hyE+Ma$0)Oz4S z)2I=<$^@OtX$_MG@nDTn*GV{}ni(qdLIh@8%VRxN%4hEhO`NrH)t%`q@08R06 zGfGC@0XQY;5n~57S>wsJBc$*7F!^4}?9;xoqlBYqJBb}UgCDN#ku=J047l0@heH?$ zS$|I)J%t)?IgS&UnZL+jcnpsAcv-CwP)+6y8E27jt@w%%gP8Ngv6W-8Gw4n4bI*q$ z(2fttIci72+q=aJb8&UH7Hc;QD`w%Z@*>slH23q`9pfC=L2@`a1Mrpo8I%VxgSEA zeZRx`Oxnkr6F7wLT3Bh>(Ka~SqkAp*)TQ6ILj=vMcVu$88TY9);6YM|z z@!;iuyr2Ca6{?xGlw3`c%Ng2W=kdf3N6t(%ulI!W)At8jzf`Cb-`m0n18&Bv zH-C(dW6)^UH??y}6&ZZI*?ZRwE%$`?;XgL=u_vJr!Dpi^KsUxI?| ze9SC+0)*q#Gdkq#c<`}&O}NrwqRv=%x+Opp3+yngzu$}X@?nJg&vW;r%>gI|Dx*s>>qiQz(#Jh``Xwp?Ewk5 z`LVBUU`D)%q?~vQ21f@N2hWqYg^8GyuL8C`-y3q)e6oA;(w@63o)%_I0G+RNvVlU* zv4P{XoH-12oQan7C^}r(0^WBsn0jcb^762`Ioc4b{oH%i(vYt}x{9BTk-a9JZ|KE+ z)~ApYv^zQ+4*C56O#!>pAQ?v9a_XxeAjd8)p4t3o{xm&yEu4IJ(h_J2p`bi|g6YiR zP3(zHD5Y&v~ps=#Hx8M&Ri3YJLTZfD{L+|pvVCwcAf81G;D_uu!e6G$_5@+42f3zh}HwoyI%z z60RM*(dch>!Rxt@O^g(qxBN3O%pvJ7O+HllApDvnXIUwH$(9ElnBHa)hlOZm4JUdm`@TeS zYr|*i*MEPqSI@|(uZefZ>-`=rzRr|4$@DvhG@^e0j#I~Ew*2q;Rb;%yzoq_MVH!^g zfJ4s~Cp+pPa{Tiiw;YBEiQAA#lpr~@oM+=Y_h)MFmxe`K?>LdY4_Rl|qxR5#e1e@N z5l*f60*PBAupTCAL@pltvzBgWgjYsxby_PDS(@E`FxNNs0Wqh)I3Mt3LdKg-%2Nlp z5_Y*X8LqdAe=Y3Fb69GO@IGz@WY^GZhPE$_>=`6qp37%An=mA5TWzKW1GLK>Csnf1*60`ur3S z#>(?o9hdDP#io(Rl`Hq-rv!W@lWv?N-3ESB`%kl@;jGAXjT}9}^85AQ)*&16A!kG8 zw)>q_eGE6LaeXa7x)EiBwnuSFKrB9mQ9si`g;KQD>l+_#qxWZ@J5L$4P2`Bdsmp&Z z@?W)qMgvyczw36kFzLl0SIeA-9e48Xqkn8`U&=d*zEt*C^n(}w5hVK+9>BxhwSRs3 zpVzwr0kr2<{p-HpV@`%*2gE}h4KMvaueY6xoSV8ImiJRuUKJM{|O*l9z`^ zzbFh;iSt$Xk=W#-VHjyn23AiE?d`>>o1feUkYLjrs`4<c?0;G%;C)J z-67BdB2}F8>dOl^`7gCQoH-OG{j1+Re3)|f=nAl@v;dv^4mi7VTX zu-bkMKt?|thW-hTl5l2s#`Z}*o3qkGCNR|iD&gV9ydm#F`}p|1jd@T3KJ7A*-CamW zIFReMFcJ=K-%DU6cCNfedB5D?B&;!a^`C3p??^El!z6k;xydS8XE8>=^5`!5Qwr86 zBmUUN3QP#R`BE&{^77)Y0>(M!uP?5X!Qg-)%+#{aeZJ3X3i~JDq;5~;&IOP6^3k&O z$hly5ir?Co-rMDpP9B#%c=vfo{jV2FbBto!Ue>m6IqGS_)gIg*?}!|GAbi7VXFhMX zI1+1=(Z6~|XYuPp2R0QR77`sZeG4)*+SH z*w&yO4{9r}tYM+BKeO%}a`};DHlERtdqEuyKL1!$G&QfuSD1-;b@Suuc(ib@8+}*# z;$kBIx;*$Abtb-#5ZT;{twQZM%|*jzCi__<^}HbN%+1dnXW`*`JnwPDw>N^`8L=FnytDX-@054_bXdQ0AtP{MGIOVWtkli( zQxKFebe{HQ8G1UxMORZ<1JpB&SHrWT;CLDTei@8|(iPDDvE)bhz8|DGBz$Xy&$)UX7)9yTFvX9oX$Aq<>Mv-Ff@wbYFrB_;_ zu5Ep=%c-3{M{Xwf^7AQ)u`*}#sB`Tx!R{gJBy)ysYoCSayv<5~!mfe5K~Xw|`dwIAeJTR$Z;dY05i99Vh-E3n)t`&DAR8g-p^HZV_@ zMAvpFh~77HH?X!1os&Ht_V!6-xP`xR>5H;GH`4%s%4?g_gM;)UUoDUoPfx6Zpy_RjSRJYuXq^>PWEbJ?_(Xc;~mOYzI|nWpW@3K1qQmhUmXs>L7f76u$P{S3x^P0R$GIQMCTWf=r=1J!63HPR?Q!N=KU26 zXe8Cx&v_h!8f^}chM^CTipT`5NA&QAL-j8%W4Ny9e}4VTC;uTK$en+p{(9wl41(!z z0G{6gxN1dj;1Jl|g1m_p< z`(-cVn+sY8;*F(Br%AJP+C_1=loumDP9}IfzezJo{I6d#PMXD4;XUUi zAY*bKR0#Rii<*TyBLR$_w_rWT_oqar)yt)Iv(Q)f7xAlzfz_n9+t&yBk^UODdnr(g z>v(>Pf93bzbp{PVp7D8qXtju>c!U0@mH&HQM_LaCY=s$!$JPS#Ay@KbD%7v>iPl{n z!&ezX3gr^d)|cOe!j2Awmm2B3hx^;7FUxlMyu##vLhL^Ge`lGzjtRBY`UjS+)O;7` zOEgxv(65$I0_)+|*1qbipbX{(DRdIdz7NkRj|iBQP*R){dGtwi`Emgayv>IqTF|Ia zN$_?4^FvDa5*a#R5j3nD2L1Div0<;Gs8~a`0_~sv&v5gfE)xFTNs0{YAE_@bGMa(2#R4ZhZY`_a}|hcoxP(VYQaXe}fu1kD*uB?Vl_E=inNQ@|T5~*6CRY&A$;b zrE@TD736eU{_~q*5$%Q{Bh@v$OaCu$mkJCTI?73H|L0==`}?nv!XIqBPptpopGR1U zEfHpryvMw;|C*pVrx@W627IP2Cj4ik5hu+JlZN^^Q*1qy zsbC0EsTCLc*B=aaBJ9QbrKa~oxZ#gVIPDbvYo2u>oQ?e}EfqP;(q}FkvVj=?@pEt$x84A=nGmmgr}On|$2@es-2e6X`~K!&h5kdD-W#wEys-d{WO240ha$i;0$>}5av zTAp+Wc<1%lgP$cF)&H;=nFL`%Q+Xun9YOYN40_2CEV2?|5(~Pw{1Lb5bV5|8 zpa)GdFb~j!9Y$i~t00}`1x(}#C`dj6eY2p9Y&ih6qJLH={^BGtCODc=cuCUa<;|3v zwQuQsLB41W7JmB57B+9tH_jd%tZM9@z4{LAe}uaqO?TzS5EXgb-X_8K1`$jRwDL=} zLj^jDkN|Ne@fX~Hb0DQVS#!gUn}%re>G>^!k!4aTl>`oW^**5A5k!DwBpTFlJC<4O zJiw%JAPuWqex(Vm^&u5C$;G2{ul#l#eZVB*ZhQzy!P>w!i@+b71&7MuP9ZV^l>CO@ zu4qR_X3L9-kMQk*H^r@w_dFh8BM^SOl63US)CWx!N%~LJ7;{_Q_dcM4OaqR~`cBq_ zfEYZQH=96lh%S}5@l)U75kwyz?1+IU;^V-}m#fe#3~AxVWs^Lrem_#_u*`D7@ir^< zR~TqFVMWzDE%Jx86A&%u22|!!J;*s7e>NT;`F)3dR?gc3CO4YOs=pB{MsrsmJzT%{P@27Hlnv-5K>s{%WWws zn+T-kPsdY(jGdtT@pytx%oGpW4Q~bqL7L$w9>m&v6*4obX^<9SOfrLHs(Q|?4r_I$ zW*nzQLP{}sFfHu{&NK>KX(YpNh3+op{62y>fdtp*W1{Ww?a$4$#U_CQNh1&pwj_gG z(Ep#9W4GF!;ZnS6KSO>@39ncZ&)f{UxpT`yWnT0bISlKf6-1x1 zez~2aK_6;V*a0C?xn*8INX{(D%Y@BLJ!@-=#W7HyDI{~kzd1uli}B1lycuM`Cob1O z*jM_=a0h&B9l(m>W!q6y%kvq%mRplzx4|*+#C*$=lI8}W!P%vaa0f8v%!6Mb-+ZHp zJJJA8aAXrZI~WvFRS14D0Rp0og0x@T@^ui>sH5?Tze67or6(?2nV%S!9=^<{9fpA* zHu+9UW9T&;`_TY#9ZMJaUxcV3&wJfaTU--@eKlPn+I@xJ&`s5oi z0&p()-xism(jXnb`j9Rg4WOfIXMU1s%(I#2GG5lyBDR6~#tIG89k0eUq@GNF|8n)W zX!nYfR4mm8Mr{37bZDRh-I)7lwHV*^l!8T2DtzBs8tiN^Lz+Z_iPJI(P|a#<$>;^% zYvDgA!r3pI#)^HR=>y2qwh)G2`?um07~tPaz=# zgEi{S8vEt0NxC48_p&n56QJ!WH8*pa1s^Zu?`4sV-R%ygZoGZ7d1|dF>Kcgz&D8mG ztm0PtdQ(SsBZ}RA7{hMBTSRP2t43cqhhl!+;NZe{UI>*OYr2i1M>~Qlp|pr^7vYlZ zd$ve-!~|&#lMo5?RfnBUVut2#a;|9o&D!xweHLt&)Kj0OyBbJ${>nH3W)Af<%Wmo! zWY?8C+CC9}_p3WqxNg*0V=eouFRP+ovg9!6`|{N`!6Id~JIVAjNJo>zf^l_5i!q7u zwH({^zRLqN>hOiuE?`$nBZSL-=+te(um@9DotYmjkHLVyO|X(p`26AP4vV#i0?vD~ zs2SaeN0HAXE${6lN4TCscluUDOh5?M6M%~zG0Y9MxB~mg22zwgM3i)pIz+hMp z;cvVh4UUEMh(N!Q`*}n-Try8+Lx-O)GOoM=#w-bxEHdCYEX-5ey`uD3;F|~kRXI1| z{%RM?lDs|UpN!}}=(R62)Bq5QBix~>m7SR>h5LqyVU63JTl(6$DbdIRR%f(9Scr*R zw_so7zN-Wdhs@Ln6E|UC?tI*mo{c#QV1=UW&^kJ9tfxe(tzbg!>XvyfIn< z=i$$9rS(iG$}AANTmqJ@Fh7Z1YiE5EAD&g9V=UX3pFziap@x26V4!I>K&@p!MP=;- z%ReWORvo&+T_1IvUcBiG7P;F|!mpF^+zio{N4mNUcxdxl?r$Zp63v{i%Lb<(4#oM``OT857Q|^71MwJ3@gs$0p0rURy zm!rUmyJk3>Bk`^IEiPdr33F;j%0mfp22m0#e>U&T zMaA+%$(Un<@;lcux=Wjmd;>nU3LgzkkF<4SRP~-LWIT%AS3g02DA^{3kC!Iw!o-Y>O4^R5|bFx0RttmD;8X|md z`2Zj-C+{`H(Xty&o4RhcHRpK{mOlCA!23GEQ4c=)vWU9d1u7*sobrC6{Lbz)w}iww zcj-L#Fj1KAYO@vau6A7#5a(yro>TW2;IcgLF_=0^g1)Cq{pF^$tKCBZ)L5nb6ooa( zcYQtWKr_MqY!CTo9A#M;bEv7j94L{iR8K^j&pYcCPpR2f3~05+XJgC>uv_jBX+KB_ zEMw1U<6cU2fXTAg?T%UlB}z?AC&IAFT|M-HU-IpYdL8y9*2exE(kMY;aevmP-}=0x zx6I$$@@H-2iE?+;b;|gxB&)4>0r$P_X+oa7v*K00D9IukD>2c)T`}%FDI9=|++Mi% z@T|%BS7$OyKF$?WSWpxfYtBgC*T6cah@|Cv@mNE1`fM8?%i}?RMPaYPJ%Y$pkTi&N z2!7F8c}>QCt(9u_^0fo@ot*R>Xh9vv2ijky7d{=RWqW;P1gl2+9S2jJXgmXt^C&C% zdx+$J$;(2GTMgcgTTXLlo}b0i=S-tKCv0Y}|ANFZNO>0fTK}G_ZyLj4_u36ybl=-- z#=IzOjI=-DJT-U3O7U2L$H7B_Kw_SIPxXBH6@6|lY(x+<)GarRVf>|W;@K%2cpPIIrc&Ze zmK9>m@_5{CqtZ1J3ehV}%3gBKf_;ta(S5p@h!mmdidge61$&E*5?4BoK`mxgV+)6| zKE{=-L63L%GV29lFS%Jeny+2fSsSCGnHk2-I`P+^O2SETXL!v$L)n|K@u&FuC{bY*lcK&i6L> zdF`aLvxKct5YZN{HXTc2Z9 z6AkqTnOYLQkO5ITY@q|jP`ai=Uhah`V<*hQ?X8jJcM>ED-uQI1t_xA(U%B6wP7cNx zNp1NE#%jBvujI_15?k9Re2q;~b@#JWgFi>NL3RG#3YlxsTaI?7cTRji-!3yP3u}LQ z$RN>6G<0P|N>_Kw$-sPp%n62Ny@3}6pMFgC`&5$Tt+v&5GPW_~=U-)z_QWhMx77N`CE zXuc z>W@i2S&NzAJR|G)>ZtZYf1 z%}vrd^^ooppU1#m8iRfy60@L2nuTqhsVx451D#BdUlXpVvAZ&OX#bE;hSiYTo!vT@N8>v z&zbt*)#3^L&i(D>Tu{4Yu}2e_Fs=?xf~-=gcpXPRy25Aj%G9~gQem^O`~s{IemR|d zkwh2sJHsPwp&7!fGaqX*QJEzP+4ozszZQD+uFlXK$2XB+leN)NA-t24C^jaaPHPj> zmGx$l4y_e0Brk7T87Tv4r4`@lWVpg2 zBj0g}wobibLoHCMUf^!fdheEfw9CnD@zRz0d@bYOpe3hL@Xk64U4ES?4&)~bI1Zp# znSRb0j3ojNw6J{f><#EnK7u4FB_U5^*MvJd1j8aBI(>NOrzX(oDG+N?%s%!9qlx3h zsgJ!t#7QoT!x>kJx{mX_Qcd;4Z@DQD;SR3!aITQ9gMckz?Nne(Z?0wj0bo%H(z^lvU~%*HRmu5*0`w^nl#jA`_Lpu| z*znWbEkW=rHezi2mPnkiB4svuRNTQC-goVpL>FO9U)4E#PS%Mf^yG^-r=#G_XGfTX zni@zVpelc?|d1Y`y_ulVU3$;W%@H z{O~a(Wk=0icifW<*stykG-+m~t_w-kU^GvHp^jI_s1$j;Xf}XWncnMLrnyBo z1I=LtcZ88>5TIDIXabh@{zj2Er$gK!TSX&%1-6T?+`rwGy1k;p`}Y$I!&*UN*Xrzk zn}Om%R4SkvM=AS(|0>96h7?wT*0+Y#=30B8`rbuto(Ne`G4L}q3({YDo(k?IZcdnI zcXS_uq(BK4M0hk80Ti}aqYKiy^$I6P8c|3*%LSpu{+5!ZRSuwht}vpdr2ox96VjpK zZmVAW3B1YqwI7c-cclhlE;~d@7%EIa;ep?;A1C>q6{k+b5YicNddPI$%3s&0VUbhp z<`i8vtG}LQ3qw7X|KEDI7S1);jJXt9Wz)xAxXxHQjPMDB5rZK0-UDFD3WhLkfL+$^ z{L#0g*&!_#ZOg0w^nU?ustd0`&fTB`!j;n_HGpzB?!lq1>ohre;2FpRA9!lnpv*=< z^c(Wx1YmOmQTM$cb6a0mL1X#^bWR#EtWqo1v`_y){7TqpNW0NtrE&8=O_?L}_6ZB& zJf#joHLxuJW=_z6XB0%$j*Sy>vk}VrI7Op4(s&GHXJSmw;7_b zG@#%w&bJV)xFI`l>&O37Rpe2KMjWAPQKh}GjUo=f0Ry-cDYxyF9_~Op*#QFCf=-iw z?ZIz5H}SO+r}2?@gAn+Z>nMC$Wi}RUaSCwEA-6MCnGIVNNA34&mzi1cNXHrm&4#_d zpjBX8Rm%48$xmn$E-c!N0DV|VjK|*orN}^|m)xF;al7(OEP;$uj!-S=Qnm^CSJuJG zTp6?C4=!8cv24%+S*5Kq^WC6qnug=_Wii&d`>o$SkiVkq{HJRhELw0(R{^O~Y6-!U zSmwXq&&j;Uuu}r~P~8DuW{JZf+;S^G*NXDXgYq{?0y$_{1 zD(Bqz#C89IMn4)X7<}Gz`v4UnGOP_qcZo_ez6UBJy!4F762~K;c_^50fwhH${& zgAoUG$|y20GQ5P1p?UGgJZn+f72svu~l0BvIHd(;L?RqLPUolX9j-=c=|EL)e9>d-VGXmA}i+K1r zK4)z(@z5q4fNh-@>d|2Ks98wMbJvcGcXyioW9hH9rp>(26FzR{V${|;IohLzUP@j= zQQ!{$!on@HrA;)1*945fOSc}ykuVQHqN6G=&hLYg{=?XdQAUS>)O8itx*@#d-ym6> z3>$#TX5!p}ha;?Zl z%3m(V+c5*HpF!m^AOupew&zMu&FIcGZFF2p@Th*m)TYp|IoEk1TQOGMxBf98@-K6u zy)}NCg~btKP!|z_*eBPG|-p5_ZsKdr1|#QFXLHM zVz&bhVR)=A1z}3;?+B=k&W;8b+des|J}{YCi6O{B7Mr$-eE&ieJyBV+H=1A7H~dA? zF?Awyhq^0K-XFKV_u;9Mc!-viA>fsxzD*Y#@f*Wt2#nD+dH=)cV%$s@?u_s&!N@w< z2U*ewxB6an3>ON|BtGy!9c|qNp7@^bWhESTh`dgKE#iB%iu%Zk49dZ$}47%cmhO+eFey!A#KY0z8d@_$D8TL?GSD4 z!Tm_n8tghb<1cRv$=`ygC9j;pMl%9o5ey-((QD4m?XAbc!kq*feKFqC=)DP~k)i94uW z&^07N%~pXLipP;32kk{Wq%xy+9=cP=)rVc6m5iV!O*`w1iy+x=7~Mq|=u1PRee+UN zgS=BWyv`8ihiel~AA5B@e@EErzYCVeFKC20fa9iky{SpfBN4#wfuA5$G_{^BL&fh&e z&55-m`S*8W&j}@X_{0C?yNG)J$UoL<-c7q=8CJIRbg6^Oz^g=z<+1L7@;xYLWGp0~ zjdV4c2o$7P$VT7TeMh`CWGpZA_LJT9N6Sy1zvycRUvj4Fy$!LSASdqK8>Wv~LU^!; zs(2^#Y-Wf`${>2*tSPV|yFR&sWty*CmegWy{^wJzzI5E6vMRk*pn=3duuJi{(7kd(`-tu287A)PRr(ZgWgweCLi}VY zKIk`0&jVq#Q>S)jW!^~$w6mHNb$E_POSzS78Tz6%dyEo$%nzl1{kXRR$_NWw>_p-l zO&9fApOY@xGDrJ8sr9IJdUV>9h2KJY%6oX}e?FNcY*E{4_yI@!qq}*eAWkvKMS=;b z>f-|cg_K|02gEdMoeklL#Xcr-Iisjpn_zJ}lcV|TD`->m5Zx%FLIc=;E;JR0;ct&++TUtZWXIgG607sQhpkt-dgt#a zbOiUtmPeR))uVQxrp9!YJ4&v_o4PezR*->_q;RdYm= zu0Mdn;{3wfIC{5e!sZf+L4K_jSWLdsJly&^ie1tc-sw4hmv3@1(sJHJn*FIT18g~( z+bZHHpY6?2K7IH}2Ct#t(Q^0u4a8M4tT(&woG%aDAucN7D*P!aB?L(#-b zr_uR&*5sJQ?8w$Y()sMj*W93ks8P&-vajqSID{HVnRq7KDJ~8IVCUgSo!fBdhjo1ZV z^G1`k1E9FSP{v2MJgAA|X2!5>9E1HkpQ9yI1MVB##Y0PlhD_R*6%Q1A!pzENVdNN0 z4TTS@GFn{a;CNKN&tLU{jp8i-83RP8E{%_zPEx##`8W{`@%0DUBoe~U@@q@(-qPoF z>j6asRjpk{(jaa|w#i3uew~B?etqZiiWfcAVnqO5CFs9$&5?|=k9K%2w?FZ8P`}bfdMmLsbhSr#K z;8C05NN3SELy)H@w{CvM&c$05`G3JEEWHGki`ybqCqo*0+|!nNOv_Bwaa3L)_;`1Q zJP%A2j>ijMN>YtqoLW?4BOlNa#)ndmJWe|7YCIvEUTL8CbF1=$iu*HV37Bu@g3w}Q zo0hj_7q$r+4TNU-Jpd<-4y0^+J0k0cjnt~adOhO|{Uj^E1o%QvF;B8QOU7*|Jd z97G=0ZZ;#fdZT*p^89dxzw9y^9R^0BUn!c%q@9I^dV4$SBH4WCUMWm#PmjPFXviRF zOI@qVmNb220&$~SB6mCv;RVg-5a$TP)1E-oeglM6Y$JkWO&5T*UlO_Vo^TPw4F_UM zrz)2uyw@y*of-R?=!Ju}U3EYilbeUlrlij)d<=7|0G@W7sE_B;vx#34Z$NL(hAuw6 z0DaBD-D=>od!XUSW+5qk0s&d><2lm|#Uc4}JgY;bfLD4<;lDg8yUrb3zPVIXChANX zb)HVXZmVC_R2@m*s>o_Ib1Iq;`sL#O-hI#ybRs><7t&_IxZKsEbbhW0rAwih2WdPj z0mM61-mKM;tWtsEy@c>6goKZ!y?RZG#Z5EZC?z$Z=1!U8fUnR4aE=_-$FO}-Trrx? zr|<_*DL*6)qnX9Ne3lI4eo)gh-?9e6pdDSo>i7aJjs{lPYaW%EmlmMiuF&9cf?%2{ zr7Z7&I;q>)13;yAv@uts19knZb~al5g}9H2LYUry@K|~xfyPv{`R7*XHxg2t?%M_W zU67`=@fga6Gr_4o1pv-#Xm)>!=FKa81gPOMxW0x6Zsb_a4xW15{Bv0D_%~D^yp>=x zdHSvp!Ysgg)3`YLmmM3q!ne5c-5_GX8VJ%C%Q+!I$^a3$?*X}9{X-mp=~E-fAs0{E zhD3l*xxlyY+cI3ei&3|<;w^!Q7FAe7aZYQKX}ikEX;j#C7Kykm9TgtCbEV;3b)Y$n zQd#xp{bW(mzN6pf(>>D!38?QdBLEkQho*e7;y#+jOuA|1eDtH)vtvJ1tZgE zi-~4VTZ;)(R~2!VMc^g5Xq-+k>7>RHt3A8c2SEV%1ZtRx-m{c7kP|=lt#`Cu+}U8? zzXh2ADu1oJ4;&!cpt1@Cc?iY+;rPlQdog{jT+M!fah=PtdK!Q9*|U9LHjruJa-IP& zW|t2eg@V{x#aWn{wc^>XQ=vF0y*rN}j!_8!;jLP>!{I8Y^itjJO)QmU>g|The^rC#Te`3hYmYe+dT!*H} zM;7?pE08JR`iMAlnM3u8BrJf)ny-+R4`2>A@2qB>NubHZeokXM77j#*6C&I!kvC+e zT04uK6Qx?$JU2Gv{ED5cv>Jz+gA8y%7QtTj5x^1B>845&4+sgj8? zLzR*5dAaG*enX8P;Al}W7;U-|K#=`SY41T2E8V#rc7QEu{X@7xh|kCAozXIS8(uyh$1De;B;uZG`yOEiR#<<5|j2>G&|1BIf0YPZIyQGZhCG9(_3g zoZ*v9&Q!)D*KyD6hmA0jx)hboH;UTkj9TYnrmGKjOM;1a=M7XhStlgL(zXG0M-f_C%IDafxpRibrV~dgiQ<4jq2;slSX^H+xyW5W#BAq*P2jX z2~d%E=VulOGP`fg6IG)Q1AY)9-h+rMY69LJgg7xK(5P~QvO$Ggodtt}NjLIt@o@^Q z*FBpMQVx2<;Q92ZNa>XO=1paWf8C8nt`CE77}rIVNOQ&rm<7pix48xpf=hI6DyrnHn z+c}b7El%1-#%+iJsB6^Mqqbo+aS$=U%L#qkY3!z#dCg?NB8DtzKQ#Y1XPXCAmGPk|>Y{)*TnhM1PO_V681>n_ceAOTmR4>NtW_;y+6!;f{$VhQ_3R$-Bf|#s}4GwjwW8W1b4VU|H>$a<-@tz^FnO_ zMaASe&6n^{7yyXdt4ua)sj5D*j=;~c%k z0n3-Obph-Bac04(o@W(86ii@CvuJLjEB&A-BEispd6T~6D!ZSh+#KAeWEG{EiGM!w zh{X6-7;R`gWX&YYAChV?%h@@NRl9_3FaE|w$6?s6C-dB?o3I^us(x&(>ZkMRJwgyY zr{7`2UM6hDLcH+xkT%y}eHp~*mVMJgfC#H0 z&2|z1xB7X}%^TogzEFPD!p7%}WX*^hGBqo_i#LS=2g<@{SSy|Kk<@)VwmHAG;>Wg? z48hbihZN{R8QbJFZ&I^m`U>&ph&C*(mV*icRSKQYJ`Sn9u0qQV)}3dxbQo{AIH#Zl zC_$fcQ>i)xnsX{NbWC*|wcgFzly36!#!moK(S#o794pZ{S4*28kHM;wbiN$KBc59E_lhRC&!zv;mJR+3=ErWgsUptX-*4l{0Tdt15T$nxJ;oU z=mJ8o{$iu6pAKf-TR{C~iX@l>jDU9fQHt0bP^lzmkC`nIdAa!i<^pDe_Nuuv1TKp% zU0r$Ct*vM z1rs^S21R=HDkc#6+E~$%9dMP~phkO#|H|Mn&G-u|x-Kp_8pR(@ycrDJo|zH+s=1r% zX|o=ZTpe;vIq)g;c@dHNB}F##YSRQPTV4CbU!=5`a*A0l4IjuDWV6!IiMTJs)lC$u z<~jY1a0-`#wx|w-;igtl>Tt5LI6{;x!=(e26vO?hc{V^%EUVGh64j{6*Psg)*t5`9 zl3Lkj%Z+p?G)&+B1>$SfY=vkmIME}gNf(;Ij>gsr}b|I+t={}NWV@YveJ)Lq2ybm0^D z!)KuoKLsG)&QR+2S|}Uu4GNUm<-F&vDk_DMT6dZvOMKs9wGv7`M7>rC=4duL*LJAH zMoMJiNS$T<@&-To1tlEXLfEvvz%6{gUv5f$K@YeUPH%6yqgg-){l>d#=cc|idMxXw zRf*^q7Oi#MX@aqRQI`XeGgH@X0GhYd? z93~JK972M1z9i;i+X=r+cnw={K;wj?Q`4v(`12OS16b*S75KzfWUTO%Fs{uH?td?)Z{6e7}Ti zNNWt?A$6v66rN9(D~8ASooL2-jgZzsp{wN1?V^d@ps-*1G8`qgtL_FuhGh%5hlS`w zaR0qRV`Xv8;L5PvFt&o}rL=HhR6hRcDWu=+^@BKg(VvVVGBOl57eE4e8bxx7#=AZ_ z!w7LOqsgbw$x!~}HG|#Z_lIBS>Rs;z9%wvUk-S70CrGzQxF<07;w&C9b-U-$>?E05UOE;OsMlEcb6L zNRV#^UlFG3i*=C-AKe859ZDV%$fq$(bP46Dlk(zJ%4Y<>peBqWW zd7zfrQH6;M16`wAK0)D{v!nDb09bdhgNhuR%%{3 z-heXu1Kh6zE6ujN_FWq_4M+lM(tLO7g*ObKQ%pp*ofWj*oNwWejP2XAQUPej8rryz z;#U2!poBrDTo#~R+^50)d9zHxq7hc5a%#s%EVBge-fm!Nc@2(p)^d_Ra)mg;UOZ8y zk`J|yrFkheIjZ-&kOua-=g9pRc9}RAiP%*wS187;raiwKocL%V+QV^sYAL!#ief>t zQ1_yjaSk5cArd+K?L*YEkH;hDNcZ~ZhC>zibq;TExF4-R>ZvtsNOT1fwS%W>jH-%3 zNgl1sl*~t1r7ZZ3)hm6B*W%6qxh>>>zo}lk*0$forsPf8LxPRY>rt6`p4bxqn~R^H zzM!=uoG98%N=Wk{zD4s1q3^?Zb424HE1U~v6a~A1{IlR@Qn!45T6%hc+J&ac3|J2fTmI7SAGS;L_%BhAlxIfRtLnZH;sS zHc1A2Zy*^FV@#r;7biE@=&l$Xs=NB*%NwY5A+ZksL3qQk#T`bj`GCJ0?+UTAdF%!6 zymQahm#z%%SRP6ycAV};_l#a{>W_`HZjj`k|9S`g+ z&0(}EfZjR8U77li`ef`Z79T;*MOjJ9%Sdl{b11e~4$ww$UM|@$<%K_8hR?{G-k2hK2i!4EVenhF>gH zi9BU(`VV5&S6T;o#~)#}=hi%2p{D{lG|#RK-wj*?kNZov6oeP@(l|1Jrz~Vtu`O%hh)8E>X5J96Y$d6Ir7U=1yqI?o()K~(s&-HQm`^{{nQCjd zqGnheXEoUfk(Uv_?NQlZ?=~1EEmZmB-*j(!&P8q-9cW>J%9cg`sruv}3{!8_aOjdr z`jxB$n9!Hvr@+TWEUAD&+Xv1tqRZ9@J2zIPb z_6Eun3NEXFfUF1wn#OF_tT4YQ3QoS)wrB|d}*8`4rj!D zC+bny^Et-k=kAm>otfvi&miEr5BkPq?iZbc&~YZRF?hbQ9lZ9JNvV2F+Jt_?I@JPT zWkVD8Fql}ztj|BvR)vfuXY-~x`iH^~RBqt#xk6W;Q!)?(Vi zXW4DJk!I%%w9Xs~m^k1R@^Wlr#W$2xVp1`oCny>~#R)EHYjnsR71x9O%5KZr!c;P!#ok3j_aux3FG#!@&8I;`811kaxGxos~9rbv~Ib75lfNd-vA(k05I%e^( zHdysI8+nEC?q?$GIZLO;sjGLpdOf0bFH|c)jMTVj?WK*g9NO?BIpB%Tocjp*e>}ZD zFq8<`4sb#APp-O;pzoFq;;V<0p>zGW1ImGI2W+%*DJ_BamZr{oHji}W!!5|7?SRag zqH^%VtH+Tyf_6;g(UrCy>+A`b#^aFys$U8RZ!Z zl$6-6q5;7aftn>(>i!)7MYk9hiDxyM?azQd6I)^};w_ukXyajhUF38o(Pf#-(qLH= zXfQ#mE%2H=?eRbX1Qk6%%wS6|uB;QN}t1KDh85IXWvi_hh1#SSP3 zGE#Nau=|%0Q{z2YAtB>*=ZlmK&SMLoGbiWNjxfJZ{g7N64|}utF9r`+b>K)FCf`3g zEwTDNDa4lAc&Gsyi5OqIm6}^f2XT-Ft4d$&tn0`aZ&#MbW={?mdc7{I$)IaSEP`OE zEL_!UT@#$l#CHa$b4&oSQ6UMRRvevuGsA^Dkl~UBuu+Y}{Sk)ARxGnW&Fgzl@bkaO zc6gUC#|wCswgK)5g!z`JSWWZ!LgmsuM{mYk}OVam~VGJj`b4NBqu1&(w z)p;(-l4-FHj1};3P(6CmUxZR+<#qu zEa3>I*pQ(5px^7^ynf)vX&rBK!owJ|X)}$3s2)PVGL!WlP+oxqA{k;%B}CZiNgwMI z!Zh+9;Zjeg>PLgTEBKW2LFe~xXv?<@)uQ^?UYj~$zFlDVp1%EZ4OQN7x-ZfCSTz`* z-w%E_EFw9!;Gi&sR5s=hafX(6iK&W?UVdT}f{}?gPThG=D-wJRi*Zz-LTw5C$jQo9 z$vS?(q07-`;IY^*AGb<}tgc&>HbkOw{DDx(>K7-PS8n<61J=POKp9$o-X`I@)jJn2 z<~G%g_7$-Svj3Z}lc_tOpk+R59nLltKh$ULW;R1yltq-Phu*6+U6tFs&SCL1g^KF^ zFS46$TUzqRjOFhYY3}GD&#svd#UFCYl9uB+1xmf;!V8NBM6I_ea{o^!^d>VZq6blk z6%o&O7MNuG>4XjqoBtHSpvtA6ZFYQ#9)*BmN{Y|>aTEsGJo)bZ9qOQyM?A6}$4qO| z0eB+sJ1LaTWu$+(FKXvJB(?YR8%&rxQiQwy*s|pu(rvcOSwK<`5_qZ#;Dvz0DRZ<| z7$TzZK1GZV6K0P_zEioMpqd?6kIL@ibysDF)+? zc^8YqJGgX`QRoO8p%)>>ErtYo&~##U7E3GY(!O?%h-=F2mmcjkDV01X3-ND_$a-YK z-5&lZyafn~PI%$nCFScWn&07L%rM@eI-(JO<7Q&ar8;Gz%{0h{>O>?a4awO<C5V`bTBT<+f)-V-fmSLMexkF(q@69@aYeCuc1@@0M4`mooOxM z(I=fBY;lUDd&gc~lVkw}=l*U3PUa;l!sd2ltMm3&ICvtj5Gmd~Iy-g3+aBVOD>{9B zl^L^!d0q~XSO5L^BcNOe!B+DDuk|&hH&HSj!Ch0$Mu$N^xDZw%c>9^oCUd;pLa6Y4 zpcAoi-Tlelmm?NWB*(5+86bh;Jn^#@4Cf9J%=cL6Wf;=z{P|y;I}R)wLjR(Y>0pzp zBmM;9*H{$b&XZR|LGIsMhi>zgW>)1%lW#^wbv|GWq7(^J=>UBh;g z)=&RmNwyN35H@CO`{$@ZtbUto;QOr@NWWc(f9IY&0!Rt~{V`iBpiI7xSrazprGU&dPXu zj-u3EsR%8+Y=$J}Y-o2?S)aUdFJgQx)g|Yb@@~YjiZ6r^wkR$Cf&#ltZg_T^aX#{! z+n}C-GK_|yrFUQMkLK?nVKx9mI2vbnUm39*sdbrY?FSp*&EjhI|NFd#+14P+tR1$k zT!g}<_*M~#Wk794hnxp(^%JlFbb`V4FWqg!>`O+X7GyuI{>=LY?t^u3^@Qw|6vSni zjTTt6Sb`Y=m4uZoO;GjI&<&_`>eIV`!{%zFOJ0OUw=pgL{NA4NO+OS#jE-m|m=$o#>Xz8=E0}@6!8ba{5?JbLRz~+Fi zM)6O20k&0Z0zJeW6#D^@vokAnYdFb_1nBiB!u;NG{PMqBguG3U-%wgQksXj!G)O@u zmJFme-kultlKRzZG584I`PE3&n{!Z4&f>IF+91A%S8sP8HY%*bZ#2m;Go;_3W-J@T zcu?)&D?8~<^WH{juzd&2g)7GPl}pjyqy^8uuZ*_d{qPvc-qd#4uz*jycmpf`#aU;M zMwksh%a>I9(lt!~vQGdh$lnsr;hGoILXsf;pPW~I*rrlAFF^~8vDkukZCQ!0+Y3LH zx&O`Of?l)n*;$gptvWRdKTN=ZMhIRm?B| z__}dtE@!EbX53EV8ph0du1HYc(Jx?T0;vXM4HQxPn-sE3`qNR7P@ zMexUXodKk5m3AVl7V_jiOW_<5@J`VYh2Q%GAs}z6Izf4Y*HVgPh<=2Ojb=fNNdT1P z&Sj*bLj>##ZAi8`f^2BmaRcySWoVN63$>A7!=gT7r;K+3j>zD^@SFe}MHN>cQY(S( zuEQCG3fIrBgla<|B3?S%M^NdNvXZPzS3~(H(c^<$!93}!GE6+6-92fHCp||UMw&l( zV01OXsL?mHb7JDh2~P5@_Pk9Y8^a{Rjx=r;r+N@Y5A2-HF#p|>0ljg?&18NVPl!JB zAUVbzhfTt^rafy!e5WfW;jAYyBHFBspm)2moFE2EKtg{V|DU$5JRZvZ``fY;Q6ePD zzKktnxmjW?C5<(MX(;Nd?1{M%p(3(WipW~XekC)=QrQdLES0hEGewp`*6(?`x7Y8V zpO-)UWtPwLoadbPcIvkR#eS$+3zwX*VM2xdH{1(zSoxag5dX6 z0BeQBZ&t`jn7sbA4mLljJ7>3dG-1WXi35=7pnln=u}oA`vZ#tjTo3*BmBiL*Rh~4M zDTC{(kjis7eu<_yhLAl&5itL7cWhx7)liixD!MEmk#f@#2gjgVl&O6XUt zO@E~sx69?7r?Ge(KH|jIHkz*bce=}Ke^R@-cKUCI3NMBv1G$qt_-0Drk3}dct#I8{ za=nyexJHP*7qZwUqXykyFmTZyo;O@CJyULEMQdo)XOS8~2B9=SovguH6TS^euphn3 zeI7Q+u0a`}Jh-!V8__wzD3jVSBZ9+%WVuGz;l_`!RVF>r?mnL_(~UQ~hycre-4b&L zm~s!Utbl=m&(+-6(6-cc`$E@KB>+sD7J#E1z$wW=lB z6aV}9+CXH2Y^82)B-Hn$VuA?gP5u!dY{^^V1UF6f#s0&F)iSYEBNQ7ETp>E$b{b-Q zGxJE+JHY&oLXG+5eo^=ADOEIJ3h|tLEQM*n{30B)>huaC&(mq01PC-|^#H`>N#Jz; zpOzt60LA(PCjY0=EY=x@Bhpu#qb##v)Ng%tWnj4rp`Uxh*D&wSvOANJwPamaF+)VXx7e>iR~#VaNLbY^Qu)@%MVN78s<^817#TYmPeovIIic8SA7}We zp=!m;Zwo_A&afaN$G$0=^2pgQx-;X6x@H@>LN-q(U10yZ(LCTE(SBTlGjYKY5RGht z>rDCkSG_5bKCp~==hR9e`@2Kx6q8h=88plEF@X7u$z~m}s^;R+*`M`OY*Qat&21ah zT?V0`(6Eih)X>o=Au2g5h>=gD+{OF-PfO#3eLO1W-8&zLz|iL5MBbT-^sXAH(gnygUyN#ZSPiSq~1_k}jB z#>%~FKl3Vol@Lp!luk+eJPqF!C!*HML%4Chn9(sP$NE->kL1kEznRMk1((v&c z!|<)nCWDA|`R$g)>)ey1YJhq&h8&)mSOg7Ih8pc(Bl)X?9-!)RGqTKNIybx3?Oi2OJ{@==S1@USTqndLw2xSbpxLTDmGR)NiYC!E{oqj~FGgvEXItdAJ*^ z{E}D?jqy>uFU%Hp^QPj###LW{XA)OmvkQuN4W2uq{zGzO<6TJ@Vq2N__F5+Vo0P|g zVQ54Fw>#3wVZGZLkx>>nRy>FFZ@Zf9H%Ino0H{Fg;hf?_(jfptYue=QSp^;C1Uw4^ zY1|=zTNGc##N-0=Ziy%1#k8TJ?j@UX9+$Y)b?NQswHtzerkAx|wmb^F|M%t~B< z6|zV3fwBv(=T0Vqli{$#B~{pY7#cbtoGf#?Jcq|_jtWr{f@jwBOC{4zAhQ9o{dv_j zTtgyZ`{*%EB;PQ+h#%2XhAI?!(N96gF1A-fni@W0Q6PD`+Up-59^TX%i3bBoI%U>k%V&DW*wVP67S-ZQxj&7-ON#5B*pswe%IcJtkr_(JfzYB^n=J+ezs%mX%% z*Etu#giTGFuc9PHY#Z;0Qy%*sysfJ>hfQ3*6$+qWx|96}j;mmW+G|@$4}qot$qI$- zGR9R}&e;pPNrVU2!CiWfsG8m3*0TnwScx%yE+UAdOr-W*CUV2J$z+PBvV1(GY8?RO zukGe8E@TOVuT`YR0mP3sSynXF0;9gwE+kQW2iB7ZJVY7giieHcTM4hH@>_eS5vc{l zfHP#PmccHoUBAcY(CQ(-7s~8jeU+CSuqv`waiacs%`V5h<8NA(F~PkZ4Bs0gWE(Ji zg66{Sr!GAAW?_kFxr=b)3;Nff1t9hS_(=EqB;k{j`aH*U=izAKfQVH6j1VS4N3L&Y z?Gc`6{#RJa;wCRTTtkCeum`F5=btn{HpC#-k|o02_<`=jA)y=@LxIV$V%iX3ACWIg zoU>f;@(ve`z{UWID+J!@9oXIPc&gvf9=I|2le+_3rv0v&1Hk25Se9fFzK0)2$Z8kKMMhUn$v>G`NSN947DrBryZ z0|q7Y^%5_p!piH$|GAdPA#sz>7o?BIb#fcd|kW$p{Jb{?1!u?aAUL7&Q7O>gLaS`N*hy-(s;j z(5RU9ZUA?x7BF)f(vIVmKf&seV5q@4$*VNhf8pHSU-PwUH4^G|lL_lg&f^PaZ{ z;p76Ev4|L+ok@3MMQsi5ota>~WxCt1ZcKdI#FkU9Q6?XF!`2~WyF|tw)X+r?Q=ln+i{Qa3a z_m@_BMOVt}xRi<=$4kAlX-o5!tB%8~r8hFRhWo~P4Cdef=U`!6(t-QBuahp6)zu9x z;73m*(_xhAy8b2=Vib;xLQt_k=r01(SS|cm{BQSP4Gh$IIvXgYf)CK^5TgRT|Mk#% zerYy;mHol~7+v!L<5N9Us^_1>U}V3RCMnxCTkx@ID~3%4@>MwJ?KWq`HvHkov+S^{ zvBF$(@;)tatbe9IndLG^A9cAQ+`#@gT9v@PpSjt9mlYJ+vuNDqi3~ZmkQhZ36_soC zH+DLGo0f%g1R(5p)9f-Cwe&uLn>Y8Mix*erSl`U}52j+0e^uCwl-`Ei>P=8Um?-4d zzAPzGUc@JVS!kmV_VbouD0(&XXp&1QRd;gk^3;|L>tzm&;f0g2jOQBmx3(~Jt?lXJ zwZ3%e(lO09+YV2EJ_C~I@IBG=)i=A)Ytvl^b^IpS0f34Rl9nyqwRi8{fXy{uc{#bZ z!s)k$#U#yj^d?qe?(mfbU~yi8bVlUZbrK8j@ffkQEGO#t?}ufiG)=&whi(9tqa6z0 z)tR2-K;w@^+f(v~K_O;tcSvl{c6>0MX~9Ao9-pU++`bO)Jyjkb^o=7`u9NI3z6Xx1Ftde!N3oy5;Y3z@Zqg1tKI?PC)6+8>T)z7# z&$m$q3Ui?2XgARO$H28~1EB8p2(JyE-NI!WX5#j#DVO26wz?SRcS*AZ{UZ(Zltk^z zXbc4x;;b-7x*`&R4+kk>b0M$Ks%S%PhIW|h$Z+L2zB-`bYA9RnqgCEaecBn~?N6OY zS>XgS9k@3)f8K|sv4$FK(fM{Y<@c-b zKsaDw)(fkOdK3|cY&YHHdhk6Td#Bp1f3YQQvP-(z;23KWa~jC!JmB3@g$-~C5VUPB zp9%M#@gf|R`j}apnoW#`nW&Pcbywb-JgbcMw^ANPLIm zioks~{Srb@Xr+I*c`}txM;tz?!VwP)QHmL;&AY`o-%^*FWJIIqs{lge4V9iR+PM8| zRv&Y}t2KwgeaMZ52Ru5gJ8N;Q!Kx%HtI>C)P%#a*&-g(8fDiOxymUZc5}r*TZ|pBU zD^|uzB1?TMIa~kKU1yO|pSaw*aJ+I{IQjGrkZ_LG#{@4n6@P3QjTiBSkNd~vJj0E9 z8+bdr$?mBTLGdDg(R#zOrJ>U;jfvMI6F++tli-D9cgzth3b-@<^*%p*tgGufqE5hO z{+#o6`&A{RnIA8tqgdx`E_OiFzE9dym*sZvZ4hraW#YDXv3{bgy%Gdh`bu_*-B+Tp;sX)C;P zgRP)wZ0!e}W=N4|g|-|sbA7xwdmK_>%sLz3#(MK*PpoXq-q>oWw@#hK#l>-d;XIdv zc2|7HJ36McN{5cH|0J%?xn%o)$}IY?y7GK4;ydb-pxzQ3`*uS>Uq)vAJ@d^dR84<< z!Co!4!T`XnENxo`L$dqd4eY8VG^{Ylh^eANkV~!6z|{2i$b`FX678t+3=G%;Os<9r zC;GHFd|)J(Oq_*ylihTlYi1@d^jWSJ_qJ4yo~l$|$?^rtCI`%S%cCl}OaF-Y>IW}I z0+Vh8u(H*KXOHa)TpJb{@}A4T=R-c30P`7Xj_8W>@}Im|M#BOE13knowpJOmbQaCe z0~HkMe*cW$1=4+#19~vC1gg(bO4QQJH(QIPXX|G`dgTUUVM|NPcN;10=HFnGyqoZ? zFzY&~j7myMkqmKs_$K4HxVWTrSLjgc?6STLv~aKWzu0)pS5y1znL!|cLd%FnQXk+l z+{)c1cWJXm6i^zZyj;v~QsOpCL3$GVN?+Czmny~lu|isj$3ZsQ*RO%2dc{8V@59&g z3ky$hqpwm0UF~fn;5cEgO2JT4(w?!AI5pYAtU8`^eyrO}e^j(5EzGplGQf|y$tmL! IBf`!90k&y0OaK4? diff --git a/docs/articles/intro_mash_files/figure-html/unnamed-chunk-12-1.png b/docs/articles/intro_mash_files/figure-html/unnamed-chunk-12-1.png index e68e391deea3c6752763646dfa22f9051464afba..1ebea88186cf7866e3827586aaccf7e66d64f44e 100644 GIT binary patch literal 26714 zcmeIa2T)U8`!9+JSO8I?q9|gaC?FtRI*QVXNUuTZy-6oPM8zlw(mROs0HGr-L_w4; zy%#}15+Jlt5<rTs55^m{}-TI9j;cnB7;nML{7J z8metr4VuZoy&CrMSbN1gA04ywTNQ;Wo}yr4#|vdX^hb z-2UJT8R{F`O)aRr%W~?X3+uzu*&N#EVF#A8wOc(7_z+&R%3usGY&`s4c|*35&*$A& zvw;RTY5!0g%WphWJHKU)rJWx zpInI!nklPuGZ^87R$p)V9YU?Q;s(?;_2%Nd-}kjAG1m|6WX47XooWnv9`NR_+{WD0 zZteC=JLPEjA z*Q+qtZHm#b?DgC{wV54grcW<-#eal)HZkvZhiD+%U(_MRWIT}23-{!ccgeEArxL1xL zxHQWn&4H4*!_t@QSaKiJZ8i?2)b9~@#L~B)@>J=Le8`Lx;fhJXe1D#E=GkY}SJs-r z!wv@LsBz=#Ve5$kH4Gp_UWFtg9U3LGvOwgWdR0adb3PFE>5Ai-&Xz;t)M9H-hD%1? zv%FUHqn+&CT-c~52zdFEdgn(Q-SX$I2bPOUB-c;i9`#Ghk`WYEp#cZdQvI|K8qwk? zC}b&=$CW}{I!C=R`Ayf{#wERc`JzHhxLJUSo<#he3ecX+QVi;4sLT(R6n)a zKfuBxS_kxJrviP8s-vYTF`MT?ML5k!%w7j0#>BL}8`pi|$gh@$;gQh?2=>VONpv$Z zotY6>Eo+FTcvtWsWnaI;QU^`pWm%)sun*{I*G(x#`}x89dLx3|rWBH;#)t>|+9qF4 z^HVt5#c}>ABv||@eu|W#!an|eEf#CqGZdA=ZTmTCJ9>tq(|*AA%s!lzQPTyEFwg#G z=EoaNErE+x6?VcQ`#2b;94UkRct7kbwffDODB)lhkHO^rLStXV*UAd&1C;E`Tc2e7 zv+a;|9(nvrkFq5e1BPv~$L#XB3QDLr6ipkR(Z?n`e>kG6dM{ebg8J|o3S^*qzdR+? zY4Bu-V8@^bvTlQ5z2V$AwRoqQ$`WE)PueiVC*%CA@|b;kH58I-eejsb?7VI+71PI5 z9-}Uw)T~mW8*Z_ny4iyrKVZ(y{QiK6w1=nD_Akm)W2-af3zo`I60V+ ztU;t#Go6T9B!?cLkH-=UeZX};Sm-%XlNbZXl%0gUOF!fjCkdkq~x)o0kW+S76GkNvifHQL%s`i0>96dA1iw|5-nTa)s=^6bCw?lX-)C9n_O z|JQXLi~J}NOb-SmvfDnf9a9?RDpq?KPO6pJ=3?8Y?~D8^lzfWx1MDJU$Ft4P!S~tm zVV)PVW63PKK3Ye`9gLn{3;uPN{{bc5e54q@$dPYC%5&4!NlAWJ4B@~6A}_j6MdK~x zef*_x^COEXk>oZ=%Q`*$C1`*R^J>{|zQjBy{I_BvuP$5V_O@~qKV0jb@6v|^neM%v z+r-q!GZ4FmYM4bQv9UZG=|DiDl;^`-v@lH(iU`)znRi~&g1-iO2I9Y zXz!qj)6ezyF{kGbbD>)ZZI{f^i83lA*Ud%P&Yr++5a$>*b zSe!gFK(Uox_SGucW2Zx);#1*1{}(8$yY~jfa{a=>SEOK_gVp$x{Ss7tf)+c3MykEE zU9JI7eP}FY^i*Gn`$q;p!HF(8Dg-ep|DwE*Kxr~@&HQ$jZ<)!!6ohS-cxVBsnsW4C zY6&qmH99!Sa=(;wJ$K9z+AoY4z2?ZKx{ox|1y*9RzDdG-V&Alk6Bx2LHd%&p_56}df86nAsMV-=*AOUar1sC!(7 z&=rca8gx&$+3jAp8#HL)F9jfkj7qPP4sso*ui?E5LDI$Sg+lZe zwYua=Bi8lM7U#9?jcGn(_|IK9sQmds1z(;QP&G+WNK5)|%{8)A4)^ra`d@iDp6Yey zgPE5hQSVkD1?(f^V3I=3g(}PTg!<3N{Aj$2qoO6;D6s>`3DMAD-(hBtcHse|h)yEe zei&1LqKZ+u=owHS+j$_h)`(a+F@GXQq810N z@yKsT05-q}o*ACw>k%M-Q@X1y3Rr+D0&SD)<#y2vGRv^k6VA*dBT;-wW2Bc%F@E#X zZ-jGb17YpH@7=mRt8JvjU&WWhncMuOi;W+KuZNi1(zJtF%@2D2w$@fQQ;J&EWe*jA zIhP~(XEo)5Gm@$ihPg`%{nMR68KN9LOcG~VB|p=w?)RlI4JrU=g?wdskBp~OMBC|i zr%RLb4;@3nllHr2Q?Wf;@?2c-fMx}alN@ZT3k3z2oH86+ME6Cm1C)q8AnhHY(H9ChhaR5g;9S2oa^T#BGmf9Dt6DMmZMXyI z_w36w0!XPit2}%6o;`Rt1bgDUrLNOwRcLNsY$A2R{HT*RN*t@MJ}RRTWcK?#zlh^W zK#$7Ji+MbfJ6Z*vifc3mdCoUF#`}w|EH>VO3xZ^n6@T##vpJp<=INklFG?T1*t-%N z?{Dz}gUH(_nSVLU-$3~9XWkx3ifi}pf_GND7%VNX)t6aVe_F8GwwK`k#~H$kkJ1s@ z_3)VtH6FG_DhLjQVXNWG;Dj;%oKb;&`6mVJ64f&niUH9hCf(P_!3CtzL4$Dv$zJ=h z8!`XSMEo~ddsn_Fm^!4oH{=hC#D^)4URrr8FU2Tj?V~E7vyk%6`Ag`xD}T}mIKE3o zx*pY4AYTT37(#rc&gJu>`M><;LE8MYVslgGqc=Y4XbqdSHiGxXoQwSaum|;B1Dnjn>Oi}dt|JqdL+P%~PhYOw;=GNb+3>_w3-cMH0 zVL!3>rkh*P?TZr8zoYQ4M?$!Yd>ZK)K6@xsR*bQ)As1`?9kZ|o*y{HF$?f6BKDf{N zcEl6;KU4uEJ{%^dmnJBsRzfx9_E8__51XaUIXxUf=tfkp8RqU5y#Dt}Wz$}TG|s#e zTtCuKK*({&VFac~B1a4nqIRssMNMG$(U-Q>-)`oo#$b0R@$qNpn4a=_=!<5dHto2a zlmSPg&va%3V@?X2fIONiy`Cgx*qVrMP=^e1b@Kem&P)&upj^EycG;nZM@RYK}7e#~c3=Q!y=_1jQ~m_dVoQ-D$(4 zlI);^lXpI7|CYuNCF)cMDx~QWoB}ug5MByGp+L%E z0Z--FP(L3hCcbhAz6gS-Ns735-})EAUS2Np3CWbcXaN-}HlFf^lj8^DzDI9#zSv}2 zdhO*j)dlz8E;ZO&1cWv*Y<}s+xW|aYrg8piw2CSFVaa(VFrfy4L=jLuz5NXgGCQ4M zDSWS6c5Vko*))Pm#A&?NWvtqLLVTqWgiOxx$yM=$xf~1RYv;ASN5nJ+U{N&%G5uqc zNHz}@$xAC6KLM-42A(P^WQsV4%o-jt*;TAwZpV^35(Z3 zGC#7LTxx-3DEvL{K4OG*4|Q$f=<%Dcq}`qk3zn*#1=7D=mN$eXC+N6m+i{WaO* zTy=iw^F5asBWpU{6v~-zT(dDVq`fB8$+*zRd$=tHT|Qvq`>D(}*2Z>-fXDK6Se+r8 zC1COksW?<8n|kL!qNC0YXmfB`;byEj992?qa*OYZg5vVbH#)B|Gg97m1 zCW*D}X&ztZtxDTp-Cnq@^UKMgGl13t*tRo=33%aU{z501BY$dQJO~skN=r~hc2;YG z7*w|?^QG_U?6WfT2kIvL^)G5_m_nQ(qZRvwC@;k+UbqxUf~`=|C}zuaDoLI+!uIcQ zSdD&n2(9*Xz-qQG**YfvUjfMibY}C)m)8E{`;8PRt*Ph&iSv32`&_2rm7+6Yvbgif zKg%f%qF!|+m1+Hx^7E@ zCVS5sWXtWR4o^AlBbn_TRZa6gHQwhG{k}jBR1Q>2Q8Vo*a?3yH+}qf1xMEQ|%@a7v*lBCDCzhlgS%8on+w#K?%_QZ`3h6lE`GReC%4txKat}9=VAS?-DY%X(+8*pxrE?b9+pTxgif9|Oq*itH|2%E?z9BDar`=5;P7YyUXVGWwo zFBwTRD7S$+Ka4@{hkf$pcc+~8qpDuo@xS3R@ahA7Jc@gzz40U54s^(4>UI2uaO>_! z!L^Fnr`LWN`>%b^RPJFhJJXf#yVf!(URrcztOHvW2^0-iR$3a5!66(QN=83RtGyL3 zuP>)9OulOr>wJ2;rUQR+Yy^~Qw7_yLbx&3VTKmX8_u$H1{(v8KgjUexaMsj(ko`r5 zUoSPKqfAvf+kW%2keZChvkLCVz%p(f)Al$6f5n#En(Anvm9U#twH;7vCr&Btt7S?&bvws{Y$avy-FB2 zJTy&EJyb^=(Q-AmvNS+8iY)T)t1DnfPVW`_AJorzcCM#g8u9)ASy-H)oXa}hUU^m8 zA&|L`d;pbGsz&OZY&bM{_w;XfC`N@*5ldW)u}Sc^Js+w|vnM1Tqcqf!YH9MwdHLPz zYhmi=69%tocqVRwB`>ZF%R?XNL~i!Bsr*tsERzw)Gq`MsK${v(N4xW2xPM1x($aZE z70c@+IFXW)I)myCcouL?Xs?_AD#NtE1Itt%FSr2rx|dodC=b-%vSo0sX(tPKn}iSY z{)5Vx`#^bkt$gt%unu&&=nUEC`{S}jX}B4q@EMc}r$_604cv(nQ(AAu;s0OGC7qV1 ziQ*lXahODE`SJjph>jmHe7iw@Q1ZuMm zr)FGpOmZ1rgEXll&}!Em2)rgQnhWRg3P2>@(5hcK!Le zWc4e1#P%L=9=o~wsWA~0?5W2`%i{@w4b76QOETD{y8CT_2-|sf|I57vRSp z14%CI>EUl~VC1kv3)pEk+u>vegFa?XFklfu``pd^mCV<-YT#`R1z5V zZo$s2omuxjC+g@&VqEGev^1Jz=tj+w363p0x$v=IYq#X^6wo7lBVS8jj*pA#@@3Cv&7OT9$21GsJ?5^b`RBtqXv-r4W9VTF_p#{wC?N4eSzbl)`` z^o#r|_twFY`(K!?M;9HOOm%-TYPOqxsH0eobFLaz8UFegC%>|Mx@-Ik6kEjP)y<(eQ|mQ@GQqTNBljy}*=Kws|XBfn^* zJ$1%|Vh23*6`!2EJ9Nu4Wsz)$2nj2j{eh1Eyc}l1B*ralz00&50FiY_8RE)D(fbU( z-=g$z(EKICr?gu#Xzm>**)_mD^5>wxHoueZqI*o{T9crwc3B0&uKz=`^lQ^$f_%So zuNY;(2jpff?p2(E7ngg>-kh!NUk`>yf*6z+BEQGjPVUqKjfIU0r=^%7fnNecUp`vP zX;Im!kY%yU^3pop8o1t8i^fowE%TK)NKuY0cFhoC?@p+W#^vx;=LQ7_5e{-MyX)wiATPER%h$uQgPGL z24U;;%On1V4A7z`2V!}bh>Uppt~#0qcipXIIEc}`Py1h$j|vC>(kFip_4ZA0?_40a zqoUC`oRBrgeDSxB-KV0bIkAk?1$T()nnVn3TE)7M zJ$`zyJ!{?rIJfMiQ+HE)>v?Qx08^7LnlF(9aCfHcwE6nYp#vU2P{4nQQUWY1<)O;b zuP!U41^a<^o&>;Uq<3-c&pSDk0H8+r+&2Eb|4+Bk1tjuJp*rvPL4#ijs8k|*&tZTz z?H~Gk$pQ2?t;>$?FW8lU@S2h_evj!FE`P4j6M{LME-nZuVG>1JI&Q1IBz? z^)aoZOdjAs*#MZ83@(unzjL21k>Juh6Ez`-5&a3tLcD5q;;QwDsNp$)jkP-Z_#8c$u@3 zHAz$zhrOtVoj#}@{Z@LZa|C!R*3G_i*w-iA1k19B-x@HzOsx(sz_{7Bvhaq0zafK__k0lg9hK26 zOKgVgezm6W6{*#}TED}{PYgD(1G8{mrI~#eik1fk+~7x>Ip_aSH6VKx053)nEx(5Q zKeva$fw_6g_a%*+`;>hq7C_uPw&>sOuAcw~0pobduCceSKl~9e`v*%n8O>JsUm{Mt zYjZF1{jo(+4lsi&fAsWMsS|Xx_s;x`Bro44h2K5S?R`w`!{R=?{<5%rZ2cD7-&KFb zvJb33!TA?Ie+}d>68>7jJ}bD(s3+B}BXXI)8r?OtE(~A1%lA;Ja+K3|ByOuYVyUm= z*(tq5`VF<6%PxII=UVGCC?(9DeA$8iYE6c3Md<@-k9L|GHvhOmp@pVZzFgV5?0z=? z+6>=v;1;IVx4uyrPCnydwjHaL)}iT?x@D72u#4!f)|z#Jt8nDr@6lp9IBA0^7|OxA zb%zXe+|?}G>;3#x5<23jeaHAKML!>qnN>JnL+%aDM=}T9~ zt&U=9A`p6q9t$$2s!F=C-&U0y!CoF@WyjnaJeQS*;;wo@mIx&mG{WHlhRw${zijYj zzrsKII0hO}lY{Eoo?=jH=u|R_6SK^`tg2LlJ%&{Q;S`g^AGo8$@m2O8FhbgWLza2A zh?3IF%VXDUK07A+^uG;oFN)yMolH~j!`xtp3RjV|go*muK6t%&6vkBbyE!!5MCOYw zX!y3ME@J7hcJY13!PwN9B2XBiG zzv6paEs$okc^1>`>r%@eIYU8|+GoqKQVfTm4DK+y-Wg4F^Ky|oqCHU(o$79u)$H|&{fjqy zt|x^yErwf4N*0gV3t43u7JC|Oro}scT@YF-v}uvD%)Ti&7ICG!`k-a?)6%YY|}zJf}K< zp=&IMbleuPv67Ki&F(&Wg56eJG0&ekc=4>RUW!6y9avpUX6({RPa`#OO2Y{}8h1u_ z6=e`-&T?B_akLmTU}b>IbmCA4`>_{)`KDKX@4Yr+L-B+=9dcR$-?s>2;><%zM|(lj z36AG&9FBS`18u-zT^3L$*pFic*2u)z?MykxK@3oml+gGT_mtyP-qWMQDHJgF6JkQX zL*av?aPb>P);?B8ty>bE2rr*5*;6}^`(y0RhFy`8dD`2og!IpYEiX+Ec7?>Ote4pt zr^in8>2FTnD61Qft=s5w*mnC=)Tz()5Ggb}y>L;%2pe z#SJVZ@tyX=OU=c$mwXU9Qk;oo$*y&k%K4}I0a_2=*i4*UE|;%__LiJu1gk0r(Modb zpX-#l;8%*e$$}uP>M89F#O!00{cd*c9LTpJtr%G6an~;j5ZljDFl0e@&i9DB7+Q

    !CnDGLD_JZh#-utM|4+!zI|xuf zc+PtrW2$_ZdMV0;kL@PVgw@5RX0h5Dc$z|@yj-XFKzQ>zrT)-a>_cKs?$Wxya95P> z-1Te@jF-D*dJEDWzqyt1M5tbtB@zRVZF*#M(DYN@_TmiC1&OCNi6@B$>?F+JI)rw= zgW7&ec-=NIT}^t}nIB7_U5M=?zqTgiWYMxvOz>I=anie=nEEU)(U1#X#&Ka=>V z6zw2j=s2TUk7H12Q``Lf*p4weTYZ>8F(g5~^_d!^mu1EumupbjlOvJvei36{TTF~? z!%1d`j(l^c&q$E8nozY~rJ%U7y!X2R+Vo}GRB(s~WfdKUVWuLa<@ap^M0>|-!_!77 z)~hU{q78va76i=`OvK_q$wHnyUe0*ZSL>l;p4n$Cqnb=#T6s)p#IDFW;hCL@wtkTlI;Fcp-nJB( z6?$*imwO_0vsrW^kHNhqH?oFhw$+}5q{5Kag5>eq#vy~EUDx~4w&vh1x=hb1z#$Q} z$h#UNC2GdcbHeDfSbM4K{lUqS>5<6Ql*BXafS%JUEE1V4?r)sJOH6PVw7-}=bKe`+ z!)=y2Q`uZCzT3Mo0d8a2vFuWtf3Q8&J0ErZiE3!wSi z;f0vIUb8^_CuCvCfN9?LZ7;=Hrik6E=R_^sYq)B|OLV6yI-{cre7Dgfoy>BhpOW6CAJlA|pAiLe%Eb*t*P*2x~P9?jdy6qHVJWu6-mYT>= zW4F=T<>`Xjx4SIssOllN!75_K&V*!VlebtX7etEK!Veg@hx zDjC2La2)DN>nduHzK>blM2m&urDOU_(7uCKfgfkcuK_>0?ca%&-{piW;|I4lfK!K? z4aK%%NROq~wW$fl3FY38>;AbUPp5kSv%dMng8o&8%GsLs{+|2Mj;UUgYHt2fvZ0rPC@6N6L^z zlz8U%5@AC)5d0cGhPDSlUr)$i)cNWop{g3{Xl z(CH(5pYl#Gv2|7-lx(hh>-h4Lv+2=-Po`YS@nYDlsufa>Oi8fk3k3Jprivf`N^2uK zaDoV9yCTDWOkJ^#+T^u%FnghjxoKrj`%{lx>`(*tZcoYm&G&S@GT&2SzNLHT zF9gSvznLaFL}Ql<*qJPLKf*DU)fZQi<)w}qaIST#3+ZkNa?@YE=XqQve6S*tym-)2 zR5Wq3&*=3>UsA})v7N#J+-?yNE%b`JRtIjp<*gRAi9!+Z9Zh)A^=DaW7kXy(v=O+` zE1k+lMocf461|+K1YaJW)GL4f?gfXesSr$fXrWf#cpeed=m_WN8*4N$aZxOFelq_i zSs*V-{H9aOf=EH+h82580rjriX<<*%VclYc6LagHjwZ|-vXX+2-TUloxNUcj*PUrl z;$@ghEx$(C9nmKD?#gi5LssQ21_KFBElyyW6#OBMj9p3vb9@iTimgy7>OrzmC9Y_0 z=3X<=_G;m4!GY<5dPS|&KI8Wj>U4?c5zkJYyH4gV)yHezF{Wb`8RI%9G!oz+7>XNe zAr~H4MXvL>zphx3vY|WkO504<=*4BsxkK!f+;=9yJzvD<5i%ZNrQRcr?=*>m+U_%z z>O6K|NN{A?*8cGXlH zAj^fW7OZQB9hRlf8J<7nz4Ey+Y=kyA$*CTRaf@K$kdE0JkcWzF0rL>C+oX>3(Ov;l zCU6&N;qWO-t~$YjT$q+bas*R$B@^9E98{QX@bFOU)~LK0X}B%SRRKXtam{CbbA@K` z+qGuRIhN=`}V$}cNxNhsHR14%1(;6Q(xtnsOW$gj*h zsH|2IVG4&r$_w2Vq_mg2S=t>I@iJwt&)o0%m~l@;$JZOj`f053Xeg~_qLGXOZq$xMS8?rdxRm1BiS2qxM8d(_Yx`*?vWiA^WJY>_F zygh5Gl}b!b9OgXOXZ1E~BXHySfur8Ot3vo{9L2V-Z45zAj^wC2d{cvyNd_$BK@Q)3 z(6cc#`CUF5Ph7M2Q|%We=FysHAF{E7VcyPm+&5Nz=Qec^-c(4ikCH3e)Sf(jZA@gw zU!-fjYg3SHM>`m6)Mj7m-l0I`70Sovg;b{YI;72e%s}`{{-%~^mdm>KW`qI+qbFQF z7U5LY9)asRb|ge0O}(E$+TATFO`K>gtIFHVYa4*yP9AMnb9WujAdKFP%swW4F7cFE zihsHG_*G9sgGBK9!bPYylb|iy7u(p$$fqg<4IhGJViGyDQ-!_r!iG7Is->9p?1j)p zfkX?p>_y2%hq>9r*#UP~(id-LXM}gZry&W?Ty1#;xnEd>ke^dd4escWsQMKTp*AFIGG0phUa; zRA*ntBT-eCi?smn`|^>SNmqMahr)QASaE3yuRwBC4y`|}(jv5Cg0HT{elEUmQ>z^M z28S^5SE=%2KaYs%cFTbloaL%)qsjtLo+#iNIN~KWJaGM$!wdp6%h7ptqHu^S54T{U zZ|Di(S}e9Gc^EJ*a=jjsjknP7l+8yYe@f&cbHO)58=((oqjWn0!NJN za^l$7dT+3qw$dZ8@b`R_4Nd)kIf|F<`lIE<%Z`L81Z$t_eYf+&t9isukCfILtKvtu zr~p=quV7w8%L6B6W+iSQ4t{aX5I?>YKX{`{`t^cNtoXfxrTi_AN;<(3`*2jA1X?8? z)ipfihes&bFf4pSO4n7CFY>hS8Z)m2%-VcmkicT6=2;jIvD@x8!2NTfAN&#iXKV6D zI9hEf?r9KHag_2sy3N&?bi$1wn>f_4@FH?z1%m){3`1YX8fI<~9uQlIw%i>Msft>B z!gUoN7DP&XADLSNhg8ny#2h1q;vdohDzj_cenmY-$m1=63b|COyWkre8$ghS+6$jD z#&wBod7T*CWHj6)7YbWhs}pEPuWAr%ixU_1$_GD~@zjf}T;>m@oM0{L##iO09t*Er z;#lGGAD53S!gFAV4fJ_mShAz81fG#|kxI&R&arHjkL2#;~*G&sh=jOq`QvXy>^B z^L072O%(M+WSh&;d+BmwI_&Tx2+*|hEA27%;6()Cu)6V%9Zfx?bLTdrjADpjpUa$F zG9}gv-w8b$AZs_q!-{;ih&bJv-Y%|low?xgk>h$iFwcxh@va;(^8*swj}S>dRCy4B zyp9Ob?^bD31sSHZ^3f-Glr44H$X5g(`G&UVdh{#Etr)tPko1( z%x8t7p|0ybSeoAK`(QSf(EQj0Z_qJ=zwxT++m1JvYij;PZ!@{Udv(KL5!$C>Su_*O zK0*)2lva-riSUn#ue51(9a2SmeG$Aqe0QsTk2;Ju;CC_6$3DNnf}%ZRaJw6}4Id#d|J&kq+;>h|tYD*yiLhh31Xl6>3e8$#Q?{y2O*H1A>F@p4OFx!!OEKN(cm15 zc8nF1*6{Q@S084MvY*S6Bji=skWIWG?b-_@!e(jSd=NnjIWfJ-aK@JnB9WZJq^2J* zVbT@vnU8raFquwZpu#^HuXV=khD9+?!tKyR&_Q&gX2jZ!nd&AY8F%?bMqN z8yY<-p!)c58YUo(qC<2NG64cF=o9ayUn|XOPphn3s3LikJsubnaX@xLD+(};&?Ow9 znn&VB*L8Z=`x+y4{9SZY2_%QlC6=go6mKi$n@M#hx^&fb3FjeN&%CoyohWP#@rI0b z`+w0TCxo=)R*HH^vkTwWb@az*(9vVXw#SA8`B;OpUMtTSJap)S*jKDqOjL6jV93@mNtU(NlvQc*5Dkj z>o(Ev@){_Ej8A+fgta$AMn=PPU|wT#kv_De>7f{``Vp~A2&qvh#GF@oX{Wj~oqYFTJxropD7y`oXn zJ24GUT;?`4#*>sLPhSH3^H~0D|Ky`lo7j_jxZEHR-k+eHPEci5H61XQ5Yq6+e|H1g z9C3#E-=O7sCDYsk#kPJiDjMDDL(OBLLu{znEuD1UG{s2vNj+hG(i@S#E$jVLlzkJ6 zzqj^{qW(@X{y$DJOs!$%x1ApRIu~s!4U^z#7-J-|^|SSJim(scgi9{&?Eg(0f#@%k zvF0yZnT}`U2d9%ub|yfHbB(O!s|*p@(a<3)Tv3SdmfwakX}!QAKgjsck1I^`s6{e1 z+z#b7)JoT>V4adJCcbMd9TeX_)ZCN#-9rBnVgY{&&U(FdymDcEN;MDSj|y=wvfCS8 z#a)6qSms<$_<3UN?ujjBkzuiK&tg5A$)h|qDpuQJf~0ylnB7Cg@1_PowV^97xf7gq zbW?T5ejAZav(K6E=gAU}Ef>B$EnrpSIqUAmwEe)XUziV|;+yf3joT)^#UlSR&SD}= zUGm{stm(ksV*cBdOB9{=tZa|&VT}hND>XivK{)SRDWsUn@kE#_)H-&J&0O_$R6zL$i*4r4MJSyd4nvSC1=7xuHbSTk9U{2sT*Q3@ITW)?eA7OYer#)UNGn#eO#!^OPgU5QrTxt)bq>;3&r@Q-O|)_qf> z)B`A8g9*Q-%7HQ(ySjeiR@R#e#VA#*%9_1yD(5(~!@~_CEkx^|V!hGhKtSbB6qi4! z-(hu{IcL66(Y%hAuOx}I+6Hj0UtO5$sEE!^&b!dQ5u{!}%9|z#@%)CRY{J7$W_H-H z7vg_@OHV^JdLRX+-b=NI20alN;c?FFk~O`^CayPQ$+rJuO*hbKT{XIYR$~6MYNwh| zP&c3X?rpdVV=()!%2c@gm`0t`lq1|7UN!Hczr4|i>3Gp;K`3tY1bM1TO1UHg-!tg; zRkk4HCKEyBxN1A2RDG1lN{oCa(3-T)a|_wzNT?Z4ZT6$Du#+7soBq-^z%nlg!Vhlw zG6rRUIeYkg%xiD7B07Yjeh)|FGl$8eT6k|os|WZ22}buh&)~*7P?+*4uZwy*uX!!U zKOZ;j&A(0asBDXRMptSAW^vN;=SSaW&Q-LGnl(xaTiZ|Efcl#Z7`J%jz1OF|pGffZ zWvg1M?Vqd#O3>w}aLbEHLy7n47f_MbE#6x(AL|L;E>_0Tg35#f``##0^#$tMwCL=S z$;#Y@PG9>8r9JNqvo-4{Nps41fGyqe$+1VFpS+j1@4?+RLHH&X8jXC{gRCOKW|yMQ8K- z(6~=)_5uEOP~ys1a_J7a(sT3DK0 zb~WI51c%1u`{S4S$`zYinIj_-*k|A;MXKIDct&4ZmNi3!CLs!wlh;P)P&lha>99`i ziMdS)V8&oPC!V@CfHxW-Nibb6EDwW#7h}K@xxPKFGfF;CHG&N}+1#^cr8J^YAb@jz zutvDX;yig1V@j6hs|=8y_Z@$IUQGuIS$lyiw<$H8_v!=_Sfp;;d~L(a{nI1;%qPSo zDc6oJzm62jNA!R>{o4I^Y2HOnxCC&16Lk#lLUE(sN?zl9yIZYM6*k?Go=PgD*u2iC zQf({VB={;oWh}ZG8xhQjZ8_7x!Hu5FiZqN^xe%L=6~aT3hR2fD#`4KG%LBYSPW^cr z!9EoIAc>+gz)qaQCGo=Cktw&;>mfvxN1&6rRad7R2O3I@Wv@3_7j6<}B^vu()h|Eg z6Y^l!&&DTgHLmwjrsp~_#kcI8@5ivLis=-!u-J(@ZsZTEN~E%fjb}Nv>g_q%Nv+f0 zJ1IiDuUA5t`9#FVZZwUWt(domD>-9Eu6ZQVabzhnhj=fq zE-E-mdYRX#M!(pm3c9FZNCFAC;Q3W^Y*c;E>xzWk;m0>Pq z1~=5wS{up#Vf7sQ8ui?sdYo-(L`a+yr<&rd;*)yr}w1VHH49-o(ZEu}HvGWRkf zo1Gzr*+Y2C^W}imoY(LphrW4H&7n=Hn55#o_tIE#loiL=aGepMvKpdf5?^apJ=piP zE-EZ*B~EBlG?d7O-R0?ZoFU}Ov38Uj&dKuKv5swS|OT;j@t?bG+rg$|jyuLs(;aX+eGM*#dMc z!z;{_cblGfssgfW23BFC{7swhL8ChzxnrkP7ClZ6#BidK>d*jbU<5~5zq74bZnte~ z5V8X0m_uCn(5EgfT~0LNKDqb$L6ZX`UKN`)g04UQ2dDsTc7Jm$$+Np^VD;Ik^V%a# zU8hf{#}#J$`hucGY}SfLi7|KTNjEb*IVTjWFOT$LblGv>ri==(kH1=pk#y3F;ukw} zNsyGCi}N~I_3#?}R(Hk>lTzvhs4s-4MffJl*=3h>9;v+h+Rss=a^!2XjWdt;VER`p zw;e^-+uhAr{pNF#uci+)*GH%?kmEwSk1T^C9J{2i*i`XX@y@r{w`NDaLL!C=3`JkJ zXHKWSfmTRL&-di|LR!n4Ya&uP+{O6j2YHnY#Jl|cS0m;BJcN$!KitNRHaq2v{NxrR z?Z8?2HjLy5bBtwBlNQDc6Kc)w80)fuRdk{^dYUos0?vPGS@s%hcHwMnooUZKyToyOkK zx%erD_8ov+g4^CN1+eesZps z-436VPkO$@bIjxmVMz7hS%>b#QHkM4v}-q;hm<&DK1HYpGH7}m(X6d!Am-d19aVmS s=x^2Xr$hNGjla_P?@k7~j&~X28IpLB?77lO0sK+Aqb^@~%k=U80wC|X-T(jq literal 26710 zcmeIa2T)U6^fwwnuwnrVpnwI)RjTw33W}6~w9t!G=^z9QkN|Q;BPuG=4N;^f5Rl#z zuLx2@4?Reg&mjZ~ z004kh^Om{+0KkN&e*gzq=q=OAM@0dEeVi_8YPa3gG}OG@z3v%$-?4YpaP)M%=VEW5 zaUB4VefG@Q>h3w?lgbso&X@RF4BaIk`emxH^0P*RCEpd6hYSu%d&X)?_g(zvusYWg zU*94tQL*(-@5b%mtK_7UHMo+{^M9=z>sKxmjXA{M*LnUg<-+yw+utJ7rxO$;5?snZ z3^~pPQ;_q?L33r!^UUAM(5A63PcocN#0sAgAMt$@^;k*p#ov$Ko?&@sqR%a8m8HWd zt#>$}$bzKyPmoP-*WflPr{pI0sq;R(b_ElehaX2VEMyor`ZEMlU-GJu&3RVszLi{6 ztq~0@{c2DAb=$lJxlW_Mji=ZpK?dJ` z@SeZ3jq><-_aVjOQwAXcrTq8&9ZQ;^6#wK`{TglbE7Z-+10!x8u;O#W{jo_ePpGiG z7#K>-O+PWOZkgS`A6<<~%_(|v0Y3S+G(~7c_%T6dvcd3U4+b6m`OQcFni47Vr6q-Z zctp8H;$q!BnulTX(&RW<=G$t&hnzi5{^I!2|kE#xVOqotN{H4l5I^AiC0{bFHC6uSpCoI?Jr+!!Z%7=3a>akQhsncpKJ4Z_UR7CMa)A> zdv+DLkPM@W0OxtO-f)7oxaR6hA903@&?kzZ)#!8i><;tP9xeA9)`~LSTa`~G4APTs z49XQ&ytUntxT0feNX!;GS{!ky@iL9ss%o>==7Z&wkc@uu2q&`Kq-DcqV|r73RdCZa zbvR_qWU=I4oq1D5zRsr^cT?X0V=fPt(bEfU37-+4xhX!2VTFFz_2ssh6P+N!N;hBd&NVJ)zaR*Wwd8j+*bUq!rmeQj!2 zT`64Vq@dGKjuPjoShsAg3zyta@tnT*kMvjlJK{^gBM0oS>-U7k=1kL0qePRHxR2WN zPGrz5t3|*J2-*k6Wze>&x6=FOY2YHTI``Y7S3dOB=0~no!I3JilxBsL#XvEc5)(a@2B?c*QYx3lWVsv?KUxmBksF$jI|TKztQ9$BI^M2jRsyi5f^Rp0! zx2RxaCdYx1hOi}9Qjc20OrQ!Qd z!1N_WUUEDmJl1@AO!KK-G8fF}V@;Gc4_S>-9VT)?q?CWo;`W=sOw4K=6 zmasAyn5wQCs9y75+e#)(hG5Fcomsxs88k2oO`h{8etF>ddDTz=1LM9wU;YT;k0ty; z!5&T{-sjU&wqUw4ZC@;D1Xbwodd$ z=VCXhr+bPfeT*!44_3;aDvDSEy-}jSi0kH_0*eO&({OPFbWg=4-i+RuFl)|w>Az45 zD-LDww)mvDXHZSqp$rR6{y7mR{yPIz3)_lSrtj|;tHfWPbCf`sh-$yA^o7s(ojT3wrN>mB=_q*}XI~9EKabArTeuii| zwJSBTBt|(nR=06GSlS}PIQoh$Kru_{z4g8WIuG&9GNfR$E9EX%u3%n$m$2>9KDaE@ z*xhlagO31E99;IHW($KjyDt5balvBy&!(5b{n8h(NxK$mJR_ujnt zU%=BPeet%5b6pG%>VfWAlRFI;eqPmyR~_v2*Ri7=#1MnK%H#zJx&w+f0@HNh7DACN=i(RM21hUlCgEz5)ABlGlw*OcMz^JtV}p_l8D3 zu@;7f>5d4l&AGTzvBiNJD|Q8bzPZ&)qu*bgw!af{+>ISH^cU>KWx|Y-_PA3{;Ps*^R@AdfpVLq&(GU-lnB6P11W>*hv|d+(Ea~g$ zK0NpG2Cq)#l|`UMZt>r{U7S`Mer>4Rd%s0)Qc>Cw^=g50__Hc_Lg!Gn4Q~6nOHXx2 zcWGyk#~gK?;DRZkvYCp6S#E7b1s`1>8}GwyiD2=v^C1{PoQ>{4mEkyRLUPQyzn#Yj zwI&RC-!eTwOL8hJ0ct|Qzy4{AVquJsb~qR5WDrU!Oi#f0qN**GSy!7~8ZyF^z$2~~ z?V7h)Spz}zrNf!-6aW5Y=J7^hTmYGQNV*HPBKi=T9^#J2gTu;W%CQUGABQ6mv+Uui z&isbH0O-W|vtPc49{Me&0`x%R_@KmsZpBZvZTqjD^1IuRg(;6AH_SatnxyVP-SRs< zI%0S}IF5+!g&YTN7ji2P!7-kFLw*5~JHU$?3s{zhS4D-r*U4p|z&v;R{w#Gh;gwFZ%0s_x8rJYtQQzIZDbnDUw&Q zClY=a7|r0VB0RmVzzpe&^aH2J7?SYu8?LB64 z(%WeYu9?8KWif$zAuII3H-)8tKOfKQ`>>R1F-(1s>jeML0ps}&uXCnq&T&KN%nC1l zKoLlDhaW5D9FXo{H6u2aWY~OJl4`tdQt``$9Jxei9#3N{ceSD#<}=7RC^;*fkR@6~ zbuD-+g`^1+hz^7Gi2VYt-a@~5h#Tzj3V(6eJr}ukVNLrSvNUMK6WMdAW1hb{x>VNZ z=7`3@Uy(qJm2qVeJI&Uwr{<^tsg&=?1?j_#V}9DzUT3y<_IpLVo0Zr~hOrm)5)>B8 zz$b!AVr8ZuA*)l3lg;*OObi@$2e%JXy?kd)m(|~`%fZ+U>b7{hK;cM?h0|}h{*BLL z8n#KDnGa~dWf!56EE+8$A0_4p*0SxjtN-f|KXJlTiBNtAoTE9hA9z%2^Dq4p^<#-} zAgAD}j-~9kn@hh*aQt`=P+_nW0eh5r%n2Wh(Cr2CxE`J6a^|>|H z*!(tnIDMkb<*wd3b~riXGSIZo7T6nR;fFI)30m$)q#r%A+>a03`%K_v3QUIbi!Ou#+BJ&47nfK4dZV&@_aOXQ9Igmjo zw}oKXNTT{6b#v=q?Ee&*fVJ%f5~I@MGsA0pUKI}yn>d_#y6+XW9aFw!kp=I*^&3B6 z=NBUjx3YH~i20WqJXYPR8RKSe+%k*~zh{~BRE1NrAb79UyghN>(5ZIg#=A(XZcx_z zR#HO}JzKKvW{yx&FvBg4N)On9bW@NWx3`yY`*)3>o zdYF329I27s3_WvxQ4m)^8R&>Q;hA{iQVv}89z-VPRIgk#!qr#+E%q{qMFGhsY=*f_Q zYyX)&mMN4(Z_RJ(zq+?I4WDQ80D>}_dl_l;Q7>)acVhO`%C?7LAw-jP+is?)$6)w! z0tY~p4Quybca_&p&jPC2Q4D>0|AAuIsf%JWr-q|y#}_F*=i(sECR4f1n`hb22#A0- z`964!;A=jy;o^EOEUwhBvzze!PDTD=#MGLS!wtG|e8qZ|Vv9I-8UDaDru-2U+nR2r z%HrQO!5anUyQUEKnN60`B(Y{y;MB~01`|BaW_plO587Zjbbq#pIOLVs7Z?>ZFq$SB zveBp_4uXJ97^C^}TsqQ|x^pZ7+7Q6x`&)cn+E(RYBmXG1iE9%aI2c>68WQrt5@G>K zBlUdrqJU8{LzzN17zE zL-mf>*?PJiGBDboTK{UQZTKT)H-$1jPcv+t`~*J;&K1a+%kB~E7-XLgbezfGeB6>h zU7lU=-E?AA&zE>`qa=vAP1seN2ZO!LVF{;7zeb#$ds90r4Pv##!MFN?BNkOMvRcSm zw*Tz&)P*UR-T4$D12U~Laz4S#^EzJ!Vi&Ky1DWdd-oHW6%_=&Uj)(cc4*P|h;Vwb#MxHo*RO~og zFxOe5#$F9@A8Aqw7zoiR6y4JW6670wH6vcf?YK9Qp>?U;64(9XmOo2;l z{^p}1Ma1S9ybZsuBtS!eiH>_IvZ`(GzFOGf<$S4{=l=y3UGc&6q?s~h`aR?P*MCV| z%Ht?eo##T5G|wgW@axk|W4^^H#~(TMTQK6Nz?bgFi|VrO2z}^3AeH0wTpGZ`&?4|} zN4sBIZEW8#4c9pbWUP?V&i#^M!l@qFhr&s<@_H8vuCSjMJGe{IiOvX8w`;43_)ct1 zUriB$3QdJiwZl2Hfxld1DL%;n$8}q`aV z5&!2yO>gkg9Vt$LO&ez{gzFb0{Gla7Q88@4Z=yw(+wslbER#0Cp3rcgsSk29g`YPdmhSz`7X`?`%t4`36(s<#rH5LdU;vj{iu*0O1+$F?Gj}zXx=z^G|>Xn^vfoHJQ4Qr;BH!VwX4+cj&bh< znc36JckjG^N^Wo1-Q(DQzf^Eu4C@s(=cI2F3^5&P-A(@Zuv(T0vMP(n-y?B4_$^C$ zNOGyj2bbjaz;2~0zo&g!?6Mc7s%5W|@Bwq>Cc0LF#_@YXP`a+|n-W!G3`rC?H>BV5 zV#*BR1?(2U>YeLB)Hw!bs7EI!CImzhrOdT*0A=BrAfJo}yiLo@h<-NPO^@2?Z6-LSs3*U#;y&E;qd-uRwnnS zc`0Yu`++Q)QU|aeSCZwvu>4C_YR@k`eBD;LKOudpuC~?x#x8ICe8^oPJ(f4|9vSVK zHp%Zfph(;ySVX6JCSjK!3$us(8gQh@ZQy*yRgha%Mw>bE;PLotyM%qd?L*mxj@%9t^sGDBC$2Urv zR4(|K+D)okc_^P>d|NZkzVgrvG-N0t`irZ^YSEo(R_g2{x*bReExakr*-#pG^6uA} zzTzc8O75Ot$Vs@v4*FTi70{#Ky+l=iwyUTZ#sIyX{c@R`7dNcm^uVuJWVQcrLU_r; zN0zM9VE@1kCI%UmnhEj{Al*|kpB~M4BytNg7bh#PAO5uhVi%=@i(A)CQKkj`H=p>4 z5utLt-H}jCJEVe=5NREx@YDPMnmW@Vp>Du^{9xnvdi3aN*{$V{fFXKScmCBqlc);! z@e}cqhzqoU~IP5G&|)0-37vFvn9=aMQ#~btNq-5 ztuQ*fSV{X)#<&<(sa)LaOHj&{rOgyX$xp=FlkMZ!!YmGU>ajmd1Ur7p;IB?tcNjT} z*l6c2Seint@r!7_d6Oyrow9I5kFX)wshqitG>CB#$i?!64C&-MPmsx9~OPm@d5M7J9ta*b&18E(uNugB6BQ$=~<}=o(KDpTEtrvhfXY(y0n}2uYhmR-# zw!2zHrnPgc3l;{wD7=?PfyjGv;t@+He~U@whv@fK8)m%LE{~P&`zKO9iCeL6qWHI% zv$dN%>=YV0Y}M43w)-O=bU(wWj_>vMVP%Y;j)sP`UbaQ))UEVGANe(GejR9l{ubwI z`Nd`crDaxSt2-97V`$TpCJZu>PtCXr>be{98s1 zzEP^W#DVR`rg_;CV79{=}&8F7Y{o~sa3C5DyT%-^GV-d(yyM&88O?uqU3e01NSNoIxo zDna}RzBky#0M?S`%)gSF?*@wJJIDdhGw{khyeFayCDJqWl(s8-xs-q=-8sDqRI~bJ z+uyZl+qSRRqVUxz9oqzEdOG?>N!oMZ{z7rT=X8K@T$$)hG9YY&Ie8`9d2kzz!6vG7 zuw}z06-ZDNeGPDSg1;xmiatJDhu!*kw!&QLXtlyU25HMS`pg8&Y0f2!^eX$@B^i%C zzixWy2uK_RyK*ux7F_fmd#T|`c^MV#(3?~+jje#gyRJ+giL1I~M2`E(z`iP zYDKR2lXI3;-}O+u`}g9uu1oK1Tm@uf=x@=| zzV&6nUz{%b@QS44Za-uv&t zl-O=P`s%eS5qlhgfCYV0`818$y&BY^3)}v|k9hpcC=ck+SGN}&)}s1;5vv~t6Hi}w zuG^QzvDf6>>2Ir_<1M%BM{X7ts|119QJJ6JgPcNMQh#)lG zrJb%aqtiTH_x)D;j@IW(S;gx0rIar%K2*2TuOy`!>%u(wZ|{H5!9EF0_Qgo?=p^DP z3b19F1Tac7T5#dHQO@?7l9c|=;k5cw>w##nN%q{!9qlQP3u;Mo=Id4cY%OjUAkkg>LaTG`5IAo(h`>X{wP|aaxdR(!-?5%><*caD`J9 z(uOdptg6B^Yn54}c_7s_U*SrTn;rVl zbE>VGYwsdGeSl1R-tfP2fMG+7MCYz&2uM|dsl)|FlK}F&T!B=ziL8j~#Qo0yfTu7A zr(ZzPhOi=KpAJQ6wmE%0Q>sfR4tLYFRr~GKd#64@)Hr?g@p54{+L3eezB!4_EL-42 z2(8z>Pp8PWu5jWiV%8aLT+ka2>&SjYaO!Zv8VBe_&MAGrl{?0XTPz8blm+&P>sFNi zFs)OhW&c0N=3YVXQRAOWzZDJl*xFu8{=a{z)zH?MAO${dYQ{(a`Py{5EHO!J%`qnJ z#JlfgGBlc^f<@1X&1(^gEda_fJu*I?BXS6zR5gPyD8A=9ottS00>5&Stb~upb+T|? z02ukE)N2Hb;AV^CQ+~BxQF8`jbXdN>xy8Y3kxL5E-=he#Ryj>;% zx3WP-HZb|cTJ@Dd*>>{DrNWwMwxV2l0r-W*-kL{jHd?nn9SoN3-LV*fBIIj@B;~-u zMy;##B|rrJPH!W9!yhtJCgLz?9>f5GZl$Z(d}{%>uFkuc#V*RWM;hF!lAG$%lR7@+ zdrl#-O*}clKSz6csX}0y=W0YTaXO!AH78q_LCAejjgejiPN%ag#%WMS+=l7ER5v5!p{sq04_bL+tEPi4O-bNs z*TCy`7qYPTC25=vu7Kdk)x;)v`gnlm#Evt4xjCABb~13>;@d&qT+I3B@%~NzD?91X z=Qar&L;<*{et3@jK~pBA6HLka*1c>lt6XapkVO`JurmiI9M$f)a zFK?%NX=p2d!Yu)j?mn;Koq<8?VuK10AlEr+IB} zNOe3s8an+nn4-5**+iTzr+gkV6KxjF3jyyqr%b(16!8ND)99-wE}7BCHTe6oSXk#Z zC286HJ*gJ^Jd=9w38Dh4RphVA3@RMPf{c;_JKz~LnNx-CJ- zR#lu{Gs>|WR+EK36^hs}qv*+Hx#>sHZxE@c-3Mp>;RFCK&FuabKsR*ybW=L7@~e$N zWHV^3>46{{sM>zgiDTh|eOKshtNbmkSSw_vawdCllwK1Eas;ZDGL^wq*qyK6zSD3M zB)20&fh<)cdE)ms?~GDyDr|z%HZ15RK8=;3+_{vyI&Sh^h zl&W@e_TA?fV%FeQ;w=T6#bzdTsJ`v}h_Y~!P~RIKBo2hs?alu}l4Cs9G8)PL%rNNt z47<+x9iuN;oeNVkh$&~y+urChTjkP`+BtBndGmVvyz^*!wx{#MGX|?#1FH|Y3G!xb zJbqQqVH|M*4LR2!xiTkS+?i3#Z33AZ*Wxqnn4U?+B`pXP1}oJ{<9uS)u+=5yosnI66A zLo$-N7$(|D-+O{y3CEI>k*tZZE__h6-yJY`h?oCnRiC2GMn^eTcVD-aGInk( zcbw|D6qCPlbH8I-HS+VMvw2D0iWYCcU; z!C(t9*a+|DD*KTD3{oP72>NK*Rd+5MBi+kgtDh=$bH5(9Hc|3sF}2*D*(&aI!;GTSTF3P^k+j@4uWb7l3Uqq|3etE-ywllO%ah&-uN3>|%DU%|36wk$? z{T|h*$q+US8Ks7xo+&K<#0x9e0OPfU99Wh2->_z~ z0OV<=^7`Eo8S`hsH=Ut5XY07FpBoDgXeL9L@4pJt7^H7d#h)}U`Di46$!B7sY;h|q zBwxet2}WZ5*sWl`#mJg9b`{Oe-jOx_qnfv>n5|=s9|1jYZLX^3y{IBff75F>b?;w z1?{1X5avZSgcNdfp2xeQthW+7@kD57)H&_jM42^V>rS+NSDxMb@m%|xNI$RkzhEmE zi%cnw6p( zaV*Kdo~3!!=A7*)t#An?rMexHoTl3@T@8H3@aco=k;mHl>afwb>b7YIHhs9vy{Kf= zUi5a+NG<~6kf+K;w5_SUiDTvnxJVNzkYwj zhJB@Xm?l6TJ+kr!dTS&cafyAj((U}`W}`qab3>*qe|i36Lj%@gW0l0N7q*uiL`RPS zxu%KE@YjOov&7f-T?e?Y;sx}ylLJZmQ)G3dHIj!)1V}Gt- z9s08`Se${8=}@OxTcCtxIEe`9O=o%AY=&2>$maEHvngpPlcYR?^yVGa^70t3xI!tJ zf8~`Xf-dOpIs7?yH7g+gK?Y*3lk=4;jM zib!I z9tRaKQ9!)NPIKjk?A_2&d_k4IZa`-Jr-Xe)Z|;G5)kacMpVt66RKk^R9G$0z&hQ`A zTGql8q`)HYvYty7Hkn8Y*N4poc)H>^Gk52qJ;Pkuu@@m}PFYc>I(;kJ zv4}w5#R7HvwY3CZ>Dd>oJh7Si@Df6;hBK@nSN^Lpg~zkBOmIrOQT*-ege-@E`eyJ^yyDoTL?&O1hiJ8nYPZ#GeSOL+6 zdnJ2Jn#lx=KUIX=^|j5Ws#QP<_2Jt6hbNj!RKT}uy|jas7sqQ>5rGBzCAUJZUcFm> zfpb4})MOkPz3w_Pgpfj%GmRcbm}wuL#n7js7&YBYdWg^vXu0*~MJ79Eu8yMoRh^~s zK6AI$oe&EzXyYrcY`Cm`hn5>Vt~a53bT{00yN-Ut>?GLfxt!Wsl8dsyuq3V|l3OI7 zSCg%mzG-t~ZkAXq3DZAHeHX$O4|FCKZ6xDx-)1ICBs^NkRRWZi+IK?(%^v3|#Pm+O>AVJqm3ks*1hnz4Z(3NQ&@|?T~_J z%R}gNsV!J%qA?-2x=j3o5k162`cbM=l7gyup>88JEe-PZeg-MJ>%x3RYrXKQyDLsB$Ri=U2Unuq$lK}yGwbX^q#x_+ASVeJqok&lTZPVfQ>(z$Oy-m- zrzAzGRxCrgebzUI6Fl31^jra&&bybC9t{IC(33pbL`Z1%a^_P+%f*8 zBVCJD4Au?HIo&PPu3XPFrMzeyb(z{-Tp&p+acSC_>Ui5IKRrD*bqx{DF115I&?3LVq&n?K=|E=a=pb?7Ldtq@{*iJ_1amPtKKnOzWCAc5j@MZ4oA8qx+*bPC*hjUV8xQh7q9gx+E^^b21-FyDp&OXlB83Y^I^?NCpuXVA{gEttnD5P>QvkS_n|2f-W#_C z=OrHfjh2Bw+cA-_34dTc^^Q)?tHdbUFz0Kzfx?qZR1OXMrH9I=%51v5X2x*YdPCNj zDr{i=k>bq-hyE9t${0))F3I#MJ>&Q8TBStdf4;S*^{&~i)Y_)19hpB;t3Hk# z$Ka~^R2r73x=(6XQ}qws^%ynn`bSJV6z_MS@R)Nf*ZWV6@M?ao;Ob?S6# zIU8v`r#W~qnJ8RD>2kBfQvA)jE7M)eopdiKO!y3bZkxei@A_*)PEjoo0;JJ$NAs}8 zhPC$CM&4Di%pGHr=lJx*KyEJjZZ5`vL#fMy-0ZvV`2I}6hD>J?u`A4~fzAgWh}Wc$ zbcWJVWQlUx(GRl&*;~(zV|KI=eWY4?Wx%&X*X*{*`y_d81)1s&gv@eJO$ zYP1quwvC<;TLqh@9JLG3=~R8kx!E1VVGc~On_=p!7?#gsj%;@$BCOTlgvD1kg7-uNEDjlQX7 zXzS~hfZ0<>XdPPQFD4PR&G}3P9WGL&xXN9XuKM}1EE^l+Z5ps|hiOMBYC6OgDHd!y zvZ-!}3TY0e^e#4OQ4SoD0Pu|MpzOO8d%!2oi$NLZHDeOP>)ug}yk0PMaTjcT&RIea z$a{A=>n6cEmtWQ7@~3`eMtgC6A{T75pNAQjWWyxeUYm6|wf0-DKG^Utd_Wu&@yI2> zigU=%Xakefav_DQZNjWe=_$oQnkk$$QnU(@#>5RLYJ?C*X01JkIXUT(FRFiV({-VL z+O_*FY#ah{qg~_Oaub&(jNZwh5qj&BEF2+c0Ve)+Xt}#g%E9VOto6^UHy$Oi&d;F}x+prE|!Nqg|0P@+ts3CyYh5TVB^6l^9mGHPdOeC>nv@T_2#O80?{Zjq z)c2?Kyl0ok|BBM{|5GLGKmGjwW!)w09W(e?)kD@F;|+r|2Ixiv39M?nCz+yp;_qf$ zTPma^{W{h06On3%gHwz0(*r74z6CuimLX z9kdO+_$egMZ5Iad7s6(pGKCYbf8U05^TfKjbh*nHR?(?fJIrSOB`r&h5?eTg9&q(m zu#~>b3q^~2B0D6#v_or}n|>RZ6qY^9F=A*{+M$KH_ov03E7m4_d{1q(`F2Yk+h5-? zW;N&HyRw3I%>Lr#Owzlh=+8NAOWb`}OpOr>Al!C#J+!;jDTGlS92!Wap7Sn|yp`

    A5hflh)6TtEQ!@wD>@es1cU>4tv9&XH&u+9x7$5ynU^IS32~q zhNXsFI~m-e6(#4bqc^4%q6qcN-&%rm6<5;CZRW>Z1l+_a%QYdI1A;!bDb=hSZio=g zH%T0PUBW|py62*0Wj|=7czRXKF2>euAp8S8cxK-8U<$m6I)XN7ed*oDqRGHBQ6R;u zix|WfdmJD6@TV(73F~)KECO$w1}l-BQ>}X%Zc|n~`aY%hX~YnPW4yXPq_61=yjV8} zZYy_C1zdma!u<|x!^>h=0)RK6TLwj&mCE^Oo(pMSs%zHI7VV&Fp=QaHnxK|m&$&+; zjvonZ&G2%N%8M0+BI!4xJ3#XdmV_!R@vWeget42&5sp3ZWN~qU$GA%dTBBg=ZAWhL z;crdEQycPOKK3alaw;w5bw7TZAfAE2B-HsWyHwcjB4jM+ZS{F{nY?*;Gn2ej7A8^Z z3)b&iicRvoNv*E@m5yV;#A{P=;yZ2hqTwUpw#7&*eYL(SWF#2@L6l9c3*8T}qbE$0 z5?l92f)EP8lVf@Uv*5G}BRQU+(+Hnjqb%sWpk}X?;8N!gj-(|A2&%7~AluB|=QcGMI_ zEavsBQFsAYs`SddzxF*VXwFCIcvjf3He0i*Y(bJC>r8dh-3Dk=>}T9&MR5VPfs=nL zP};Qtt`@~;JCr1#_eEAhXmeIc50O*!j1rwKF_G=2 z5`2qCk{09@bYp_(-~CJ;ussUy)o8w40W}TZ;N^3D0fpN=Qdf|IaQFi@ftOQVh$Rv@ zjf&Q$`8Sz_3nt0b;Kg7Ra63o0x`B3V@A)u8ez7d3VXkZ`rV@qa6H3J~Yn^gXT&G#d z)sAwjauYP>pA4X`<~iiuj3Vf52qy$!S#n#(Bg-Ok@~CIQU0?ITCcgw}2nYRmS?X)U zSx++DJMJE;$IJ$b-r8!%0oBHnj`JN-)uEPOP)GV&moKy`9APTAE)1_XT8RIqQ&Pv`62@2r7wSc6uiu^iki8(nkuun=owEL zxPQCx#5_NDP>EB?)6pGk?itdVu~kt47(pTWeLZ^K&3ds)+QVOvd!AVn_Z|oYiZ?qx z{O_v%BMnh9MrdSz9JL)DhPI9PGo_c$=GSNh#^6CWCV1wp27HyT!0+^bGnQOQ`% zZ7TcW#D9jIoD;H7*@rj0Ky`ec9ixiZf~~6``jBLfEvpTkK3xu@5%*Xuj|2MVSZmxBDqPwo0>_vldJ*8XJ= zPQl&(-ZU_7alX|rA72$5{mfEVkuGgz#XbH-wv0E*WV|YAJG-?pI!=H9;)yR*daZ7oU+q&&J^};qFW$#V*EtLSBxFHm!*S)-O$uC;CL(t3Uznf6PXllb^ zSb&Q3ar;wC8T(bC-gDgj&qIo?sZ8^b3kx_1*Gxi9SiiPq1;g4}_&6zy)yqp3x8GX>buAaHxmU}IVchE_Dz zRbP=FX~jLHy>Xu_O4B!_qf{?{j&OYz>>1Cs{ME7|2L3!Ux(*1#$Nr|P^?;l5VAiNN zsroibD`luhwh#_q-uWI!OE1(=ZghRwGK1kL#Y)=7RDPbX$l`hb>KSdK{5y9T5JS5x zGm$<-Iyi;k@q@IbSoOnY5*l@_3%85Q?)fN(YQ)jpV4`D62Za-Y?QPETha4=CeiU+3CnbYBQ?p%JBG zToId_@tK1(d9Rv$!E$yi+OUGR@cQm%{;=6odC#1Ka0-`p2NSk+*pE?B074=_fyf|X zuLjl<9T~`qzx!91PK$|>sY|pp>PwZE#d1GNEzoNzCsi=tDhCc)`skqJN=uz2@;8pZ z&*H#*32E6FuaA1=Sv0q=%pyt@WRBINE`L*?+UaBK7e7Z*cqi@~FGY<#u1t*-9ef;7Bm#SshIU?L$l6bv%MX$dN>-z|DD?WO1p}0(p>=$ zCNxMXxnS8vp^O_g6L8qCBKZDNLRN0uBvO!Eei3S(k1j;%g6pl|X_=$3yKikQ|cRbKR}wAKqaXx%9(WN{o-7jEW~KsMN+Wm3qZ%(;Hr|WCX0_26L9L z8Orb>dz!cW42X|zDP1u$!G3v|6unUmiKandioi3~2|~CK+NR+>$9yaSY3Wu$b7g#k zjoR6o&+x63N>Fhoxb37lbcGyGHUF+;5BB@`V%DDfKhAyZB&e3SqT>A6R^w7JX)^rC z+jD7el55WhI13C$3~2h4n0gtsXuTiWp*ai*RMl6#2bGASqTe9>?4UwZ`_{!`fDU4W zTjhb)lM}1zdrpM>nAx91>A!mZBaJ`O*u}-bB;yXVQ-=F}-s4Zh0rWr38@lS)>$VU6 E53rwue*gdg diff --git a/docs/articles/intro_mashcommonbaseline.html b/docs/articles/intro_mashcommonbaseline.html index 0b53f40..ce34061 100644 --- a/docs/articles/intro_mashcommonbaseline.html +++ b/docs/articles/intro_mashcommonbaseline.html @@ -8,8 +8,8 @@ mashr with common baseline • mashr - - + + -

    +
    -
    +

    mashr with common baseline

    Yuxin Zou

    -

    2018-11-09

    - - Source: vignettes/intro_mashcommonbaseline.Rmd - - -
    +

    2019-03-11

    +
    +

    Introduction

    -

    We want to estimate the change in some quantity computed in multiple conditions over a common control condition. Deviation in any condition is defined as a difference in the quantity over a common control condition. We must consider the additional burden of comparing all subsequent conditions to the same reference condition. Failure to deal with such correlations induces many false positives in mashr analysis.

    +

    This vignette illustrates how to use mashr to estimate the change in some quantity measured in multiple conditions compared with a common control condition.

    +

    We assume that we have measurements in multiple conditions, and want to estimate the deviation in each condition from the control: that is, the difference in mean between that condition and the control condition. When we compare every condition to the same control then the observed deviations are correlated with one another (even under the null where there are no true differences among conditions). These correlations, if not properly accounted for, can lead to many false positives in a multivariate analysis. This vignette illustrates how to properly account for such correlations.

    Here is the write-up for the details of the model. When there is no control condition in the study, we can compare the quantity in different conditions with the mean. We illustrate an example in the no baseline vignette.

    To deal with these correlations, mashr allows the user to specify the reference condition using mash_update_data, after setting up the data in mash_set_data.

    -

    Note: In some settings measurements and tests in different conditions may be correlated with one another. For example, in eQTL applications this can occur due to sample overlap among the different conditions. In common baseline analysis, we assume the correlation is known.

    +

    Note: The correlations in deviations induced by comparing to a common baseline/control occur even if the measurements in different conditions are entirely independent. If the measurements in different conditions are also correlated with one another (eg in eQTL applications this can occur due to sample overlap among the different conditions) then this induces additional correlations into the analysis that should also be taken into account. In common baseline analysis, such additional correlations can be specified by the user (we have not yet implemented methods to estimate this additional correlation from the data).

    Illustration

    -

    Here we simulate data for illustration. This simulation routine creates a dataset with 8 conditions and 12000 samples, the last condition is the control condition. 90% of the samples have no deviations from the control condition. The rest 10% of the samples are ‘non-null’. The ‘non-null’ consists of equal numbers of three different types of deviations: equal among conditions \(1, \cdots, 7\), present only in condition 1, independent across conditions \(1, \cdots, 7\).

    -

    Our goal is to estimate the deviations in condition \(1, \cdots, 7\) over the control condition.

    +

    Here we simulate data for illustration. This simulation routine creates a dataset with 8 conditions and 12000 samples, the last condition is the control condition. 90% of the samples have no deviations from the control condition. The remaining 10% of the samples are “non-null”, and consist of equal numbers of three different types of deviations: equal among conditions \(1, \cdots, 7\), present only in condition 1, independent across conditions \(1, \cdots, 7\).

    +

    Our goal is to estimate the deviations in condition \(1, \cdots, 7\) compared with the control condition.

    library(mashr)
    # Loading required package: ashr
    set.seed(1)
@@ -102,16 +96,16 @@ 
    
 
    data = mash_set_data(simdata$Chat, simdata$Shat)
 
 data.L = mash_update_data(data, ref = 8)
    
-
    The updated mash data object (data.L) includes the induced correlation internally.
    
+
    The updated mash data object (data.L) includes the induced correlation internally.
    
 
    We proceed the analysis using just the simple canonical covariances as in the initial introductory vignette.
    
 
    U.c = cov_canonical(data.L)
 mashcontrast.model = mash(data.L, U.c, algorithm.version = 'R')
    
 
    #  - Computing 12000 x 181 likelihood matrix.
-#  - Likelihood calculations took 0.91 seconds.
+#  - Likelihood calculations took 0.97 seconds.
 #  - Fitting model with 181 mixture components.
-#  - Model fitting took 2.31 seconds.
+#  - Model fitting took 4.21 seconds.
 #  - Computing posterior matrices.
-#  - Computation allocated took 0.22 seconds.
    
+#  - Computation allocated took 0.14 seconds.
    print(get_loglik(mashcontrast.model),digits=10)
    # [1] -105525.1375

    Use get_significant_results to find the indices of effects that are ‘significant’:

    @@ -127,15 +121,16 @@

    data.wrong = mash_set_data(Bhat = simdata$Chat %*% t(L), Shat = 1) m = mash(data.wrong, U.c)

    #  - Computing 12000 x 181 likelihood matrix.
-#  - Likelihood calculations took 0.25 seconds.
+#  - Likelihood calculations took 0.29 seconds.
 #  - Fitting model with 181 mixture components.
-#  - Model fitting took 2.33 seconds.
+#  - Model fitting took 2.26 seconds.
 #  - Computing posterior matrices.
 #  - Computation allocated took 0.07 seconds.
    print(get_loglik(m),digits = 10)
    # [1] -111355.1971

    We can see that the log likelihood is lower, since it does not consider the induced correlation.

    There are 3358 significant effects, 2932 of them are false positives. The number of false positives is much more than the one include the induced correlation.

    +
    @@ -166,7 +161,5 @@

    - - diff --git a/docs/articles/intro_mashnobaseline.html b/docs/articles/intro_mashnobaseline.html index fa74f8d..c58dadf 100644 --- a/docs/articles/intro_mashnobaseline.html +++ b/docs/articles/intro_mashnobaseline.html @@ -8,8 +8,8 @@ mashr with no common baseline • mashr - - + + -
    +
    -
    +

    mashr with no common baseline

    Yuxin Zou

    -

    2018-11-09

    - - Source: vignettes/intro_mashnobaseline.Rmd - - -
    +

    2019-03-11

    +
    -
    library(knitr)
+
    
+
    library(MASS)
 library(kableExtra)
    
+
    # Warning: package 'kableExtra' was built under R version 3.4.4
    
 
    
 
    
 Introduction
    
@@ -103,10 +97,10 @@ 
    
 
   Beta = matrix(0, nrow=n, ncol=p)
   for(i in 1:n){
-    Beta[i,] = MASS::mvrnorm(1, rep(0, p), Utrue[[which_U[i]]])
+    Beta[i,] = mvrnorm(1, rep(0, p), Utrue[[which_U[i]]])
   }
   Shat = matrix(err_sd, nrow=n, ncol=p, byrow = TRUE)
-  E = MASS::mvrnorm(n, rep(0, p), Shat[1,]^2 * V)
+  E = mvrnorm(n, rep(0, p), Shat[1,]^2 * V)
   Bhat = Beta + E
   return(list(B = Beta, Bhat=Bhat, Shat = Shat, whichU = which_U))
 }
    @@ -148,18 +142,19 @@ 

    m = mash(data.L, c(U.c,U.ed), algorithm.version = 'R')
    #  - Computing 2000 x 181 likelihood matrix.
-#  - Likelihood calculations took 0.12 seconds.
+#  - Likelihood calculations took 0.36 seconds.
 #  - Fitting model with 181 mixture components.
-#  - Model fitting took 0.29 seconds.
+#  - Model fitting took 0.51 seconds.
 #  - Computing posterior matrices.
 #  - Computation allocated took 0.01 seconds.

    The log likelihood is

    print(get_loglik(m),digits=10)
    -
    # [1] -10891.72235
    +
    # [1] -10891.72215

    Use get_significant_results to find the indices of effects that are ‘significant’:

    length(get_significant_results(m))
    # [1] 141

    The number of false positive is 1.

    +

    @@ -188,7 +183,5 @@

    - - diff --git a/docs/articles/mash_sampling.html b/docs/articles/mash_sampling.html index bfadd29..6fe4ffd 100644 --- a/docs/articles/mash_sampling.html +++ b/docs/articles/mash_sampling.html @@ -8,8 +8,8 @@ Sample from mash posteriors • mashr - - + + -
    +
    -
    +

    Sample from mash posteriors

    Yuxin Zou

    -

    2018-11-08

    - - Source: vignettes/mash_sampling.Rmd - - -
    +

    2019-03-11

    +
    +

    Introduction

    @@ -95,11 +88,11 @@

    Here, we draw 100 samples from the posteriors of each effect.

    m = mash(data, U.c, algorithm.version = 'R', posterior_samples = 100)
    #  - Computing 2000 x 151 likelihood matrix.
-#  - Likelihood calculations took 0.10 seconds.
+#  - Likelihood calculations took 0.27 seconds.
 #  - Fitting model with 151 mixture components.
-#  - Model fitting took 0.30 seconds.
+#  - Model fitting took 0.43 seconds.
 #  - Computing posterior matrices.
-#  - Computation allocated took 3.33 seconds.
    +# - Computation allocated took 3.11 seconds.

    Using get_samples(m), we have a \(2000 \times 5 \times 100\) array for samples.

    If we fit the mash model without the posterior samples, we could use mash_compute_posterior_matrices to sample from the mash object.

    m$result = mash_compute_posterior_matrices(m, data, algorithm.version = 'R',
@@ -130,6 +123,7 @@ 
    
 
    x = get_pairwise_sharing(m, factor=0.5)
 corrplot(x, method='color', cl.lim=c(0,1), type='upper', addCoef.col = "black", tl.col="black", tl.srt=45, title = 'Pairwise Sharing by Magnitude', mar = c(4,0,4,0))
    
 
    
+

    @@ -159,7 +153,5 @@

    - - diff --git a/docs/articles/mash_sampling_files/figure-html/unnamed-chunk-5-1.png b/docs/articles/mash_sampling_files/figure-html/unnamed-chunk-5-1.png index f9e21ae6bc7a6890442994cc1c23075f49bea648..e8a667cf2ff1193d93a83d33fc388f179ac9d252 100644 GIT binary patch delta 85586 zcmb4qcQl-D*Dj)k2%<%oAd+a&qf60)L=e41FQfPN5D6(5jNX%|(HWhYNHIk3%xKY> z(R-cq_}F8LDvN z(7&@^o0XpK#1V8bVGoeWN5+spTFo z*uiX{X~nU7x{1t?SFCrLcwMfzKm1bEp`UB>Sb*y}BO!N#&@zze6??YzlloE3Kz64v z;^>QCj>2GhMNgyGa$vpB+OT>wSE2XjwAsb+GH05o?Z;<13R9CaO+L=M5*B5~5}Vuq zJPAEsz2f28LgHXOFQ^@0%a9u*Si-f;9i~)GPD7xTEZgIAJ51?$_*J?)9RbT(_8ZOv zU*1i+|D+oH3_Kw-s{%r3Ga2HtZ#U-^o}9s0l9^M4|LDh%#-E4zKW1rR1Fd9jpeC`ky~aO@#TM9!0@ zRPF7vC_}46KnM_qJeyi+byV1KzINRL?t?YJKlxGaHeUI;gItMWq~uj!_s^r{pJ!!I z<{bAQSC!izKC|kx=OfBzcElp?%7EK7r`~J%rSM4p5XiX-xj7&L_}3I{kIrc zq+A7LeQ^)?qz277GVkpFr*TRUdHd%f@h1TDTg)$I55SUdCK3l%5_Bo><%A5S|V&GuGF z003>b{;bBaeIqv}CPjSl7x+F&gM@_$0;0~_aR~xUe?aIE5Fv=ryO86Z#icu0d%OvG z^LGzc27YEfP`RSfr2Ux7oJcc8XrGZ`)!7#}US;PtU;&)#x5uz9%jb{~XENr-_!y_2 zSqp+bcYd+sr5h$Ws&2NULJ>Xrz9l)L1ANSt6r#`)oty4rZbZbaQ@Y1lkqO{ ztM8@PxwHg*I3xa9_*^6Y2Otpc<+t=*WQ^H~;w= zGw2ToB4Q=F^;l1N5H3~U9SgITym3Q73f;P$3 z-UHWv;}K}f?MdZwLPFv%H*P8fH-Pb(x*WRG$tUW$&~anesMaZIB^sT6J4*ub>zOn` zI61M?$T!|6k=I0^!()~+h!ETTwGrNI$9-4Vk#9+WRG9IcI$N01wM!s`sWE!cfk0rD zAo;BZ8mtm$0|G&vTZus2(N%)!BLUF`(TUaabkOF$D~E(rVl=Z@dozbin;)C9vs}L= zs)~w`kJzuH6|>d+pO6S^bRcRAw^cLBMVq zG9$O!Sv_Y$4%14mkY0<6q+&I0K8jH?C5EG_4-Co;%F>Odg^_2s$vA^6yHiEDbAzcu zvx7DotF3;L7VGAJbkLy%*TD#Y3#os&n4}DbnLuEs2?rQ}i=VaA$hl?>2%+2I$c0Cn zlYu_NAdG0LgRf=Nq&C;1wC~6)qVMU0>h2 zLw`n^_l4ha6E!`mvt=F*@*Ge$pWzpS6&8xc=B-Tv#&EUl;2m{f0`3SiDg%?N`pMxY zXWxBYMj~?%xxyrrLbPDubjEy1FCc&j2on?2C~=7r66MBAd(eZs6dp#Zw4brp($W(5 z3e)W87I3_nvU<*}`<{LKM*r+G36m+8Mew$A$L$h>vinfQi-&D}agTBTiKLQJ&`w*1 zg)_jP62zKG<_?Y0cL-6jA9E)psQ>|)aF~D~8V^h` z=_Hx~@U7h$Qaa@ga}?@YT7t*Ku}ZmY#N=VgS7ekg6B=mYKOeC3E;7KCssDqNtlx%< zc1y8Nzpbrp<9R|6@P^_6W$>G6Lm@EiXu!khzkaxDC`DPx9529nW(!~1#t9Qu-QXf% z;aA^z^xwaC`LhHi%h6IJ&yHJB9g~e-uvw?ZUr%Yc2-3;@;+_9_EUb>~Iy)QVcpW6; z8r#GEOo!b*MLhkIhWpJ-ob??r_UvHp5)3>%<=^vD%jsf4&B|nHWae;X zRp|u6Z`=Wrnhr%=}-l`iVz+05KrE&eBITnO9CjN5If1i zBKxfk=F^2G$s~y>X=(S)PY;=>EhLG@D{nQDkcZtlJYQZjlbNzJW0_*gJo5BkkG#ahjVIq>$-Isu#t#EUNQ#;52*e8m zAC`;K zC)Z`U6(6doy7plhzGilihW}y$p?Y+f%KzUNco`66Boq`B?A`MIqwn4+xA|2v5E2s+ z4-(Z?62i8_=!IT~vy6Ph$&PZ*c{Ej|y@-tW{ARbwUvn194#tA6-czo+vkd}X)%zfJ zb??Q(w(ICPG-$T`KVC`XWn(;dE!>47>@q(ny<>1casEIN^j8noM#4mK!vZ`2=Vq+S zX<#op87T4?z~#Z~vnf?!FG3iuZeRNBLVk%N#Y%&a!9dLz+zRy|Y@_>s4c>o(&b>s* z>Q+rf^ZxXU%_B+>S4e5z|C$*7$qq_HS4)kX?`^!Q@xjD4q^O6JH|TTqz}o7W43%k$ z%=l4-EF3wu%ZxQO_oY*$rMcrQFG~Sjq^I2Gq;5 z2X1!l2vK&ym~L9NYlIx(I?<;c`moP%23SZa*VPZVz>N{v;C_`ts7XiRDSXiysc`!Exd!AWC}wU~b``Cn;Lk2pMX-)W;HQkCcTUxeK~ z6=s!h=CwNr(c^anM7&pN2n5yNpOeExC2w%NCE|eA5I%>}FK&eTfdCQIXh6cEX$0-kp2^sFp3C1Pp zvh59z+h}~Op~eIrkSGley_^q8ON)Pt|Pbu_I2>t8h{@d?$?9HGPrjCQ=|Kq#i;`IiZfv|x(x)%(Oa$U zl$OYU8HK6NUSb*cSBmi}jUu?5?V19fOG_wd-q(PjdKmGN?b20Of9ZhFJuAI;_0ndO zIRP+|dS~gQ{vFwg`ng4=%OA0gKDbJzfHm&_7AsXuhDb*foS5J(R2jxM*`2tV zR_Qx<+&ignvBF+R(pf&v^<3<}+j}2#FOAvJwkxcLR_#>Fh?{^wi9r47BfZ;%7sd%I zc3#E*+jGI&LQptPgZXz_pxm6ghve{_y1GQypXxu+=ZkUNL|6HgsQceG=bO@8nnY#@ zI#uv$Xvzy>LMDnIV9aPWnEWBRm9qVekbyi5EC7ZRW~Kg#FP=MkpXxDReB&WT*aqkQ?~QJBzT*3 zewYP0*YgB$2?=kK!5x?_CIYaj1_g07aglrpI z_Pk#gbIXQ@q*A?ty?g@RbywD-W!8{*X>GL^4 z-9jtajorTlhZF)Pa3#dag1ph|7W0(zDy46uMYjZHeUl{3nz}xO2q%lr`!JNye;UIz zCE+{xD&&RvlLaI#Y638x`Q|XYb{#M26<~#KVMXjDecm1Qt#{#N74h6eyVL3UYKHKi z1QKNGI&M@8+jCI)d7yo52GzN2A;g??{~gJf7^*H9y=)50`dSJsMsn1jebrMtCf$w| zGwM&ZuOM{*!M%E^u!RurKQOVVpvtHhoneB<%!Y*5RdG79o{ruA8e?`}whkhJ0me^M zScA4i`2L8%tJJrf7H`w4*%0c)$KN<6YLtw>LM``vsW*!sBq*!x`?BAwn}_V{H{z)m27}>vq2wb(+cOB)1%f;d(GEXWrHYs7Tc; zJ}PMAgQE)`47xTzOM3gein`Z`i4-Ib<;HTC5`P%d@2st*nWE{|(sk`Vi9FYjm@Y=a zL?)FfhtO0Czds`O2T7k?8=idAE8*CC(6ijw9N9TI15h?uNEO)I{7Fjzs1gCZy+eof z_+RsIBTk(}rp;r;2^dm#MH#-~nLD}FzpGp0dY*gAJ$18W3@6(sC%}n3$;3^;IIA6~ zrZ1MuRl5m7J0iyaYJ~gOR(;&<6{GH4kZRYsC^x%R<|O4{17GD{oV5-D$y`avIA#0a z+7nXFKC{!mFrq_8Z#)2;e0`*gHRdJ+R~cpTUY+$6*YTGZA^h;5+iNl5b3N-Or4#Y* zY)g0*Ct}Ct0|Y0*%^6=%h9T?Wrj@e0DOP&8I{08tvl$z1Z@VBkre-|SuG_e7M4Q9Z z6Mme**x8FL)SOx(>p&xyRba~fUUjZhPVLtX%1!T?HF?KtX8|(m8X6M?u>-M&{}9u3|t7dT%Th-=f4w@3X^v}qg1 z@+$g0sT8<9UAKfev>jlrkGdY}q^vk4m)$XvQiqcH3>YpE z7;jcJZ`zdHjqriL%4!aP$%*FyA2>b5f%S>w1-uy+QW)Gd*=XG?C%X5i5`35W73(@^ZmTMT{t{`hpXi-ea z3MqQrT}l9Od(iq0!VF7PV`y z3M5~n*u0{R@-3H7NA>M8UaW}5LTTLe$)%Sxd8oaMnf z$ZmI_dV}^>tfrfndNNE3QD=US9)(Y27+^h?v;6)5w5n`Y$$x{~vQ|Visr%3CJr9>8 zb}%*gZOYKVJPlZEbx;?}gtgqajSMGm8DrxtubRz*_bFZ}-SNt9XuQhHLD77Tp;#3q z@iBkgg|LZvQw5f)Len_^Pu+TFUBWYYB^0x~oPrcb$f4GV(})PC><~3dSz-Q4VP3(X zb2lv@=35l{e9T;67f}VM7cM$f)B}vdi80PZ$Y)KzR(t|KZWk4}1s-2uY)@nla_xMR z*rXSZUfPsxtF;2cqu%owC#s66ytQkp(^r1f%O&$ChbhcGJhMPy6jHG%zL4hqxczno zeAu!TU)HU$sY7o)CfZM9rKIVR*&(H62xEYG{WH20qI&#ySH~8UtW|fizu{NBhYWy2 zSot>?-oIi!MNM+rE+mCOAaxfI_}Zk_eOvux5=DUjU3-wyOrHEB&eR8H)g^lSt_3ms zy5o#;?Z%>e1S66MkzTPiwMHj0PWJGsraokh{8^ruZO8dQZPnrRCc~vrv7sf?DPTFI zEs|@KQb;UNC?~M(;*bR0ZVrr!)SWm*qMwYoSlb8RXK~oq;yY}9CWPdbMvo635QHog zr`R6NcJT@pe$aFEs}nZqPT7@WkylE2mc+;&^e5qXCP^AEyV3Au199r=b(fh++iA8*<^i7UC?lKh2sUf#=Sw!p4mJ% z9R!%h#&M1gMx22wpY6DwBY6ilCRt2J%I5+zeX$oa_zU~ch}Z0!_|pO)!!?0g;pls2 z*h(HuTeNTB`?rJdrmcas#|l-DQ5=CClciIGeF~>8(x5EJDdYpg4gJCW~OJ4TYT!?w*rXjh~V6A>^Y&bsFjRS%;Lgx+8((CDt@aNIJ8B>(T4^h*m2HGboVdZMGRt%36ymIyf0NuLr-L zFi~TlVT+S!pE=z|arR`l%z%l%Qs8y9R#~lT2_jT&k-PFB3hL6VAB)^V`ty z(Wy`(Lr&{mXJpPo+2qo(#h;CNAz1g?ehcQdnW5yq5#}>9FtVnRL8+o{uzqLhpbR({ z_=K8Sl`nALbe(8AQY{`?G~ASAyzxpY=uU{pDg;Bqfv=#9p(gcXo$o71C;W{OQh4IK z;3@$fpzh0Qxp;F;oZ}Y17P=)AqO|q$N^*pXqHFAvLVUU4*|F@`sFlTPezzzotgwPV zKNo?_rb!SDupjmln)Dgk!o{4H)V&L5nu0K8ZS+|8ZA{q-7jkP$cQwb598x?B!P;WT zrb_?l+YFb$@;tigmdG-tI9p8$TBLh}!e_Kk4SP9*HH%RZF(b|f$fsMtGkk9qUH;EU zzwx9NzC(8d=0+dw%tseb%i!k@mBanWD9h=<;3@$gSZIaLFrE#I(7Od4XgSWb9@Q2= z|04H)h1gLCQR+shB1050XGEQ76;@|;p5@JZ^A^>$M(FnG-L&fg|occ)h$?8BWT79&(g0-j6jRha=DK>-D1)Iu8v6@&ukS4O$gLO*?mJ^&f|^PNj= zyzO~wFGyskhd;-rOSw)uga&=ao}h|rLb1nz*$U_~pyfm`^59}P4+W}<`%9ihct_Cd z`RG}Pem!iSJa3Aii5;n3*)Ck_GcZ^Uzn<>t>Dk?cINfNr;_5^qcZKqcc`@2xL37#&r#8! zCeXyYTv9RLLurq>UOEM=KNUMFW|fy=EEKt3h;rNE#(TLizcrjdy#9=vZWyeqe^H^? z(xx-@qHN?*!c=Z;z%z|NUhCU=ciOoRfB!TY*S6ukMH2cTQ03@=Ttfo=*{g!tl~i&6 zjgMzC-Gxv5IvS9?^8jt~aXWQ?fkbfvIXobuER5Xe*yLeD3;DrwCoUzW(ByyngV|j8!|uzI^QBN5S0nK(_^ulMKYU!u~K^B6axrZpB6-2E`H19!=>vq!XH*r*`e?}wYd+sUcHHU-n9#ywb$|Y_*=Ly;-{QR6@8)*!Yei{!ViCKhO~V| zGGV5fRu-Niir)>iU%sXsf>U4Kz&|U(c)#HR(1uBBnRz-{PJ_LAC-73Y!Rt25VBJYs2jefI-zppKH&-jddIZ^3?4r zp5y0Bea=;%qiR8o{A+FJSs9@L)ipaCn^Y`Xb5PO0_>t1ORk3=+hc-LwG1dKj)}$RZ zbX*LZ^q_!obE9*ZF6%@5UIlR8buD3Nb3K^+s$)+@m<>wVD3k{N_4BRvg(C|uuc7V+yeof&zb?9=WylW1K0ad^c&Z}X0qQP5l8 z=56-X(5|4tfLX7>+4l%SROKxZlF;4L@1@~+nJfqD^-b&(UB25gCEt2)SfuAcd-N;J zUdgp1S<_i+9c)8y9`-#60*>xb^*RaD@;l~=buGWpnkpCjb;3tAnQ%h~upAjg5^*4L z!}AX?@7BJyK$_WIFbCK}i-}QgqYq#*jnGD@+e;kiw3%0o&V0mIl{RZ~nL}^~lY%TF z3EibG!f|d}LukVa53$`UOJ_5{&9DalNf&Ksk+xuR9o^0m+(7jhGww?WDLp?gSbs~y zk2MgKnqgX5R@b6>f&hdcjwDCmPv1C6tZy2*%Og+Zi#rSBuAWZhv4K=#m+u~q>^5H!pn~F z{2oRq`uZ0+W_|;_SrrFpjC|AYAuui4mGCc~*E8;vdhZ0O0YVSgMl|8}HkEm8di)Q* ziPdB4`3wey<_);>XNpWi_D4p`!)90n;1^cc##d)av&K8qJdE82_VYBZ>CB$v` z%h5fWDK%E-uj)-ni*4v}u%%6!%X5?h3opjN6Q6=krM>^$Gz7N-vV-+v_y>`pZn!ByA zZ`kLNS~=w-!-$X|yxFvm^BsToCg>#u2IpR=1*BAkoceBZ4^Xj)(`1DU=nWDSEA(z; zc-Q%|v+BWOPnmJ7*JkE@Zt(IO0Y{@vz~<&?*&)(uXUbL3*?mX@QImnKzcRlO+KBVN~ek!T7j;Zy5G=`M>$Nr0*NUAM`^Q(<#OaA zAye+W=LsJkA|s;v=t5gvqMb5EeBHEu*w?{s|D=7sH^}C^FXzv}BBjk(B#@^Il-)^~ zs*Dj?TJ}hI;e($kU4LqDLpQfSb<$zK8VMO=bos)S6{o~$G?q47mIp0Rg?XC}u=qR% zJ}=TDA~CMh3eOxudPQ9$847mAftbE$5yt7cm0E? zZ)|Ean(|gUInH9slSMv!0(Jm^9h=84Q|FS88`Rk_7#(vNl3ec@n{8tFZSi_t@=cD; zbYR(@BuA&qWOYoi{Qi^UJ*`tRyqIuvJ3hhC@3T?>yy{P0#jd4=-Kqk6LK|>1era81 zRmklNco*TBdR!@K!1rfIrxuH;m!4=3J)EF7*}mAMOs`wzLe~VWBP0O>q-Ee)@*~d# zk+e*OkuMexj$@6!rR}fkusT2SIsWLj{1sihu487HQUI1XRgIXNOy=_xGEA!K>U>6{ z$6DfyboU)Im4}*qu#ee)_!7UqA4gC496R8XG<5ZoMn>uP%4GTNJG@|z$lKtE^gy5!VeNQw!{#PzV0{vy zNYM0zvp(8m@WaDU(RQYNt*O|R#bzfVP9L;%^-)Srk5Zv)KQr9tHTC(n4y1{L=Lu)% zG~1PKfD^kxH;$0l@?pYBu)1VuFED5kh-d18y;y&tv9N(x`@6jd7m8n*iY|@{=c+t5 zlg;h-cV~V#Q353Qjaf}B<-+$*xWBq2Oz3;&)ry$ZUHWfCt=u*X-la(5tg|m44xl%b zSWio~5wobY1JzIGQHukClY8TzM0dVjZ&IlIdMxsGk-@W%*>|g+{wxSfc8B zz|Khd{<5qOKYZAB%F6fiim!{5T!UMs(ZUTPJS|driL3F-B>O{`!6L zwuDsUrw@)Ii3~$~H%3WUj>s*fCig)M%LLNn6LFW6|9GA{Oii1X|3NVwjU$)1Vf7vS z)jHux$rjDm}|u|J-C&n9asqMJLN+g0`Uumhd>X=pq@|D6~*Ek zZ5f6n590q`ahVvFoMzg)x`R-{jH9A&aiKYA5HYO>*{H=MFQLlC&W~o6J)}X)z}3^~ z@p6U*2Gq?_K?|OfU-_FlhZ|ErdrSVB&^bTT2}u{7w1;3FHJ#ts&d9wHG=V+A_J?e? z>}r3%1=5|LS1JPxD8^n6%S6tQC+c%OaC$z<#y@+EfR>s7v-!mZ`>66|Luhu2H^Qx7 z46S5eBq~nhkgvA(b*AW-tdZR)P%q7!w25Wd$S6$Mu>$5O={F`I&1yQvmYX{AT)*8I z{il|Ia#zYE3g6{-XkM9BT^Nj1E#hl04F2bUTA`~DX}QNnh0A9V74vN{;F`Hf)~R^h zhjUuj*`>U@YTZ($ffu+i+u(4|Ua#`$Ii36L$3hoK%(?T|*;=mR6-NuejmxkO8TQ%> z3vIW)(0r}0zjMz%e5x5|!Syb)0!nxmor`!d1=(NhMu2H|HOk`&zqh_@1D!wf5OiB zY2B&7Y+9n&2yCVKVTZ*UgKeCZsA{oB61n07yON?xuDfb4Ud;O9y&|$3+`b+NP$fID zyT1`v1gzg5YAI)J6n4B}A`Rk9naQZvHwZaS7MCE(%lpNsCU&QFSLbB&S!G^L`ALB# z7%EV2BYPO~FrD%@eqf1PN8Thb+uf<(CG_a*Wx!e;*5%YY^SAF*U`H&ZQPlq6n%e`P zZDBmDxk5o_BXBjOc&M^!oy}m9Zx-Lgf0z6MAaS&cTX%dED4@*!91Rk4Z5%9~jc3PL zHaCy7r0P8EUzA!;(|*r7&?B7fXaEg4?Iep<`k_6L9l+y9d4x3Icbc1OU!&2pR-F+L zQ_QsdIXp3k7u$&s6G`6ZLL0K`>y{lM?auM*^$*@ap!%%MzrcIa;de@+X(l$?7GdC_ zc*p6R;Q3EgWRZI*ze6E7LQuu}?qZG4R2(a9s;ipMX|dSrf``JZM%Urzz0c|u3jKM8 z?XP04+8U~>^K)o?ez6-X8nClajXu#Z0y~{XFp4O@$AmLMKWo~2D|^~fuBDw8b-oZIrv^d{m1d{US?OG^g&w& z>Ndv}mjCGc$*qLY2mV}35(UuYfyS`jm4$2boXA(@z$`N+ptM0p=}HvBv10q3TV*0r zw$jflABVGyN_H{ws>HPhv zU$^-XErQ4hI$n-8_W((zKD!jx;K-W_c9};yp?PEj?dYNe*WUm#W(wv8ICTyh@@i#`&3T5qtBk5nm12z_o<@Tm(}XFmbS zk_=@-9q36ifD$FrCg=Ihy+z>%wn3(b1_T7`3~CC$gAbougdXQhH&h_D2iihwqE*SM zAI=i&vaw@!S`Tqiw9YA}?xL+=I<>kXM>vr8#ijGwLTO_KgC@@~Fxs-}4YZzDnFXXKvHo%`yI| z*xhZG>@+dDL-9KqWwKMHSgS@%*f)LG;psZaSUW>I6WuAFP|dmxuNCo|5W`L>ge7J2 znIf@mk_Z9I{kY-z2AJ&%xA5Cc_)3N&CrR)F8IaTfkyO^)YZ&x54&##Yxu(l)qO!Lx zh5pUFNWLuYDpRGt^5>yQTc3tQZ+|Y=**L zeT)qM2Ho90Ndv_za=_SD*lgj!S<7PQSe{oOcf^~)c2ZmQ^q7l^iRTcP(75jo2EG%R zam%*8Cm`qd6BSK$jIccUHRf`t3O#hQ2Kpq?<4a)J5BoV-L)?YTtB7&~uOhr+j?5FT za~BcnYJsxQ?v(yqwJ%RIC>-@n=L{9N~ zpo&>{US)jQ#wb;Sac$NbEu%cM{5CT<>49ROp4Vu8-neri^6bqbrw~#h%Qa-KLp*KpR1E~Sc}svVH>?+DTxJKsBtd3a z9$qPp&066kD8n?Qt+x!83K(r^_Gjm)veYzF2h1AHKhaj{z$IU}Ze~q>dBRkTDRTV2 z&E9kfhYV4B@Z&FrKt~^59d_)0m^;{JUV}L^N6DN9X~!57E(0n}x6orQ8%tW%j_Xb) zB5jer{nW44uTLt*re^2Tg~d$@J~xMyGMt_5!2BR#NhJm_Om>i_3`YxIGZaJEP61LC zYMy0dj^jMDZz5`|Fz0rkq%h;@w$9cOB#Kb52D=ur)fl%)36Ggt8EfYV{NYGAmJ4w? zz4A$x&PX-|z^k)58#SEQ_onDpm~StaI1Uw%w6bM&sKMH$z)KAynA>7lCWkwrdmqrx zSJ&%5p$tG_Vb4&tzoo45FTJe`mO`>>b%IRY!urtWPdJ3pxvQ=ibRCpf3 z;6x|YrduN?nLVI#q{7g^jHf@hu{Fk`CfHGFee7)4?|WUlAcolGlg&>wL)RNp#h}Ax zy>jM#hyLxA;N716)Bb$xyQG@H$@@Y84T{V7JELZI`YQuU!dS?Hddki1fF4Pw0h!qr zHL1i!iDroAER(v#;}$y!!=()0_3^*Wf2uV}z@z{E5EdGJ13znDYwWYo>p_>Vm2%~$ z#$bEoqWV!*d|g$^h0*I8hsoPbhs5b0>Z;n_$zGGTOHVeVjfc4zz&1LIOZ{!(0!zBroinM@J35VFtmNnSZ(K6!rz zQnEteI8o9|ZE59`gLP+)uQSt0cO2sy9EawzA7Sd!|b*Hr~7&#?S_3rbYHnBV$k=1(^KFCYV*a~c}(D&DKS2Nz524r0`Kl)bNn>%hEy! zI9cQAIC6iB(s-uDP=(8Y=zvkipwXZ|SUk6Q9thoPBl{x%xGXGJBKzG%xH$j{H+E2s z8TH`Ou7Qoem*Z8^?<+3ijhC@DPt(1-pHZ!WM2Y4Gc;Ul}qe7R#iNnNuFQ1~=MsmyU zgt({zd&ae@SnxW|Bm3tAC7gEHvhYBtHj=tulu^jE2zgrrUFmqmZ=-QJJzPh=b7#<& zip#muFeozFSB_RbCovcxax~J<*PuH!bYor8u8+CWv|oY$0J!zAu;ZWxI1xK%3p~=X zjuVesa`u(EL*W10URdLE-!DxW%5yZQVl{h;etjb(E9n=*!=v%(x=%wB<326V_*YY2 ziz(aQL`4)L> z!Y7k1zjy^BX?A>6yX=AyjTWZBN(x;(?yieqcy(2;*uV_!hqWr%?N+CAd30)l$l_%+ zsHkKg66@fO56s+jJ{a3XR%d9vZY_tM-|({Qt4x}K`*gP5 z$6PJdQ@g|t;2gto)(366_x&$;H`UO$CV&4NQtGDj5Is(S$U{HRH;xHY{k<?#nI*S_EOH^G`0;F*d8kaaq~sf6 z;)J!O&F1m5RuT^+ZYt=bHLdME!+AEAqv4~-c9A_m0TD1aURGWx(@rrQ#1&{U4dlEF6n!Uwlid^%c0tlP(Xtw{*^NaWf- zHFk6o-S?c9xybWv9^F=wn4JL2P~8~C;$>%dogfQIr{NAcq;{6fqGzW}&H35!?&2mA zpA0Oue_fDqg!%G8^#{-PF{}`_+Obl6eMWrZS7C{=oRN`MqZ+%xDr;2H&d0o#r?Lgh z@9*iR!hy1b=0>ogsCtk4R54jU!zUg+rA;5g7J9A&ixE*e?29gw7nP-EaX$Nsz{_&3 zcQ(3D(wMZfR|M)-W9#7%lbwr*YR3x!08{se3Zsz~q0OM<)KrQOsCKc~IMqU^%9x6% z?4{ahn`Mk8(otVyE9V+`Cf@K6muWtCdeg%o>DyrQChroisHU(*80;NEhLXmzaI5BA zxUs&AjrYeW7h>wB@CSOEzl5*`7u-)y*aVf^nN0Jw4qNw&{p#k{8%=CFm{li$z)yKB zcmoF-eS7HGwuT~bNHy?5F8F0dO)relPhq9`14!Jjd~NNUmL4`Zy-vn{zx8BK%mS6e zO!wMlm-+r9R7r^UU}+do!r6jAb|0^Kxhb#|vDkG-L-GwB}1|cWM;U4W05MVpj!! zatem754A@w?cQwGjh@x`QfS19yh=V+Ufp%_P-dvRHl}r>`|A&Fap#h^eSad_Pbtrd^oM|7%k3S%xF9$r z*|qp4D%MX2mh5iusuPm1q&<(LD0mDc`3T~o;`DKEb+Er_?*>54x&2m7bS+T5%kjRx z?1A9+Q8il}@~_{-_Oea!KwgQ<3E!4uti#Uet$n!E$~XOzmO?BngN42}l5THfygv6Q z@fzD*mf7Kk_rTzF2N}S=GhM>Dr5Vnh(tE|TORzqby}m`(Q0co>__%ThTX_-P1U7m; ziZj%I*a&-kBB)mB_V`P$_#ZkSr8#g4Dns#w9qoa54^lH_P_cS7qd4?yp7j?Q388Op zMgn|wN@eA!v~UzV{~`5v1Z1M(i((tdUzOL9C`Pc;IhnwwUjgbHgBCWMu|3mI_To=1 zp;K5wpP(A1&26c@@^+011Y{7qc*C z@&5kNYK~)4=9=f!xlSk8%rwVd(e&N_<6pD+D{&XZV6i%`EH%mD!h|usn_HonWpv(W zvFz`R+*;mz3IK2NHDkv7#4Q5bK`E$)TKH(y#N>RBG9t>Y;kKbq4`kWgiyItJl-WSj zR!I5|HAy@Q`k7+zNwN<9`h(xuKp|e!+iw520t83&HDn~mvKX(${ng1@F?9W1(VLtL zpNFU;B5%Y%`^l&dW}Ga_bOR{zYn_q!3A=r?+B&NX2#6PDybOkuo`%SnC*J2xLIdRK^cDE*pV$YV*MU#GjEK~u>UE((ZK}vn!m3KBOSt+J}~w)W~Z05&+2Sly>}%)FyqM&4klOo zCXt^(D1e{y)7_$C=!BHa#-RH|l`tU*e0hP2T3$HKXu_(Z+-KR<^XE@t|0tcSL6tMT z(ly8}%#CxWv7fQ=lrdoi649CsnNGvSd=Z;L8r3BmK~Fe0EDstVD{Y>Z>9h)wv4y&f zpE&i+Uer$tPC;c9a2}t1FZX%>=FV)5uHuhrAX7DG;FWgk)}QAiubU}J<_6j?4t}}} z2uHk*za0NwnU;gcw63p8DoGaU6>8dc7!GvG^!|>-Dj>r)Hg4L76(1)BbVe7DV7TDSS?66qXFExFr+w$AL=)TNECcS!n zt$%B$xxO6MqW_6kP!U%HV9^W zt>3cVz#d_KavD4V4~-Fb9%{oiGOD|ESqA*JHp$-ZKxki7E#)A21-V~>s=wfVhB#Bm zUSc8}5m=qjhb&Zm6v5QSQl>zntxaCQwD%f3dNjNQS_QFV>Ohu?)>C`#(W}{No#042 zw)eC&WH`YJe50Q#W*iAJOWy{xCER8jf7Dl9_JY_KyG&GSucbp=Sii$<%vh{*)jRtuTl1K39Q!B5s32Xg;|v$u|ms$2iR1rbmbMF9aRrKF`hEV@&2 zP?{m6B{zsk3PVT@C@l=#ZO}2&9ZGi(F~o0;=RW6mzUMyA^L<{=ANo>;y=U*euC?~M z;`6?GfrI=uItj>_+q|YGdX}8~5$;Uq60a?yd)lU-Z&S%{Pg)fH;&} zh?)&2%)As3U<4F^)pKf^h{)IEm%;LXw+!5f7Qz`HcP#8*$No(%8+eJm&)M>nVC-m=sr6vIx_gjYY0Y=V?cmzr;)0;5MoaJv{abmhl z0kSH_x%6zA^ZP5^54Jgk8v`N5C}VVgz_~|3bTJ>u3YWi(J+}NKhP{}IyCdbarlLQCG|d65`(H;s z{nwFyyFKEAi}q9T8x>851GN)ANxZMFUsJe-n;9-lRhR!ZGww139!p(t=6&IGUvs=8 zU6x>+TmMCXY7R@ec?~6ZfcrVjd{wu(GuMb2*9y&F5pzU`qTtp z^ieu*O420Cw5$x>*jy;_74J5}l&?>^@lgN$WNaVek%md}Ellj}*jV$pI&L;~+)a00 zejqMPrFV^(38*sfiLO4tnc_uHur|2wpQK^_AxV`3*GNU{@efhvrGOeRr>e21nm)K^ z095)ojLK!w8ZgBu5#F`8D56UmCtrT995O>Azx@w!x!U7vPa0Tw*9Th1%Gfe30ck`R zrw`-Q&yNOz2?db$WM;{*KbgNoLW0x!6+i#=fc@^>52SSB3IIRE!X5TpN5zRgKIU#C zXa1u*q(ByY1A0Irn5}Dhrlvmf@(ottQrRq@=d4cKWp~+}3YmlHAQuq1Q*d<5-ya9G zm^M1k@!Q|uT-kBt?E2<8<;QlOh*U7>OZ`F_i1{JC=Bbwka^p6vV#%KCfTc36)*}3G zHrU2AO{PxmqZ(^52c|+mrsa!ZD9+7S5i8W{1;j%kEv*}+eQF3$ZPlJ6=4x@YyTtE% z@O-4wIc*$hnOH2%#|jo0Ts!%qQc`%i4NE_`c8kda)#MzY{>0*>OCaIwG!K2d$o3~T zkOh`KaTeuz+<-s5K>1UCQs^LO3Gj~)Aw1QsMq@ZdCXSr}zBwhy4>Qu6CVzFKK8lhD zgs}t;y!tie+xl~~StKOJBrW^1|3F#}r-QC`)`Cp49ig0Zd&CJSJZN04VffFts~4S`Ysw3JwV9AP5c)?wpJ_Tnpi?7qwXvBK~6g?bXH424j4e+9v%392P0j0^vgeG$p8GQ4rAcs z5X$$M!G3Z(E;YJb?LIIF7L&DJ^Dc^K@V-p2^Be4xvp z{{|g{jy6CLWdI|a^y6ZjPCtiW6w6VbB>th1diRZ2Bn7@2$>NIpQ9*7->IeP&zQI|B zTY)nY1BlSl-||Jz5KL8naa#fsFJQj-;Ovg zZ2+iI+Y{t2Vn9VP9v2(S3TiV+yo|$Wge*X~kxT}^fO@HO+hMFc1{BjO6`0ng1FfYM zqx(p)<>xu6&BpWdmv?THSP}tjN-B{}=>0ZQyexuhA=EFO!PyYm&I>M(=lbmfYFAx; zH@(Igd(13u8ZZ}~y(o-^A{)mW!@3nXV<26Xk? z4z^}8fXVWEtUl=}dZc~156DA0x%gwTs{&xqa9F$l zdE&rJcX5=f5oz0 zJz`2WX19^Ef5i9zTrSxIXR80?77#;>8kz)`NJ83;W8nE!a>d2Pvj8?G%*MRkwV9KJ zjKT?+#?u8Vuvu|1e}gxdsqZk!hnKF=0I1j^fx^xLIt~=!crkn$3UZS>WJ~eYGKEgu zcaqMq4QD0E9hdOcaO4i9Sl_?h5+gf-p3^uPw7=5XMKW1aEJ)GQGVRXl5z)c~iLkyyfLNJ8MK!kgkMh)y?>~gRjV`-`FbYtM>8+1!d7@4!<65oSgZzs77 z%k9|#GBP{KYd(I`0jyWRYthJh-Nx`ALkNI9hK$_qQ|HU+$<%|lWQY(w27M~jOGQ3% z*%j(3(w}93I5rBXMihbwuXLtng9sBx9Z;xz@`xYv0ZKEOlU>B@{Cqng-*s)tO1LeW z5$b(MwyF`(0uum#~U_)e7LC2h3MYS{|Yc_bL`h@3P{>_IPTv1vk^{i zV!X(Q^8lLtW;qanr8kM+fsGd9UHb<(PKfCb9CeCnKQj04Cg^f~xGwA7I?F~k@4LV7 zC1CHe*RARF?+^6PaoN}KjVd;ReWX0d&V|*8U8vS)?5fkvn#a9+RSrq|QIT~G*O4b7 z;aqDSGFdKfm_`un`0ww6cN5tL$Iz>PzJ~r<$Is!bNs- zY*O*Hk-t5UU}tX%M6W6}S?ax7$s#4zal`Wt(A$-Gbn@pZ+kqLV+2?!*cZrw4XTV)P zA=Q-mYwK$WuCd=O>D@ovW!x1%Zg|*#6*pqUpDBU6EJVkK8(-l&uODye4+E#<>;XF& zUK}8`xG#mJjs9(ipuzWNdhG|>PgSw4Bi?o~s68MCyF0PU+TYL!Xow}K%`v`?yAMKguwc!_{DUuY z4*w!27{I59jT1V&D>$$4a)9hNaDl4BMjfyJvEL;C17jK&Ne*xqveX3I9vj@>M|zU0 zOK96e^D~oP%O2wV=rbSLD}Qqo8=LUw$U?`pH5Q-oeF>NKz0mvHh&JQ)zwkr(?#W^0}m5n>$QQ(L=2={hcDBvOf zZY#>U|b9_Zpht#t9xJ~Qj2{sNz5D#D2F$eJzV&RD`93pbF9_i@3*r6;0Yxr zKVMzEQHyO3r-P3z)M#mFFmmeUZBAJ5$b0)YzQ0)fbbGeV1<35%Z@;CzBV^bA!FzAH z4HQOk>%;uYI+GI=d=5u!(sc72@u5^*Jn)n#)!-HMT8biUxRG06>( z074L;v_7r*X;S0iwLdpEH-a(XEU`w%uAiYNO&S$7>+xuz{OQH@b<#);9fHSA-x?`* zoNl7-621$&n;LLQ>Eg=(#^C5X9Jg5*m6U_rzT>rgg^*&_@$q3K9uZ@8Q`p1twI7hg zMyIc}qvQUF2#p{PRA?9?S&%a_i4c3qI#>YX)i-gE!4Zi0?fUN;R6hSiEBag(Fn(AX zPB4kSfXm!l`1GOeP=0b{7t^~y?f-b%t)KA)1JR#vBKzbRuU`N~Cd?ueZraR9!)ml* zP(6-#!3DwU$GOsMdOllm{{*?WV8Y0oyr%YqkJXdKSx%1j1kR^{t$_0W>F~AxFp$7b zR$!hbCS>76cm&6n1VP;pAlX{y7w8ut_3Hs1&?nXwal=!|*2nWy;p4`5RSDaH7eHZr z{M!BU0*Sa93T#?FuarXbYn#8m&o$q2iUyT#U6F68|8x3$(*j1$TY+46HS&$c zdQx^XT_>dLuu`xwTU7U_hi?;VhNNZ6tg<8yayW%P=5%*Z{237nr^YiYC2ZPSWF;{GPn(-$)Z;82wax(UZ5tZ{7ZAq{uRDb-Y^fcWAM4G?Z&| zYV6!Gp_SnnM801JAG132KgcR}KQSCB@(QAr@Kp$oVYfbxjw4Adi?$M(bi`D|KD%#0 ztQ^QZ6Tdfe5Ls+Ba%aSByo3m|9_ykDS7@Sf&#=u+OxNPx;Fat1`D}=Zj!fw|qJR?i zSHJqzWGJ-pYxP!I?`@cqRf2co2c5J=sB%`t^X}))e z+K0Do$2>|Y($zi;O-@G-LxWWNt8!xA2jJJa9VN|gnfk{VcMAyTZOD^C^mqi1hyz@1 z;$9479ECPLchpmmpvLZg#4*bngP?xCvDK-}>A`oHa4eQgL9|repYErsN51Lr)*^V9 zR!_=+Oc+9L-y_R{*g##^zMDeN&d(25MdwZ>JZ&8p_e0r5!+ZF$A8+>I!b zjTdsC&0pkB%Ladlht)&3I8z!9qnMGBzr@naROSGhY>O!c`6*{Gub<&PD=kpM-658L^iRQ=?V?a8l#h-rm&EfxaQ9yh4-`J%(IdZ2hicLEOpzaw=%jw`YYq z7lyyINnqcIKt=inDO$31DzXhZiYSV_Qof= zN*^7`B;O9!87A@6Fj6z^PY*8S7?ypJ1o<}?z`Y@Q)Kd@96J4fa#Q9JF^_-xZ4DZjr z?Y=S?L&}1xfS>4bRajt2*9Qm_wxXh<`P7H1iNY~B z$E(A z@k~r|q4$RNR!6jh$=UA!Gfe{s z@YHU83t*9xmk+^Qt$qr9CmgBssaSMx2Q_((4o~8_o*%c~j~}18FIMgp2fL2?BnBP$ zxY(aLR!tAKoXjhuFQFWc7Cd3;-hH4n>SK$~kOI)wW5xvGdTnvc8?F*p^+b#7A>5#DYKzQkA(uUhNq?Pu2XR}HfRtg_X)*-sak{b}fCya}x#{F7 zu>Oh8rA6E1cwX7|^{Imj)v>11hy}V~Y&J+OqCGdLMQM+OMq-9MsRGwc)eO>P?Fzi&7H{fSv`Vc@o9N+Wu>tXZIZMQW%`B5jx2=#GhXE_&E= zOzN(W7O+jBCHM|CRtws&UhVIBXmOlc&rTmCZSLd#;?d+SsTk`#29ycb?`rQdn~w%1 zgEZyAiLs&Q=M)JLq2vG&BZRp7vI5Gm?2!yxy3o=>3-sehKSQFw)%L(^1H*k~e**lA zkCOScQZMH{JQujbdCzOxqDMBKPzt0{>$30~X~kS#SoX|~UY(8QKAn08Pl*^MsD9zR zksxi^9vz7?sz_)_#j67#lF`e?-fOK-er9Vj?TEI=@ovD&s8j6$NSxYYC+0qSCckE# ze?Mbo3VJbfWoZzV$d*eegJQG=D{-f0hr1;%_UDJiAbr>0gLqi{{b$hF*bZFgBaV{= zY6`WErRdAVsz|BQhnkX04-c3kB%kpNp4$Myp;;j8@URwGnECyCwr+otu zfOhP#WdC3l5msHzdXM71f81s?+gwYC(271ZzGc`d$1hFf;NDrhuIqnuEe>+|&K z#VRCX@(qh+yQx&WP~>zs%n6{ztjt`>!a=$9t=6^h?Omrdt+k`;PLd4zTu8PoL{y$h ze8Fqw=qS$?mAIs;ofolJy%sD>go~{M`62~8O>KSW%H*!BgqfB-nF-~+C6lxnLe8;{ z4bk-x`z48zDv`%7rbs2B+lGo(1j@go^hJ>d-VC||$fD}*lTZh5RXAjOrgnrZa+2~( z;Q}S$&QymE=qIqi9-^>>j-iPe-{$g~Lzk8(x$nIaRF%f%1XmtL@)SN}2_P zjZ{jhFz7b?Jj3cuyv}x?8{3Y!YILGrnV(SO{oWy?F)#J1PA*5W*PRf3K=Lb?wusao1Wb^i;pVlp7bCiig2tmc?wh^^N+BZLu*HzJ)#J#is@0<5?bGLBH({ljf3`YVs%VFQ4{GIXq03KsFtWzOy=hblDZ zD)8&WZb`sPlF&2ltJuNOlokJzZs(7>2~+ljm(#+`7Jiu+S#Gaf?PyDDb7Dp(sa1S;5G_~S=!|lb-l9I<)t@|qytM0Hro_pd#74tkysm2dH8D1qbyd9e6J4U@pRqiYeGCfks9_#0 z9F#a%Y`K$k|IAx8QMqLjN>MCdZS#9$$6f8{;8a+w{JH1%gN+^K{JN2n(D$1Z2T%KN zhFixIyI7}a0tPZS6H1p3em8~G1^+aR(c9dKHFD%j z)Z1O3Fai-^+Y+t`xx~1!2Io?I7t_~Olr|dB3lPq=x;AW>A*ARYe`?bxETrg=H*F2P zzOKeR;UWDv;6UqUswLZnKS&XAGA*gRDs%kvG7Xl~D5#odx8I7H#$phjEu<_&WiK=} zWQohD!n%&R4#agm{XOd9mWA zAI)vwdFz*`Mdf772JBxH3EKZUhPJH}ds=?C-Dr^-b2L7wur_3C*DqMR=)dfnw{^B= zr@!-b&0Ib4HsTx1EKn2MX}ZGS{G!0QMLQV)j*!Wu0RzRVMlJ7ogpU>_W8PTB4K_o( zAI()03dFB{1UbzMhnFa0jp)6)Bjv4yhmQ{Xv1-_!GB)lO026Ydvr&OV=SLBBm$5FH z-<4%kiW+K=Mcj`3$RfE-ZG^PBlgN>;yNo%M^2?VeYd-JHRCFkkd+V}PpK zEs5|i??d~7LHRA4Z7|16F&ro#~olc)cT4}C6 z*AXWY@fb@W#h{qzyAOKjjF-_Tp3V&U>rE(Pl?NKLSW$dmPLM|={mZu|Za;jZ(3Q~L z)3BFv0w`0GQ*L%BEy%KRpJCMtHR280G$jZbm)X}jm1gp?yW>mfak*G@!iqotSm2(1 zl6hawN8D=_DqNi?ZLh!E)%Tw5x<&W&P*bIIr$buURp}hX(X%Wmid>ICzriFP!;Jh{ zUfrG|-<95xa4)NG+Gf~hs5SvMXy;ZmV^PZPG;nlyh8h768&Uh z3Q+)Ib>#y+*^ot!k(fKj0Bn%KJ~ zp?c(yK*)W9XysoHxapV zw7dV|3|yk%r7=lq=6K^$$ZW^qS!;kTNk$)$bwgf>wUp!KUTgJHhG&u2Go(UF4-2KZ zPxs`-kkjK<-8$9`wM}6&5fvAp;?*$$I#b-K*F>kUOXE*_*G<^2cyIVjB)3SXNI3Z) zKL5?j@7B#h&F_=pd+%v%e_ds~REYJ_i*K6fQnlzW8Y2P)LOO5}5C8Yss!3~#a8cgl zk;BK%uW8&@4L*`;2XfaVi=@{L_60mbR;1OtmZt0^&1Djcp2%@O;Z1>P(4z}7$KwlD z>sZ~OU5vH8-@X7Qj@lGGJffY1ej?=$4zBv{@=5PoAxDADO;~*t;eDZ{VM0T%-^K{N z?w+skju_5`oF^{%mc>%|IiSx%0u9SHUC|9*#zh$f*0nb zNvUlWD_xTjmT(~b1oF@9|5QpUZNY*Ecd?^FF)Tu$k%!0sLH$O%J6NdKm`^%u;kQU(S-gFfi-OsH*+`5@W zI#=`xK09?Nj-nPx7{EdY(hp7Fq~R5$HEmx<<9)J_O6 z*xWJ7pIyI|C;7f5IglCd>60x*i$4OVNE`+i+cR#s5G%}o+ax(R!_NYg?m3t)dz zO!x3mkE@P%=;zK2>!wknXm|SLt_p2v*TKqQU40_` zZ)&*L<8Qfyciew?r~={Xu9wnL{QW*qAtXh?=4++W^ngzLK2hdC`b5ASh@nszMP3M; zCKi>6ypW}$bpKqSD|ul;KAi8AkmXV-36^+rv{W)9z3h>Y!k3ePWtsaYtQzNYG0_uH zcFc`zz4W>nyOW>K)JI#=W0c~gjY?V!tsqs~Iw_}nS$%d^PV@yqyoX9hD63TRR<55O z3%I6I?pEjYVM0q)$Gd&%(uv0@EA@B0d>r2Ri!Vjbkekm$5mi0s60~|2Dovrzmed8m z5u z*#N$4Q9BP4%fg2?uZg<+w3%N~!qi_x>DI?CF$&O7U9#%^YEDUgA~dRfqHH*9-L=%s zKpKmTm|UV>JGq)hnSENR?IM%rq!5z0;a2cW`tk+AMWh=?kMnH(x@y-4Zs|#gm;>46 zqmx)lB*sGRClaxEQk|wlH&Iv2XIL&};&Fm1#-N+{%~y$Cw|;on!I*qP72E z>EJZWN+x98z4L8K0O{zQTK&2~9c*Grff!lT5oW#m&7hCA#>JER6KBfCPSx&mM*us_JD7upRNuL8pBs%uS zpd!xKJM|`{75EB|uD0~lN$vb%pRTt)kv2Ek&qbnq##||WLe};h-4SBJ?Dmr8PTM=7 zYvyivnq{Wb)~f5ncRmV$fcRx}`G`4AXD={J|ZhsW@<3FsMjU7<#lKo701C=5> zkyUlIY59$^%z4{}85)kaeQe;$UMM zTFKrOKdPnpP`_Ti(#g6RXosts7sk56TJ*#2+qk@-@Ul5G zJ)AA;wnB*)#FSr1aob>?B2_#&ccol4zVj76TUci7`s_Z^{lZ#V?kh7ERL*YG*u?C@ z`!u@8P;9QDw%aRDm~D8E_3xkirWZvUKM7}T91bVT37qZK2b&pCT}paBQIb;!+tUg1 zLOYtmFR$A6C&*Z>*xlcEhO&lyWz6^2V(<#wql7vr;ld5Jjc_{4-!~=APw=TZNomE% zY}_@5Sm;O0W8xvlBdSCHVyXamlj#;%E{Hq`8{F<1ffV< z=!Q5k8nZUmH9*r@5;y0RSjxp~T3dMx;xqHCi6TyvP#33QP{U}n?Fh_!`^(5SG}qCk zI*_N9+<5-_ut^9xdlb^dYEgJpeN|gsibZdqK-~KTI0F`w*3{mNWTgTyrh*i)U`FJjjPUm~aK zEd;3t9eG2q9-T%tjG|7rOMruR@kcfOmSR2+Z61?{=UQrq`3b>3%0UbJCC#;AufOiM zV>q+~IKGJ3$rDDqYkB(Ex0VyzEJzcJ2=s0DL{pv%A(SF^%`={g`@Hp$cUY7q%Ac|x6O8XCW9v)1 zTZgvIq|AqIp@&6~wJj^uKJi54LGnQXRkHiM0oqnDAd8V^CilDBmK2pc{-@nY-hC^QS&2RuIi zW^^hTQ}dg*%yGi7)V2gOC+a${;<`z9l9hFqeR{OQyCd`ylDXeyexX^5rm|&CacR=AlsVVAHb^6;JBa8NT1u`2tyBhYuiJDi!^3i&PQ9-rHyMo&Q24+AIZkNX2-Jo2S)J@38q+g?cC&wr%Kqw`0MMs!>^NFGa5R_B}Fx{1M}QC>QxP=g}1UiKj!)>ixw6y{Vgi z&-#P04%;t?Op~_!^LB`WnxqaQt$L@iSyI)a+ae81q+Yu*`;rr?tB0P6yTGt(%2_>A z(#}tZE@NX$eLMrqC9J-wH=cy5x}9uPbhr97EaPm%N7^}JNT57$E7Y&p#|ge4KGfFC z=mKG;v!u;oBKX@F>s66v0e81EQ{-Cn)n(2&V{=J!&aKeB)Ps-2yCWU)7(w3RcC@)~ zRdVmS{J=}R2F2-Bqcp^AFjIKI#BRCoEq;r4Hp(?X%pD^1rMAJZh&E8;;!gr;o#N<_ zMzKqRj^o2zzlMv=5dN1tIE!;OSow#@9bB&Ejei~k;$VfxvuD3_NUtRrgb5tK(B$5cUm*+kmw;}aQ3{jk??&1>LP3blthk(X~#Nb?bIrC9S(KJASO zOFbB=z^}R#^TA5J{&oyc-B;~!p@4~{yeEKAQl&w4bYEaKUFwP_*h z_=3zWlO82;%xUKa*auDY>s-3C+3pBT^Bl^;Y!@7D$m|UanDiag6Tf)2+0KyGLjSLIH+5z z5i3tQo3Ripw2m{U768sTc;!o(s~-`E&bNB4gXsQObTdeB0mC|$Hixnz=+*-UkJqhN zxxo)>SEwcS;K(A<%6muv9~!;0392Lf(X+<*sYex?hYKl*mN*qc6sP|36cJX6QIGKh zX<$i;cW)72BIvrkNPsIn?MX;Y5M;DwP@JeA!s#GhYR0Sq9~ zPXJv?xSa=`%NgXyl~_uxycGKMwTXmBfu2P-=G}b;e-?nPuI}-a zs2uXi$-Lk4A`$N%GF9#~d>T%<@r62}Mq#^RO%FoK7F;C4i0ZMi(nhk(5Q_W9svUHK zXqg_GcB9SwYJefN@=(nq!~J8eb$U_le4jazsy358J$H+HR>Jv)>rZq($iZ}?r`q^=p-3$4qZ+rxYp$=ahf zI(a!8Og-4nz4D^jkn-25-38e_$vkKL@dXfYvkTBc2p6cskAO{d#fEwF+U5#g$6o;A z$;HDc>jvU<##Ghf6hDlCK%8*U(7E3rGiU+ilvZm$KsWFZ{~%#ASr1C|J)9b5yRTEIy895`JDAIkiq;FI#vF0D&| zlIw|?$NkhGddTnWWFJ?=6fXpiH82IOvorJyN>pEJl_4&73IeIi;abIPLXaWW_4N1< z*>fAEdFP~Vl*p^S@gZrA$sG6kv6;_{EYFm>x1Nq$;swVnoG-W~U?k7N7$5!$778|R_6H+5ZQ{UvHr?8{Q{eX__BN*gMML>!)Taq505#svfq}FtyeK!D|FU+)env~^!=KTF8P;N@7(vQwoVb0F+Z6oS-0le&z-scW-pYq z9`5%h0ko( z7rvYn*`%-o&dBJSgi@>QJ?^A#(5FbJ{t%kKRWMJ0BS2A)_tPLjk@Z&pX-HVuUjL>3 zH6hW=vh!tD=GnckZ<^W=J_q@8&XzE>e(e5+)9G-;WbC+G*r@Cjj6 z2O`I(e0Nk!Xt{?bTe7NtwiJMp|AdxzyHTYi5d${ZVN*xMY(5GOr2QG`Xbgtrb_me6%54Y~~Kx#U@} z+7pzrla~XxiO=cHmLb}xvZ{yqJE2^rC4j>+BsM9ADvzl*&21>c}M1f z3sl0?Oj+=XE10iVZt+q}K13+5c%&fO=9~0hE-!83d>c&G^w6y|eI7z`nD4$nsBu`~ zKC48YouRNp?hVV&j`GUBzFxBMN^^SVr;ggF$}8<267X=}L>ARa$gJ;$4ook6H|g6C z9Y{z9(Y-NYarj-=219*Oe$%q+?V$XC11J=B4Ia7zk=FeCa_~G&g65W=P$ZOoBfS{7 z!~Ps|WAsq+5+>{%gfD%eRIvhcmJ>vaYfSN3NUW?p6S&^8*@PrDo*RxHch`6xHT2vP zaZIBA;WHPb-T`=U*IoL-7OnTKMg8-={#6u6QBiu zzkIaIdd02#Bj3aIx_wiFa{Kre+qw3b93HI*O>RM|yq0VzP-lgz>;u1Ldn+hnxRWg5 z8Jl#Uikx|XC^kHlQ%`itEGvyzSn+z^Qp~q*o-$ML);Eu9olOp&^0#R! z6@A%bd7Ha*y!D;PqJyD9bUUW5u*w3RtRAe}Aj{*w$kHOd9~wR?1uAL6o5~6rVB6%X zj@z^d?y7RA2bZTkCALP*adbIRO0jaHKAy%fR~@nCK6mDX3l%)&`XB-G&v6WWRJR>* zh%EUiY7O?x|4=JDH~^#||APNhdQ;)w{(FCswVTlgHJz94hNZgE&(= zrlH&aS}#72*Q{=iTbp}$F~-rD1>qyP_{8I31esQ${c=K+EkgNt9&6&2Vun<+Q6SHqI=#(MX7V>n0aHttGXx8^|^KH8pr z7xF{fH3=cSMt5K*-%?C-)0COpv0A&-YwNPPZ0PrLtJ=}H-*0a~Ti)w+L+iHqX~F@N^wakJmpgX??c8fVEP2`>D}3tpll6QGRbjb!F3awwpv**HaWIxj5&*I#q`$_rP4Jc8*h zsRH&EuPfBw8p1azWApP2-<;WTtzb9H85T?$YoX5VSg%a)GjPZf>*vR;B2UdmMwijR zByqAmA-Z&CN7SM?D2Oc5ry6>aQq&jvHE0$61hq7|KL)Eq7HvrGryf7`X%@Eg4|6-* zkna1oM~yXJ>7XrBrIWaFZwKkTu@a}5(Coh*7z*tHe|@b8qt$=#Oc`Dn4sx#;-gBj5 zzPx_Tu_4FS_T?XKs&drY&Lwo~N*(#JPmE2li75}n7K_xb8dJ|!JzSY@$D|x7wvEPh zl+TrPSGhU7&N5eCTwtJ6(I=V(^;$Vx#+5(0jA|(Rrn4n%Fl|^VTZ+-VU#A zAiv4}Sa-b;IZx?%vR2~2;n3lWSU|~`CSH6u1IijzJ)Z&#SFm8%lAL&*JoDyAT?={; zQyP&V?%q2@&@ImDAeZI$`ORmT4xFL(O+|p%=c9==ORJQUlMPl@+L#8`tNUKhUUT$E z2dftBRq-%UHd}?nE$dJ6U^HhU)+cU59juqPt9chxpX>YH@ z@@^$HDy5Z2ogOG%bz55B`#P>Ky6xqbUgy-7 z&drx8^kJLGcfy8W#+}B4HmJWcaWLf}8buZv zGWDBs_SY}GMQ%jqTd#V4m%&|9y;E}i0CD%3gQWRdOBQ-~e^h3*8>C`)a=Hxw;YjS4 zpvsRXxvbls&x@NM5iNvVpi_oE_KIMsfv1twH|}AW+(i#x5ZF1Q+V~rX$-h?6oyRrd43O6JVWMHb@UGAH5N*Tx}0HJVqJw zW38dw-csX;v=g86880~CuEI529#!j;^1Y>3`K=xCP+}vXCOJwk@VQ9ry7KQVqL+2=SgXnLH{UrzW=9&*-qGSUBV&pwV!4kJ11}2yX4lC z=q5&!m8hAnH;6C7tB%646S-PKkQr#6RyVOv^ht+ex;=Vm5h@xQrrw_3DZomacIBr* zybXdsscJvTZ@3L)x~xTwsIdM#Xot`3eY%lR%y+4V(0)}BJ5>vc_-D(|EsCF>v8?85 zN!4M{GFr(ZLexUCyX!-Jk89p+#yopA)7en4--6zydDW@!ANx8kK-#h&e_ITQU z+7@D#b_&$k5Y~=O(LFl1_qyP`gAJl(yxC@Gl-F1-EuyJ2!Od;1PNF>CZ;5WWPB}Q5 z!#o`6z|hHBLNC!mK^ur*3-4&wg*xgdB+`vzXsp(z`}@l;cTe`2w)3U6)sq&Mn(q+K zC}*u|eR*KXZVGzE{nQ^-hDfHkOQr4g7SHz&QS@4`%7CKA2);$V!{su869JD1X>$>$ z=<_}vikdsrMom=rxZw373UG#0zXmGKwKLDRvs@CEO|S{99GhyYMIK|Iwxz0-7W; z{8S*`1i#99y>bER!{5{M{O*5>8DBM=llniDy>&npTiZ7bNJ%3g(kLPz2ug#*mhKK| zr5ovHC>4+n>5z~H=>`#%?rxB7q?>nb&bgmEyua^#-ha+HbN1fDtXVT_UGa;1@2@!) zvHqY1&Us+mqe(V#km61LecrWdZwR3ygF1lemxm01n)4rDFU|ZKdaIT7qCV{5L*o;H zap*qMJm7+}YZsHZhWx=fjTcILd6h)sMPBo3ZTM+R|M=@MjNQ>Ym9e3>{uEZ~qQ;{j zGNOEU6&fU$f-`I@dRbLGc5l7>ZAY9PjXdB?w$*)3!v7C4VANL*G_(sk7(%ob;*HSD z12u5z-*q+RqMA`iknC{(CQ*I|B+8I6>CV7^k|?*q_X#BC$dN-+`5bYyvTX(z0wvcb z!wd1+d;wyqY)*6~`cKNW3zB%RFarDUp|!bW1dtuiBbDHLc_mTaC?-8p@)2_V71bGb zCol-ZN~bJ`-hfW5BZ@|APv8{X9R@g)Ga>*BPT_rMj+`EhU~1>h%26{Iv(mID+>q!% zK}L2zToOg-yh7v=zZniDd_)+f?*I|9z&4pj`i5>;nh6e>T*rCL{yO}(PoaN)`kyn0 zhi>A-X7cTHmB$I06@HU`x|MPL@uh&LJ`L}~OioBBRmkrC3_LkNn<3-$Zb`{_N7*TH^DUl7^evJ{%!v= z_rNaapX5dw8$C+uI`Ks1qgHzm9tiK}{`%1J=@0rZY?+wb@l56&mEph|)-3IF5o${m{g$IP zp4f;Fh=lS4=!=~su^xnqKR;agvtFuurJP(Jk#u+}aw6eB01JT-^TX_aQAYWDT2}O^ zA3rX~?^B3{u26W8kbPNT1SfNeN`Fzg4ex-V#xDxJx%Zd84_pYXnJY`=uNiN92%ZhL z_P29SZ%E)L;KSxKPCrScFA>@kr0NB~VIQc3kF`~M&@f_)B)c+7s)<=xJH z+wbpKvvgoc+F}`DMk2?t14D4^kqY*e@H0ySDtnhK5zXGCc=A*Aj|5}%% zz$+)BBPSQ@BZ)|iwh2bDMJf}khz#iUBJkxtr?b8N?*|@m*K>6I%%XrWlVk`IsIST` z++9hzT+Ge1hl@ zP>_g%MR@!+*%wbv#%dpemR178FeQHz2EiV}*T?Vq9Xx+L4_{sJ1HfX2%q|t#iF@eP zgGu>Zlv)H3Lvit$et`G&F2>hWFi^LvVnp#97CR;a(tY?`Ja`!-3fDupz0U$eIJnTe z6$&na__okP@RW#rA-qxIGFj2{R-Eg-=9okOkBtUM*9D{C(1J~-pNil1an3oP%WfJI zu(1Y0#E%@;haxm9%q1NTYDN>KWn`k8uCKJz3SQm^jwkG$%qlqw3Sr6L@FXPXrgj#i zkG{IWZzbf5l5UXBda;@ASKSHRF628tK4Am`Lb0GMt-d}|$Ov>JqGuZ1v(M`zBO@~b zgEx{+p0Z}n7qiPwtK9THU@Sxe+)z6p*$(J%l067lLp1mPtQvgr$1CUUElcfuRJ~Y} zLmwCvSW@AA~I`c-l zW(C_;v%HQ|FF$nL22|qnq6V5EXrF_lNiKzt0q!pqe+W8zj>v+J^T0cR35Xc6j>;#n zw|guRBYlDkG?>5c_#-R;XOI61yafyaoqJgk{N7U_OwKPY;b$-j4X=cf`2f&VF~iAJ z0ZebR*~W&^N4#GKEN`&G>GFKP{YTyIf)x-k;p=;$7}h^t7X%*}rZ2|bz={lNa`nGK zsbx+$XMbL34eKm5>dXY(DS`cQ;+_=#$s{X4zVa+Fs0GLh0N}+#uK@45@DrI+_&NkV zA^_Hjg)0Fd1KfJ z<^er$^PmT~NG}KgQKpZc0sZeSzy;<$|}%0JcSl4^bhNm8q)>$)hPcz8I%< zS$fV;$u7Kqlvl6Qi9%+#g8O0T9Vct{&glx<7{1Gc!B9_At8Rd`4V3zEZksRP3pm}V zUBw$52yO?fimi(meGS2Q9m6_4zji0B%S;tGGd z8xSZrNqi={_lS#tD~!=9@$;$5uMg3jRqg8K_h&`H~Yeg(sW52r2Oq6^FUEv37UJ8^>RNZ$)j7k@|CUUYJ$&MRaql8TXrErA%~6 zc00&u+kL4}iO~e_)b)_%FE;u1NZRhD5txJk8wnw{(dHC zk?d`j^=jtx6~kfi7DY$E9M!lh?ghdSO^XOU$~|O;=-TK%gLZU7Aw))mgt$2R_qbF? z?*3-Cz0L1uNz;ugSb;?YAM1$1sTOWta|vjWQgW$sTF$p{yl|t5(2?fCu#&P8YsjulZ#W9r%+xRIQ+NmZ!^GgR z8M@I?ETS|bfwvwNM2V0Z% zG_U;`p!yzZNe@NE-$wSO2D=P;0O8Ha#QYPzIf5?QB4|7*&%jF9wFE9b(sI1?v}-~4 zq>HULU}SUZ%WJ=`1dh%I&2r;a&@}H}iq*CY{CoL8O8cacB?I3?UPWQ@9S$i+(um`L zMJbf$LEYsWIA`S)FwMF|S?YIvJ!5$IfujR_F**PQgc3P49`nh{!kPk5jVrZVlInHa z+SvyXifEt?C+UNPvIu^8RA7MQqkz#jE<_)D9|W7w1Kl|heWiD)22YQ2yXWB2MR4XO zoJ#XEjEV(_UwzDyj=hmFvUdYaXwk&Hy1~WvsO}5i)qacOhUJ%+%ZWA}drO^g-9jZm zn5$V$|BwluP2)62;08GuxQ%N4YgtB23C*{*V-z_I`Ljc>= zWnkYiL`2Q*u^%uHy7wm}o)L6qQcesdFK;)iWEGTS_@B53mQ&)K;>YSsArfJNDk=!f zu;EQmcwloE;}HQ^65E;4-}~LDh&F>mQa)*^8`UQa8MvNBlnx%wjaVw=O9u{JK1Tm3 zzM`S<>bgIQ#A`%c?k0u)zhuzV;+11qf}vyggm_6pdvIFRs28q2q>14?*~YyUo*~5nBaZ*Opgpw<6_JQM+0{H@bDBF%lU2~mf#tD*8oKW_=xeyiqLv&fy5($ zcKnMp)0EAWmVlt3J(>6V4?v*VOGHH+^oxN26#BqDj);wk`j5H?i0XsyeDO^oJr54W zu`-yDJP z@I40flCaOE+iw3Id<%l)?bWmB0~hE*C2Rh-(#z^XwMjXlcPYD@E<--dn0laS#}o^OqCSpF|*#RH(^wA8iC6C8)9&ha5sYDRT? z)8Bz_Rfam;7wFhBAe4~CzIcmBKpp~9Q2`|-dgYu>+@%s_P5{}U{yr<6R6MBP1Rrax z0wE#_(l*lH$?{7u7<_j#o@el56ae$<;J=B{At06kSDMmic|!N@kJLHh@ta3V^1=H; z)X4Dd%Udeqj}SN95&4LMCt{w?9lk*PFLJ8Du+H6uHcDv;I(QifeGKFKyR8WOxtmhR zBfNkz6dD74A+osO&>+@cZY@ihN8I)}h(5b`@QHM}_a+)XI1Ic`;Cg5dzS(KQ4^gOv z5Tfh;KObRkZ}7VRqw4d=e2AexOcsd3ue05xdA3$=kOFjLcEMB+HDOysMTirS5u)Ib zKabZ>d1b<#5=XfCfC1MhfG8kjjBVjN{Bv3?Jo+TMx|&!};wd*0$yB*fc( z=R%Cy2=Pi&iSq$i3D^m!)4hK_$|yuP3aS~5ZGJrKCacF_`4%>E2Y_-%YG;z3fQOO@ zLj#^3eyp8JV*2Y4UV!^wehU5bxBpQbLqbQ^np^#y`HDDR-?jA)n^!6WB_Ro3TPA#c zTzfnJ<8M1w6K-cVh$nZnVu|Js3d9)+vDj84sNa{O}8|#eWIwy^-8QQ_!mg z2pPjt_=tk^Z&s3kkhZd4nef?zuJ?fYr$8tlk1hhQt_9Q= zBoY2`FAbw%G6xX0Wk&kLi3FVfo{aRUTLmT0*w?Y~pZyR^1FtE*iy{y0O@AVoz;wJc zi;zWETPl`$t6~tC>%?t1T!M71enbnC6SAH%mF3k{u4^#JKu?J0woRV`b|p#~V%Y5w zH1B~YlNiWuX#t2TJTo&BomW{7vl zQ3+ZXDfeqiVWxy?9^>abdjFtw6c0o7w5 z98+#~BS4B!f+{BYZ*Mz{8gWP7!pE)*=kIIqebfiq+^0YxI8))1&E=zjI+;^={+#>(5rM#W86wLlT-2a$Rk$Y~jC z>{pJG)Ih1E%R6TzVVD|dTIno-#QCtm1nFPz9o`$T+&e-ao&p6|_}vRf?ucQeJByle zD4NW6Zd$RZV>9V#>-!`Oq8K1zu}S(B0E&qwhGD>!&1fJ~(r)f6Vwqqc5bTsZ+8poR zI*D=f0Zo#&iE(Aa<99LXgTGz(q$4mmGEHN6mCB6B0s}*&Jztx3xdK{|jK>yK+B#Mh zV-2le)ey2BS!MYPpBwd$$rp^XHy8uF3F+*qN{va9;@;8LYjhbTXw>>K#O}?kHG@)Va5R&;V&0G$ zb>X(x%hDgMXiQgtF0j{`SyF0vBwq^JDA0I?W7%R?lzLEXA+^zC^(eA!_bw*3SN99B zEPP}w)Y@!j!(tDAUe=l&Vze8StL3~bD{{rvaXk0DL+0%Ceb=79U$pFVXA%QJ_qpTB zi0mDcXq|7>dWS1PTH7BVJCbEqf6$ra>^yD}j49k01YPfA2mPM$Uks=N@ui0wNHW8 zupWkzE;#rTD+;aFff)O7vc;Jjj3xeJiJHgkVBS)GlB}iI127QUmI`fRD=Bm%rcU=J%UnM#AxK%ELLgCT*$g;8eT^?;E8rF5t&QGscx-bYKTZ&8#G>ST zd6z|(2~G_0G|2Ge)^!>;8ff%%ZwLS9>*W;y@EUJ5{fhvSQkAvNzWBKx#00?Sy;~KM z<)L^*l4!(_l`;b^fZZgk!B4DsAQ-EmcDO#A1yrIz?L^BAh_UwCZDInFfDOPYhQaX= zSUtl98iRacaO+eLWz*h`lo3xH&?eLN0DhQpFZh$!#!51H=p%8`GS@)bw+BZL4g%+E zj?si|&46`2MKVP-h>nKaq?8XxXH=@Q_u8mEY2X*@~aoNdxeU^jb! z%YP6KL`B5K!YH4BWQOCon1gv3Ztc0Z7$yJi06UMe1?WsK_6lA{!ayWi` z{~k9w-jL#Yjnvslb$`Fv^_@oMz2=Mg2rOEVWX9%-rQD>jDM(l=RE7>yO98| z+3KpI^_{mdqHewXeGG!WCl5G1Cto zZXroE<={k%g^<#Mjlda7V=54)rUzk&-#3Z-H`~qQHz&KVU4fzDQld?xt+#2K{OpD4 z(^8P#!TAPlAQN0{%V6U@a|S|8i=8Y@gjP6Kv^tHTcvmLa4RD_dvfGT|}CT4hqVxmI!ZixM1y!pWkt$K5Qo_hB>^69Eh;m9{{iZ298 zi?vTT6H|;3JT6LaNcHdEK6j!$mB=dFyp}6RZe_qVqxTl$MG%HrDSwJJO%7D#m!SFWd{7S#q7S?6xhG-=w;PlF- z)pp%e<)30qCetr6^53!uZZh;XOwX1vJt4TIYqkP#p-tFU#!ZlGVShy+M)PGj&@@%hQrxF5klKlKWkP5o=sHV_}U?MokkhCmzG{JM|h&vdT} ztB}x+rn7T(Afhb73kRDPKBP~C=skMU^kLzpH&Ns_p*c`SNzX{IC?k>8+)x(ZADQr* zAfA8&iyH$6e4qoZ8Bi3KT=;2X(esc5F#?EOCw8Yfx?ld(SOL`^q5frVO&VJeO&MGL z0IY~!^4s-5Pbgv7#O>;I#R#aXiwVB8wS$0=r3CPT?xvoH>&F41IF7^xht<9p>`!wy zrX{PYI>JIjf%Hnvj{=n3*&quN(pN*&)*cv_Uu!ytT?@@}Wve*Sa$Xob5x&)$P>yP9 z9u?~E1y@We)pO_y#1SVT)`I@z^r4I9|WIP2JB&~wEAszD(;Zkl_9x8>d#&yV?(LX&Rtr#_EL$zEi~j+zZknh*anD&m#6z8E5x$ul{M;=+h^}YTEE93{;)otrWePhVDDAB zwJ{rZB9pQ?j|IzmH&@j@%k7&xA!tzLLzCs^_h@-}0_!^8%O;UytzeJc3z%CV>E-M}vEJR(c z$bYN5MfGu^HPr3sc;CO698ux&PV#cDgHjBvxY&Bie5tlzc(K0R?{-_7= z)M4sd1{TVfT`)kwg$GIgtN`7Ih+OLJ?Zs$`B3Q9j5$aXs!_ElU5pn`2bGOQFhoD}U z^Hgv~R1Ms4s=Ukx9SX-^0_jAHic?;%3l~?QxDIuH=Fucj@v|j;br{2blKY|wBq#x# zT2+xp!b5qgjBEk9odTe$S&Wb~!aQU!oM|&r{&@xXd|32s6a%NkPGJ1b0xx!I*ek`3 zTNJE&q#4HWJFcs+A>?^#;*fJc?R%uVagcZdsGmdHMK5N+qOZ=LW+6XOt}|P1jsU6b zZ*V9O>q#c|S`2LDRJ^HJtBYaZ^=QH73P62P;f~Z`O&Zsea{w0H4v+vC^{0y(Z@seG zD{jrMcylQ5AWIIm(~pg5%*(27;p9hhG4xhZEbaQ67F7(9V5l>)zt|P~6cp#@l^NT}PS!W!E zmJbSBHn+LX&wJ3ZGvDwyme;!E+8oo9$?_PR_eq5>s>W6`_M?9%g>j^OFp~s=Jqv9` zmw}p~`4v(8>UGUh#0q#IbQ+siSCntbCzPMo9sM?y*9--#-;(C@;six9NWmgWQ?e7Z zht2$V@&yc5uDQo;k$@0r3;0t)m#h`t3lxSw$M8eb7A~luaI$S^szgVMjqxr|Q^3%R z4N9-yvb}TKq~!VIyZUBhzewY)0O_^7d$kqVY%Qd?+byg&6{;><01y;ExTrKpT%AmZ zee6~m@6))M4R$|gsPFAzaAm+i<}}4Pkb#nEGQ9%AQr*W(6M3eqeV=N%Rol(cW^)Zma(<$mQbW;LrcPmM3-U+|d5nJ;?>%vbtk`^@PQI4_` zHb!BT{9S9>HFoX5QG5rCJ4Q1@@J&?`uto| zA~_^G<+u^ak$gF4+!?PGMAALKWj3jYxmmWBAjgghot{Q*pUL{vcshiO#a*|oN;}Lh zm+5d0)vh}+C>NX6hDL<2rhe5cO3NHc@6<{QUx01sO?-O>9qhiak9KMOMra&L&il&a z79Q=m{o=u5IhtVM+aFS~Y>6mZ0P}9r$_e!rpz_*}CA4f{Smo`=kyg`+rhGjLx#mu& z-XOV=1QMyJ^Upg>?wwxRhwK+6yMs9^diAPvPJhHKv#5TGB*}EPnOtm|tFO?ukhIuR zWHfsL;O{;f{h-{VIow;WrA><9%}jd8iz<-E=$zcks*@}+ak$g;=s zE2&wW!{}T;4vyLd$zDE!d6qASsl%aB^z$^_%Jl_hS>D!P94|>*At#3J#hyexfuxfU zSjlBl@49pusAS@39a*fG?aLN!i&Jz;tLALRkCNt9km>!3AH7YnHJMU=H2p0#S?_@= z!0n~HU3*G2b{2N*sh(co3@h>9NI9&~kwVrabDeX@Q2l-Z8GbE)JFw>0EWpKcH5>bG zq@#LsJWC>QV5Gp57{Wd(KN+Q3auM1i4jI|wQKr+3TF$@f{xR9e^ajw3Hj&GlT&ff52s2WsUxkY!flnOTMJhY{~nrC+uy zX!=lb2olrS-U^?_>EQUIINCft>WbnTny+^mig`VGC5=5NT2{+`Z@!X>K154~}hAJ(a*+;-$hPHQSAJghF~ zl;QGTT3>Y|zUr|#da@)(rKyH$;b$rAJDStte1w^l-Hp*vW6OQP?6IP%rh(ZyZ@; z&A(kQH`x?80!xfJkkyo8;BZxXOapFy?~EtFl{;dS3nt16MKCu=t0n{>1-(7f{6c;9 zI0%~v6QU3D&MEeNfF7jZtc&+Up74k7OFW_%CFui%if%g%!W~s6ks8KMCQ-7LHdn{4 zk;!%|GtZl_4eggd`ns9*O0r4hQE7X=zp%JGYvwFYsq(4Em^kuX9`&wb4#<*pM>pTm z*HP#s56x;`uXp=;x4hA?2Qn?b{!%e?kR)hYa13oZZ1y!~Ad*?kTvQ&`Y7__W8cQEy z;afwU3ccwM0q#N+On=zR! zJkgTG%Aav-hIIc+p^DD5j(TV1XSUuZ5r-Lhwl$22T9<(>9r+k>nLzj2q@Fnj++*?Y-+7`?GjQUSsA(q8_pVlv=O-u0RXfC-OU{(;Ss({D%qA z%<9eV7r62)3YQwgsT!B#(+viRwsKVLQ|yb+)?93HZFD*HS>BAX{pbQ(&z5@2IR&pY zO+tMK*221awd6+ERvBcQaePMSvUzHySAYiedVaBpxR)6PuxVnuOEBCox=g4k=V*uJ z^9IxZXpL>E@DtxnA=#g>k226WPqC$`%FmU#FaMNXL(l^hGM*@K?$}!NDGDC&(CXCZ zJZ*Pl0sZjassKYp_Y&bB# z-zZ6(vBtGm!*?w|oRT4|>pBX>we}yo9T6yhJvLFx7I~BKi0;Ys)qbAyhPzb`k_F$O zIG<~kXCD5RtJ}Qe)u8`S-t5bi3zyIM;l24YlVW^Nl56$i)jvLftb1Ra3nxnWZcf*a z&|ej&HW6*{y}BHD|3eP9E_uQLmEAE@UN`ktcLeN=hc(Wy=Nd!^$Jss5WpS$DVe=uV z*DL~iJhtV@CQ!Qgp&h^$XV>lrgBK=*`xi8C(6Pc_H+6EJFQeu9_Qg@JAs`gm(SL%qZ=hM_r;_w(?-gEwWPpR$rDM6H<(LX(ht2jtX8k_ z@VA2?-D!z;y<*;QLu7Ycws` zg-zp0mqBAOFS6O?SrUpV`zuR|p6x`gkLM66-zaPuz#~Fx`1eU8i?_;Ca4W@yn08o3 z9J5{AXf&dVP&cCs_RSMkFl)`veRL*`j;a-xmm<~bGs~OSeA2F|_jOml%uk+|iH@e= z)Nn_yBIV-nM;to8SKhnSaH<_ytGx(aqgpP;X%aVs{H8b?$kLo&Q|Y}ON)6Gj$fsHAGhJ^lS*F7bu?q8(0Jf;#M#uk zE}0dyObwh?Wly(f#7E0&7YzzxM-5UQ?-x@0E9%O(ehH2i2$;ImUt^a#yzVid*0d@4 zMVl#Fcck>mPIcj=v}u{FyT5)%;5-f@=Y0b!KW1y!OLNhbd-L?ilIzoq*#~h_AKT+A z(_HK1_RfQQF?6DMCfc>p85CYDPwh85@xfbKtLkZ)87`>Uq=p5z(;9>;gnbTe{9vQD zKq~|}{W#!3w3^+Mo5qSZ=F3}YBke=Uq=hZo2WDkB!>+#q>4KR?yi-WfA;;PgzR$uY zU515|e0Mv58=`Cj)mMkB2m66l%JFX2pm-sd{lHCTXPo?VCMM$2u$WcA^0O5l>?(C` zn}c?`R6-qwgS}3cShz=8XVq%c)mdjET`}(XjEty<@XBoY=dR$E5uF{UpYy^qR`$ZT z;!q@KL&zA=p4f+`xD6*mO0`N?gIbH~tv%TJhFjnC>c`oiY8@LcUmf^9>=&!0LtjW_ z4duS>Iy<`iWLcw)ekk+Wr14-5)-;!0l+F*@h?P6P#5Ge<-|131RTZw|Of$zS?#M(X zz5d;dxn9jZVY+0D z7O~eGT7>7fAHPC{9lkzI;rCfNXEsG(zZYuUoj@`U{7DI;dIy`!Bly~BL0he6Mt&O= zWNiU~-l&CL-cy|{P!rRL>aZu-0Vu8BH(&h( z%7t%S3WC?i7DAkOaUo3aqO%2ejl*;2nT|o}k9Tyu&1UFqT$()GZOxaVrt2DUW!_%|K41d!BoA6h2R3bRu8_cLjO6Z=Rjhmr_&wvuRsBY;^j(n)l5WaAM}iv-F&LL7Z~b-wUD10Dl20_gmAXz^Rk?Q7fZmDGzVWr7%%S8-_4(^b}{U&gS)*QItq{XXl^LEEazYErFB zv>Z;q290G$%T$~}P?m{3UU>!7pF>Z%O^NAsN)j%;O?KvVuT?+bZjT#vX8R1@G3uBM zU8*SSEnSUReqIk1U+}J>Ope+_L3&4w`u(F+F8jz>-aY62PC5;h$yVo^55v5RqID{T zL?^MgrYl-1i=Xp4r`sTu+jaALo%-^>LKSQ}cPQP8G?0poIoER&Pe`&$*s69kS^D*M zM2BbFe_5$U%xd!4G|wkht;uKDNCjHWd_Os(k*9K1x*^hA*{S3F0)nkS_#wn|)}^o~ z2NW!Nqd*h~qUE<-yNOT`PsDG=3De_CVq!NKQFs^dbcyrmGZFQ3d@pkcUfb$wkr-KrDgj zEvj}C&3#3h==`FRn*>o}!Ha(QX#+&mhk+#mJR_8zph17kYUO>C`m%y)z9=qWSKcPCV!b6MgqL6&6)(+KTi^;Iz&?1p0+u!}J z(0VyVg(fnN*pIyAAu<6UbH}Y~JuB_9f#TheeDxEGe2tw|zIHFVeV*Nsxc5J}PBf0j zq*xCsju$HyRlqzmKl!?0ze`)ILCDN&$Py;FOdls3V48tI&)&JQFb=z%@r6 z4W^4?(nx5Q=*xvsVf&kf!LakzpLs?Wv@~$6@g<+J^Nt0MSrMVsu)Y3b74Ymd4?#&W)$bWF${tdBfeFUl9Ku zkY;HI2F;N9;wWysc^(zFW2oxkkasq!pyP1&>y~#U4JLlfd}1WMeETvPD<%mhh=kZ! z7V_#kv?}i}L8nv$z_aId`oIcEdhjIJGuWEt^RQaZV4s1}^=g&t_=6@+E3Jy}P6bMy z=Y8Un?eyEL>pLSqgr!sq9@BO#Ga=#sFji0{M1ew|H`WwZs;0yYdcG&zcxEwhja$BA z^m;%(<^Ja2Oo^cXW1XUUEZ0+={yEQgzpuAMiy`1{>>SFMtz57z2z4w;IKb}EXSdpCASQg(KJK2Av) zXl#IlN0}YB$7^O*Oe)3DaY=j|U$BbFrh#VT&y@neRCoc16W001aCIs<{w&-P&q2T_ z$bzn~ey+%D)PSn=Oq{Y_cKtf&M)!Ye<^zyyubTOHWWn#p&*F!xPjoQQdQtf(l@1j~u?c6xDl9`i7DipXfa0lp|Yc}_Rk^*0{L`mkFmpW}~mK;H?<2d(?DH{Hec zE%9!IkVs%<`$#t-32%#DH?Kg&$|K$K%R%~u$uix* z=4R&)?fYI^zNtn#YUkl?ZdMoXC^_pMGDDNItcdz>PZGx4~yTE_Q&e7Bu(?~f^T zZ0s8B?Oc13c|F^GvZ}M=I?a8R$>}A$;+V5z!?q>-eA0neUMl9vN`q5Ww)0viF7DBG zBH1X`y~kPnEm0L%+0h?Un3pd?4in117>#ePC3Btchl!}qResKe2y{!grE2LA6ly+S zE4XSsa{8qUgHl3m#x<|57Km-MOuYCcd)E3ppQEk#rW&`8U|Te-IA&PI?pvML8Z1s$ z&$!ns9VQIaZ^h4ePWR1pFnUrVtp3vG-LXtdRc|O#i7yrFRnV!>T{$mDHOmf$UdTOP z5z>04^1iM&nO~R%I{Z9=y$4axb!+=`cHwjN*DrMyY=-9#_C}HV*Mq*H5vaod9Fnw} zD35U6Uk;R$%#;*&)>zr(VjNafL8XjJUD(m<#r1nu|Po9hs+nQ+jWBuU-E zak={eoMR_AH}RmXlG#n8j`RB3b$%=$$GdYsjQs$52H{&hczc6UeVFcj<<{?=ger1) zE3}6|GyK%^;0z!(?u~BrnQo!vAgXM2-`mP;*>(J3YZZpmFp3gTE;}zX@my^MX&BdM zc`6O%iC*z2(QHL)9_no}4s*kJLyz-JZjzPw_0+Wfl1ebg#y%FGWk;2T3&!h7Bjzw~QW-hEo~{>~0LRZv`Ui>x4_F z>kGPnU8uC%pXlciXg%a@`MC7!b+PmkYPV%;!E?Q^2KkTIdm+aPiIkpMOQH?+=b~mA zmVxYMs$AFa9|l@SF2(m4%+XT(;0>-8$sGWyiAJMRvhp^a`rL%C@@0>k;*U5G5lMm}g)UlPT?>JA**B$ zM#Rd|MG<(xi@q1`$9S=wUI+h(EHHAeKf$PRT{Q)XGJy&#Ti(TX6+vr5p8Hf~Z44@W z|J(jz_?*Bn%qR~zx%enI%;eOwlEic6koEXH<9c?j#oI@I`69_|K5?}0_zs}dCgjMacFV+dl)TCv94?>m1+~$ z1w%PSHzS$&;Qqncp}5};FNU;VI!F-gnbfikSZ}4BoepxX)u;3Bbtdg7_h!3OlAC#} zK1cAkSEQY-9T)$qPF)W*oPO*PxJd*fi27a!)OOU>47k6zf8dr}%u2Zo6%0wvI&wF8 zi9TM7+E-9s8^{XD-j|bK6@KzuZFGYj%}ub~LB6DP$c-$eU)7=HQ@8Z>0jELBEEM3y zljSsubAai^yY%5_NyDU_n`ELRg}o(_ebnefB8lGz`b|gM?+p(q8tR>0UhPoAYZW_< zj>7Nnq`G?blv9r^oLbpHv-SO}xxl{fq9^vTvfQ@I0gvm!kwt?&i;0`qGa0Wn$%W1} zw`975^*YKoygy2i3Hwq7bu>KJzGdtaU0bDUT!wk2xP64_6`TWUZEj|5bFn)c*@4`P#&~K zkiyNEIcb~rrTMI`%)unZA1uW*4UfKlp@g(82G^G9=G3CCy zk^clwQZpf)p=pe||CA~u*w^KgxFw%`Ov+Rt&r8E}_$~u)79j^`jgGd-`zWb?<9BS< z%fR|%^W{p6VkZq#<|;VZ`$~^f3%^r62IqccrNUPIkZv_s5Y1YCr(eP=D-&8ATI%t{&aUZMDIU~Z%So*FO-0F7 z7T-y79MSmsAbsLiv_Fgq^t*owCG-9R-V3*cjb)-;!EuvW_ZM3KP$9n*7SL^+`#SfHIEd&i=V9nU(4W<90ic%%Irn2 z(P*3B!lTX-)jSP(l!gXFAA8)@G|b(0m#bl}c7q*RVv5Sm=&4*YyQ8Ji4g#vKr^`%^ z9W5o_!DfCmS7OulBV{JWD6DYNnT8|Of?in9v-gB#_c_UTpXABPwvUHb!XbXuL(im1 zddaK8nyzCGNH4w^xF0^ynZJS{mc}QZ??M<cmejdsX43cd?};Oc z^87S=@5ZBT!kV>?4_TT~BYUZJsUad!^+rV}MztQ-0}4Dfg8-JumFgB&2I$~ED4Ssl zw^kOg;7bAm5`<3uilDD1i5;CZ^Z*FA`Dfhgp!bL60z;}M3G|1)Ft?Il1FPSo2tMVC zk#USUPTlSfDfTyG$HvZw{V?YcX}WWN%_Fn+B*H5K-Yzi&M4SZ69s1+HaO$vVWQWoV zARG8Di!<3Y_1_sdB5)|W%0m?ff*V+{&jV^Wnz^6l&RqTPnCG6e7u@e4I zx~oq$F?VzX$ss3YM_Z6vvPu9f9WLuI`eaM(vW zb6SafEMqBtpF}K`Bl}|!Y7C@hKR;6WHd6`hLP(n zoN$E-ghrxMZ~aAWk_l}dc*(Jb*P zlUT8t)!O)0k?`3{c%FW!z>U=Y5S%i8Mt1w(Ao@4w)_;`tsaFCGrc;ABA~8z9Ch}N} z8VNpM=!CKnH+&Xvm?fB~pChRgaE%FKbhz#aO@MU&Uo)izMp$x~?ezCdrGc43|E0tZ zCoyq;R|Ai+E#piU;#eax#gJlyFTqHc2`6fp%}7}z|M6$^pUib& zD~RP$;4=#-b$`}I)_?pJ?<8(dnJ4OZEinpQt?410%gB8@0#4zQ)aej|ivgg=!6`Wm zf-!LB*9#QlwhrbK4s9QpA`vx$rpm1A_6@w`-)86AgD)14*vk$EiTh|_hIb!6cg3e4 zK*|+8+pckiQ!>d#dGXC;bCI#?H#C`m7_?Caw0FTHYq#>2JD2i7)*0LA#SyN z#oN(sVlZX9PI*W1$`mA;H;72KOkWp#xxMU7NpHAe{%{f>yI|4HwW4vuN$IUNvA+X> zLw%Y!GI}4_lHM)14&O{3UMrZbg_G`2X7ED@F9FO?-$ga)AGa!~8YZUZG=##+nxSVh zWD|*jj~)fS`}Bp<-^=0O?=F0GAe8mFe9xBQwII$~maa=j^dLM3DK~K9&BTA<>3;Wv z#s3m-VPs$^3gEYXz#10A7t&n1|>J*8JR6G2Dw+O;> z2A|}T|9Z&&QFHwJA$toh>}8HJ9efheMd3?jTrN;(;U-3dvvlsR5B(2G2T}wCaZ)1V z$A9W=7!r*^4{6q`oK_`_hc>_ict;hSDZSCY{Lk6_B}C0Zdo~|K7|ZyNB{Mg$PhtYT zC;rU__{V>J&xe($RQ0pR($UAShmu;pVx^_czBm|KMblXrPIbB;o{Dg-xkvDe4?dL` z4-3Fh{5>~Jx#5TCi*P=s;QgMzNPpkpOZHe*=)0RkbT2TI`+xtC{hvSl^9uZ6m4@OC zBpT!W#cb~xMbJ~go!CO1ni88^z z;o1$7lDBP|Y5=N(#1U9m{1!WX-GGQ8OpWWad-ZTd5kLEeoMcEyul%42ikJSrWG8wg z%@^aMRHEWMMlTKk=m6*U5!CnFR6$ARiDr2?2GQd-pw0YEtrduO!8I%PX0NZ*XS~k0 zzQjnsKu4y5;r6|XZu>6;E@V83-^lzNYtGJb?y{sPT8&wmH&IyyEa~ykGAI%oehpQyG@8$AM1;F>mVt;V+TKQ&bnTWfP6r&k`((Y%S}78I^1TyQD)z z=C@2T>*^Pkmg5h?_>ba1$+P&W=o@UL`*3<_d^i$(P_e}dh$xWr=GbfYTIpEk&=9yS zz3s(ZvmYyT<OXNuSU}Ym!)W!HOblqR_uK@m@F;vfb96|>|61$M4c5d%kAkVGIr&KOM#ihAU3@Z}X>!lD zJ*6r$kpOKsE74?+z4o8_VwHY~PJpX#l_V)E8w0d2P}^vg0O(}A+Sb75-B&6eJCUN= z^~XGK<{&5D}2WJdpixV@WE)!?5x#6`&Aplj)SN?Hy6Rg zlNA4YdoKt!g;}kvw0cT(ymJQ|OBDnU%}Ad=4LMVDA-KJn)zv6BjVR0kW$+II^!7E{ zl?!sumy=S69;>vmHKmLqtbzwTDae)&I3D=nc2GeV_tmv+-@-*vLS4B7VC?XS=btx* zni<8Kr_9H4+Fc_3I9pnqwJ!%5|7C?|YW?rQAY`*A-6Al%lo7d@g*NQbp$?eczA9vVSTA0H{Q=q4V2^#7~xR14$^P9^Z?}y%EUjSy=O!c_Sq5@rL_7JsE@^|U;22yU%ezL zAUXw;=<^MekTqy+&6PxEr}JBtbp28+ld2}5ohYrQ=cyN4Yg;h;6XsBFspc^hzusG} zn8+tqQI8%?%k4SOZCQ~-|LzA?ITsgV(_1$IoR}ygfZ*Jf$e0_CXl-u_;Ad(8GgfxhNh={Z_iqehJ zDcvHS4_HVyBi&un(qhmsG}0m+Lk*qp8LnQx_ulvO`wx9)o;k74K4=4@W?QPz8*<;NiLkP^Y^4ej#f!AqDp> zgCTM*(|6t$H_#};{I40J_Ie4JSRguy?9~T9z8i%Cm6yEMEctj8XCq3)>u{gP1EAxZ zfv!V`{4={}-a1gj{DRlR%fgNuGgKi!f)bJDHsgckvO_bG86X@MqXz|e8gNkra> z`;OVx$~%5PL}lc!qa^}|(z^^wMymJq_ao~sWYgvCq&al+x8V2yRr#kV=7sP#AWKCe zT)hThK9o%7Zr4)iad;R}Sy0jKNfDXHM0gfyAQI@;z6Olx^P`z4pYf8i<&8^<~LZ zE)5qam&6@`;>L8@iOx{m)Co{aFx%?ng14fC8eBpy;Czh-JMs*II#u}JJ04w_62>=` z^r^->Jl{x28FXLg08Ug|{euN~vYn?z6G}_}Wf}zzkUUrfJJ_`o>Wq{Byn;y8_lb+b zJ(b4zv2aVSFhRbdI8)o#scAsU$ltqAzfq5rP3z`h-czoZX*K)GDcqoJr&u-y&G5Gv zDV(nlLC;}fK>L8?9hWb-@(Tn3mlUt>;$E);?}N2-r309O8amd61|Du8xJQQb5asRi z=!VbtRlIT3VQ;MJ?c2Wu6m)YyE5%7>-T!&^1_vf^pc@(KZ9M7JctUY9mLY-p8Kot5 zCy4mv@6G@3v%ur02J5LoIa7z>K?OO(U<Kdj2_9}Lwh=VA%|1$L#m^vhLq|jHK?+2e zG8uE*><(`2QoZ27ZmL-MEO!lO=wUC;O&IV6v+knRl>Zsucd0(CpV5{49S=Cl(=9P? zFdB&5fA{1d>Rf~i59g~KnCH&L-0Ocym4WbxyYF*U=()-T$2BmcT<*93I=)MR?J&6f z-W+PMMW;zyy9r$Uqx{7BQ0J-9su;@uhRfzStHKFaf-S9`sM=dg5oO>g;tPB=R^{~h zq3KCju~=LAfG6^`pm^LE$r_4ipF#5Ow6P~6bGG?^wz##zD(vu`zojyr7OZ4@)QO=+(IQzXIjy#b=V^42pPv~%?i+$~faC--6mBR9d+ zfiWdQW7`tNKtt(1;DyzpCVH=at{X%QB*wS4Q1g5oiiY;z^%Njx>g;?AMz%cKS+F>) zne+C=6%2cn&nK9~jQj2L|9s7#_4yBQ45myDk0#j>$R+Z9L?R_W@V4I;6}vp8!IUBC zxLo}4k;;wn!Opo{aaC~C>&BV_s)G7pXFd4;s|qT7GLg9TszdLGj_JXbqO4DdpBR`w zHR(RX8GjxOSy29U{;dVpU<^P;qUT5ZfDhoGH;(5WR)AfT;NASco*Mmm`d>@R`P3BD z`7UPt$^LZEo$N1EV8*dQO9^gK{YRn$lyoJLvaf#GMN#R=k;7?ilTC-2#ui3_z{@2( zm$SIRG z8x#FdC>P7L z`Xt~#F1GDPSa>YSj&@sThAqI5%;ukPmyX_)}lG%S`R z0d5k9S_Un`Hw1I5y4QB%=a<`zq3+Q{n z4uGt@E0~RK3QuA~=|Y8yo_^o-^d1(}i~5Rvw6nsDT!8b*xvY)HesyUtr1t^ZDd#WT zWCLyA_lC@=|86`P21-xQab<|@?q~!~u7`PS3cho9=r3M;z-#V)RRHgzJ=v(XsSEP!9^X)+!aN zEZEm3SRH2~S>}$AX`yD{Q52NCJ#2S%63XnJ%V8~QH z8eF54Q{y{l_i8T4M!#iy*`?5cNu+TAzj3fuMreGHb}?ZSuK28qu!eM7=A!nwN}huy6tCiFz?W=lxgUIk)m)9*WG%adA%b) zbevg|fOc!L>9oE!7;P?8-}{EBC;wwCk9wbO0I2?RAPzRP;d(<)gvz`aWVZVKS}4*iDYUyzi_U<#odMJnRR9-}|6n$9 z4n=D@#c5s7a%$bbrYJ-Y({+yJwi=QH5yMThM5mZ>kj2;nx7@RC2PppSrRbsQKJ7|t zh-1t9WM}vVz~Wpt0Ri%eD`Kj#@fZ-;++s4Box;;mGHEC&O;oOACPYx7sCu4)9Mqp# zTt*yZ0_NhBIUF$m>om-D#eD#ys?ZY%F<0dgUjBJyUr|PJU11(&p^?jLdu~wZAIZ{ zYwr4#yrR-II_}XN)>9rmWvO3QrrdTpz*RLHMy@T)gk2!s77kbc*tgi59tL~ zmK$t`{$}N=-C79nEV6mXv z8h3z4gP`cULa|P@%IW&qDO{uaB!IQ(Ln3mbmxea35l|8*Cj&<>e6>!G)mMKt%q~xH zYe>|Vn_t&1q4gw-wFk`XgoH1c;uNRTdA0@oDA$P$<6N0@+r#QCC!rj?JX{lijgqD?s_TSzsthZ!$Z1rK}H(l?4?5iOEVMOGoU7Rw-W$1A z$n4i)X4g1he}$T03K`;&`C>0%6iz<4^Wpd6(5<%AuJCP_3JY^FcX!W!M zPJS?)pQ3<@pYtqNtl;Ug8OBL{&Ju8gJT%-Z>$YC^k5SlULR^rfkOjgd2BK#}0^cQ4 z=rAxLG5j=OVSgPkq$to4-*W1)y2NxqS~3%*PzaP@8B(%@S*q-&?gAEL?s?A9qVMJy z1`&A7<4zGrcQ_4_NxKs3{s+i~KrD5DdS#AN^muD2_SC}!OJ>w)BRy1vlax+iA;BTY ziu*&J@I4g13KXOduU)HwWKqc6UgS_YsEfr9v`im+h+A5M>w6u*y_X18JGmgyF?h5O z05}D(ixm@fZVu@!bE0R*3{&1mh4x+KW(kJniY|HTd+QdnPd@C54f37<=vA`?0zK_$ zbWF`-`V=Ukq(L|q8=4-4dDAnY*aRXLJ*w-P9zmAHiioGsi`9}A3SZ4a1_lOk0LeAN zG=~nCn6sJQ?0m1^n`DJf53)fTQ4So4*rXK|39SwumwX^`Xdt<48OQPs5u-oqgS+R+IvyEY5H!NCZdoDzPFm- z=}ms*p$CMldF_CEBc)N+`}B}Y+e*;6b{xF#K_cl)sWx`&_TE(KMQZ4DnX7uSM%!k1 zt_nGvKx?C^qS~LacP2s6q$r$PuMnT~BCdSNHN=cQi?H5pxJcuMA0c(r_OoxnMm6^% zd8l1Hx$HoeH`|@`tKh=ciZ$R<(@Fa^Sv`kEK~xyX`7W-P$?kmDnB+bP`wY-!d1C?b z$0*13NnJodp&pP2s-l|iduukc)F8FBc{HAYH$Ux1%S&Uh+)8>fkxR%bjq^yHZd(zn z*PPMm?*oV-agUf`qc=Y?b%{1JRA}oR$Ljd;^P;Cq*t`7-g7Sct&jfPaP{Wm|{>g;6 za^|l1`M*G*_2Sm*L(@1z&)ED~@z>NGp~QP$bFl(u02^-C%Ipu)uhz)TA? z>&8$CzLG*=mF-hiHgxvO8J2L}%P*Io9|66I37m3wb^>HG*egvgRS!R_5Iz)%cu3as z_)S5sXV;-hT9DTj*eIJ~)kQJF^)aJ@tHptMZz4#&PJ9a>-SPPlEg2}H@a-+E+aVR-KLV6Y2Tqb3)hD> zX!Q3xRxz-}_=@!z=5DLSOpH}g2+QWvs_VC_3&AAa})=TCO6c`z^ zXisZ!J&2OnUYz!f z3|YMY{D;qze6=o|BrQ>WkdKt1abKjL*(=|@H|Qp@Q+6HKR|WK-ushnD8ZbnI2T^nm7>>oQeVicxbPyVL{f0V@PFasA%R{hk zgY{CIL%#5xY!oB4w15J){Dp!;ztw0K$gF^BFP4X;b&4dyOD z1(hdF@HmsR{kgXFPmJ~JVd@Rgp)S*DjgDT@Y5XN`%Y+$5Y(gsP(<1Wn1+z>`o_aU! z%9S4zL%~yKZtk~ezuO7&9tDQl91RK~*!9vf5Z49$SyLKoJEy`04nMP45Iu<$Ng2FC z*>p3JLQ<4OLP{Hw4y}+D0Xqt&s+C>=y^)3B|O17*gn;&N~ zm-u1vPdDN)F;>*??$ibb3|j8f=39zpfkM?t$3xw)7W71jG|e;fwo3KevoGlNs|&>< zRl2&hO`Ay&yCm)FL0i|CReU+713wnH_9jg`+4J!4$h*y4981WI%O2bD+IbUo6GPyv zGDM>{Bk+L$qppT>WYJ_%e?Mkqc>xHxwJ=VSh91-LEUh2qJ#`}@4q|b8NcI8J!l@UM z`85R)$cSqh02Fg`n7K{Hu9*4wNBd8^miIbb(CKueq&!Y6`rJgFCfn*F2)^+~?hajw z;uecl?;Bj|GS;2&1RK`8p%H<~ddQczm5&#Y0nVk9Rda4W+6DG}E}cPTCX z$reFaE&@{!m-U00nF3Lp`-OR(_)4R1Cs4@V+9cyQSsJ4*(ydB@!elO_tTsOz99lFV z8%}Ie42?A8)4f~kGJC9dbW&UVxY0ARo6CL1iy@=`oBBQop?ohE5R$X1Z6`}Su#@*I zSUDc5X-&qGbLvKXW;(f)l`iHu8_eV&koq~M#R-QPb@%e1y!Ce>6I^x!0^MQCHY?>c zF+NMI`E5^Kc+b{x#O_sSkS@3TLLPMA{NC4Gm_ecdzeDN8+i)t3;svLMJ6uZ`(x!Qu zWm(lr`*BllErFk(e@HPMn%=D0k5KIsZQfdzrfBWRE%;2O-QZcxapJ27v>H!`-!oL+ zn|mHy_9B?e|G^8vi3&ZTNv=%Rr?5)lBey;L z+7kCQ>yc{PH+wG-Cz|=+EGE5pBg8*dk!6o@v6uAMy|p+*E=PJ6@9^vX(4|%rHSOuD zT)C}Qcz2>`#Yiwq5J3B5vPF{^LnnJ>PZQ5-r`(XA5TQa0i0+1-@wq{1y0Ik#;}8>2 z0FJoMxHs9m{*EFfuyoT1{gZ(uuVmEX6Hb8$dZoVbTcE(`C3l=UJAQn!yWCmtPyy*< z+Mmno&l7jEVP<&bz>{1Bo`lU}`4?Wn_ajjcO#+HqEkfKYlmuw@99>WHW6q9W&GWD1 zW6|)S)d1DWnZaDx<5*6U0vKk!O+BrHLZKK$hCMbZHcYnr(IN{nm>ZTT{6dj4NV1j~ zffbX8(Wc<`#Y=91d9b<<2t*oTRyp}*139w8r$(oSxy;*yB3z*Y=AjfS2Iwn9j$E)Y z`sW!7<4H*tLs8R2LKR6aqeIa%jb76a#vV!{?8t9hd=pPzd44*8TX|dYFV6KWM09_a zoW@$-A}Og&fn|_B3~Du-l+RvyDO~B^Hnp|Errv&9?_mwo5!`sdqL%g=@qTS}vO!*t zLgYUDNA|qObQ5~18^W@B|LvF~Y?kiBS!-|7B0=^ScXJ>JC*fwLc85nzql&bM5Y5RO zY2Ymx=it7a<6pP6;&$0*dExzMD1+Z&z7_b~nffQ$IDJND$zGLDiuEQu5-wxj09jV< zh(D_G_RI#{1Mc;`v=ZYTmw`y1m>2@yTWjZE#+Bq1)we5Z@RQ9OeSNEtY#yUwIxO zs^(ers4+cX#l<~V|KKb$qPaU$^UUKxKP_!=q)mQ^$hf()0%5NA!~B<|Gcc$>;$kKJ z4hmvklkK?C(c=4+Olk+yv4Euqp5MPYfY#(?<xZ6pIPt1Q;APVo1SD(Z%QofgxcM%8h@Txm<2K)9B_jy9tV3r7V?6HE*t9l zwA>GYEX(WlAZX~<=uuucI|fUXxrQ#Y+epM?7qT9ErC6q*cLk>k3o?xt6aL zJ_Opsgd3xQc7Wq?Q2EqjP18>mL@qrZU4S4|KjzEpxJj#fQ!)|@SzvOpCX24cY10SA zo&9JG^cLN)#4hY)-I4No)0S^4a4a7g$744s1b8S{0C8dwVPS z){=+N%m)2fDz_q9Hk}2s^fBjNBwrmYDpy5ybQ?#Jy{a**ah*+_Z@B_-)zRRyGrgca zdy22PmB3G=gGExmSeo5!JDF+6fA%E~PQg7S zX6}vN1M)+zE1|Drt2Q3mQaN;TY<;>wIO|cdOYF6jnZVB&57Z}7RS4{ZokIFO@mbQy zRRTv|LDKDE({9(Nqc|NFK6AlIz!kc)UNasOhvebeD;t-eEw8`51ix@^Dk;^E8Y7)i z<}ELyM9L4dc!dR-Gs)yVt5>-qCtgR}m*S=Z$@spb z2Fbc3>zhSB)e{tBx)eMk9S{QyA@2u z-{+AB%H>QVp4;fSFJ+?grWD_VH@{wM9*@?K1lj;T+-dI2(k*62MvvWaBb z$lCW8?AZYsydqT~`=ct_lkDMMwA`%bU`op3CQre|B}4(Ck1W>W^f-W^rC(2!kw9Vz zQi;B=X?x`pDHhQGAcX$+y=B{-jg2pafxDQ48R+~gGRs3pLxv^*#+BhK!4*vI;CO%9 z3)-77>$I~-Dg*vo@j^yiIcy2ww=9IIV{9(pT@EOY<96K#bw>nZjzB#Q3Sl*BMJ#(I z58*fq5Vg;nDiwnqS!7waKu2r7&45^;YTqQY%ecK?Z@wDbDkVzvT|!1cb4lso%6D7G zVWu+{R&JuD`1Zm30t0<}I^28e@FS>6jheM_?0$r8w)@1P_g=YumrSG4iaL80b-vqp zufGg1?@%m_>_7sRAIP~wvt#L7a^Y!wuSs0Pl zK~E&Bmx;dJeAyF7Cx-BGg1~C#qj!h}5IiuVnR<1&p>iSFP52IXcE|2&IrN()O5aC! zr*XyW;z;Ogl+(yjVfyVffFo{O1Hg}732Llj^sj*E*9IUkL6dl+4#hwYvKjLvPI@7i z0^}z4k^lpD7O|82rG1C+B`EyV@m}?bhooiLdXmT5mCMH#a9R0u$xzZ2Dvim{5662Z z`HP{3!#o$mmq6W5)wc1(*+@?b;a1|h`K%fxcvd-Pi*K>^07-b$W{nFjWONCASZpJL zk)O49-i4ClJ to;*BBcn+|Cd9vtwCTUR_vILMx;B=UmOV6oEs@4L~s>M2ogJ}4{ z?6n+?;mLj}GM2u<(ZaTx=i&C_$aN@HC1JZ%M;I=0tn1ev8KIt1WHu=`m4uKLM!q=A zdU{4hEmO3x3}~2qvPws&v<}BB+i)+BRpE4=eYY?5Uyvtf4KiH?>VUT*~J)?K5Gf3wZjyts7hr*kXsV$^qqj6kTTyz z+d~Y*-=hVzYZZRk-8-&`is5BL0Da?86XbIin%X9OmiR;8gID4<3LYncwg>cjc;AMf zSGb4thbp6PDL>>k7^$+;PbXLSGK;iK#_|X0Z=25eJl1N_tyTsJXSQL17PAyiN1TsX zv3B)4YF}0oT_&%blveiEsrh}&wK62rzSl9$Q*v=RXOARU_AP&PlAr za7z)mpQM3tSZy$nqnwssW)<2+FaFlmPAt2dBRjL#57rl%X z`w1Dy!j%hVCvwk1>Ux*na4+T!Ll(t-b?*LgbDIHT_fdds@o>&xx6O3ZnA|D3NND+n zB+#}dMYQFjlW^_l(hlr|zVxGUmjZLY-~tI)TW`&_Nf0TXY4i{txO++167nDr-z%)t z(^t8=?|)f8d`{a#f9ND!&*VoyeH%r{RjwN$y{u%(t}Lwo9m??SE^auKLE#~1LP`6V zKu(aLM#?5#%lw|kSdQ9yJl#A9vF&)?I zJ890|TkF@#aW8%}-sK+IF7~odYo@-tHP6GhYdt|XRnOZu&ZEb3-2zLj%O{WR6};OP zzTshowhg^E%5UAIu05C)r~zc4=5)a%w#5E&o#2{Eqt%u6e}2eIJP_F1Us7?*T1b^yu{C^&lEy!5)^a~zi)cO zMljKAZOigX>#ZNh-(zI)>tm|oa|Pi3TJAT?KY{y#JBOJ_#pNESM8YQ*-Ud_f;8`l%P8?Qk+;-P};{`ag zs0s@+hrrX%SvbrD^96WcjBMEd8R%JNp{R6X^yutEPwKh=WC3mxg7pgKzfojdjsjH$ zuPjE{>&S11rUD=(IldUQL@WO^pC*y(ZsSS?5w%P)NV!tS>33C-nS-KhkJM%Ms2?tm z(d}GjTJltno$s0b9448%0C-AdmA+ISL&UHU17F~N`S7>{QXOZ5*$#lWTZA)NU z@TPEaom*nR^QG%25tphPYOOUA_b%LH%l`5P)^?yi6{Q~j_)w>0Aib3_KYa359kTEO6hdV1Mky({OEgptjd5y>mjMTW1>Ztf~^SLWPy-LK^+=~!dy#u5y z^&pr%f$uvMdI5*^&5hp&ClsXNK4yqjzjC`;AtOp`NL&4wO*6(Y%Yrc^{B08d1I1cX zpO32jpOtkr+P|+{Lt~2(oHEzI%MSZUdP}n=Xk2ht-fjI2vvG6FdV9Bxwl$(qvp98} z)3}`5)-3?x`aw`@a(>6^WfzoW*T6{<63r3W7*j0jsx9J)&}`hbM6daLa(VAk)#Wen zsvkF!Qd5xDWKkWA6sEy)eNRVrSQDfr12rUrUYA+7Kcv$@hnZ<8d~}z`)@DY*kpJea zq=aF}@~($7H|XiK=3hJF)GaU{%Y%%kcG*-?oi8XFlsk_@+lO-2BdwJUn!ztrDy;om z)vjh|t~`_I5c<+C3u9w;3bkWml#LGQj|t!?@Lp+lDz)E{8L4(Eds3t<%WaU4FGTV! zYHdk(q|j}{_tRdM=Uw5uBKmO^{Nig+8x%71oZ|F8;+oqoR3!`4w>pMZMK784M!f+T z!=4M=7E@Y~?WMn)`{{O1x+Gy+IZ$LwVr{LwJ3G(_oxRj~IC&3u@!7+t6qPut7R0Z_ zW-s1W%cnn##*tnyC|cFD*8?m=FkhvhyqA2XOVrUH3d?iieQ^G@$?q04#rtWXrFDF% zf*`Hh`0oz{)S;;X9NifM+AnYe4xJJ=CeXa^`cTkb#5xrDXE3iyf{fnJI`wh@?fktz zVfuo~d~cug!2Z|`fUnB@hOa^Z0)qkz#@pZI`prTE2^Hlj{{IjE$DUs44%^{0iA%uW z%$fg#Y#XV`ZhAQqu0{NnHK;=>m zh;i;;zWML_HJsP7bgY7cI*%{vD%@GVRAX=ah^`yFym^H3p~Ihn`#o4Cm;@%qkL#Oa zHG2tbBu#@k$~fWN2X&zAx+#QeFd?5Cve!d#HXIQrw2GO|cD5mdQWyC2~<9vPO({d3;^A?>O!0|4^Lf)bC zM{e+4#N|LSo~}9HJ}+wvtAB$|E)J}cs;x8Nu-?SjRI*+HAmx++xRq&HP_p*FnBTjY zIz`vH8AzR)fAHpB^?i<$0SMadktNvIZU*Fu#R+CZu~-is_!uO@BgUfH*xulne+^@~ zzl6Dn9_MJs!WZeoMTVV)hOVUgQvDbHPXO$K(t9y4#8LP^90M>hJdCb?Ro%_F2Ed5_ z`5M5SLJu%YuVo~1U-NZE-K`L=W7zzU1Nh4o_`W8&|NJJ`vJJfTCMJcG<83(&-+|XQ z{FncKEEcg8*SxVYJeUhBC#C5@F3Y@{@p9@D%m`^<|D5&l%kEj)SsG z3GrBgbCN3)uvw$e3Yn;ZF!(ns~mHI zeUwi}kP=pO?T=WUh->J<&l4(sT%N(h{HX!#%>H=3hZ{AW*hDxeqJcTA*B_l0bBZeN zE$j**C>4#WLD&_Vf6fra4MFimH{-?o0ZjQC5wov#_A3+uR}$Dk$Q`Ql3RzMwuia0G zCDCcVH{afR1Cv5RAwcE@%1%zP96bJj&VKzxq={bDK{A3zC8Ox?9ZgX9g9yW)J}mBf zS4K37AbP_S@1sL{I7Osf@N;8M#wyl$O*7Acp4*r|WqoK;Hg~*xD9+)tN{1pX7-**=^zl{1P)@xk4HeL2buEC7TwP@nU`9ED~dT^DuvtFG1m4kjX+d{Le;ebVN4rXG^j#Gk~K+4Do>mQN{WsW_g!{-|@8 zAJi|%1XWHsLD}9|5wDs{0O5tI-pwExaz}{=RbpowF*f{h9u((^0u^**K&jF>a0M9w zkcQGA6*vcgxM8TPU9q_ca$giVJ$z1cO2<$QaLTN#MNdHIv+XI)H1a zs9FLEgeZ^|Q9#w50X>F+A@zXIRg^d`sh~u$!c9P_>*k%I)NOFYi$sYn?O0Tk>3iDi zeFI0l4>+%RW5qS%#OT3f63h0Ig5V%-w8F{^AAo1THr~z%(@Bfp0{lo+mMxysB!EiD zwdehcs^6gYt7P{R6c-V7qp3Uqsvr={h0p$jjQ9|c5e57f%PsnYB=jA+ctV1Q>2k6_ zrpW(p1hL#9V3Y-0gTbiD$PyqFvhB|b4mSf@{7uBtGyq!%%$Jk?Vq_E7wMYJp8zBB{ z2jdqc;j+~Ctz;A2Ht~j7g?0mMli>WOa1Ibm`nHF{Qtw5%Kmb_X>hIU!n*HbsAGjocDYFhDTiMJPy-xd z!EYhLsr7NtpZ%aN?)xu$eymK_!PV(}G$|gTOKg|qp&*5>Yn zoQ8=Lv@L4^YGFUK=s{b9pQ~aD3kF6{)`vWBJ%E9ULqsR$Bj$ifaj-o#zpYe=xqNY6 z2&v+}b)dfN;x3soixJ7>oax0&dx#h*?q0DSArW?+64un_>pHd>HT0ui9@Pbfvid=J zy2Q6zHxE7K;Vi09)+KjOPa5<^6zXw!NNSP_h-{9yr`fLoz*u;i?He&xOw5!*r6H`} z4M6k7!Y$x6X)B*wR_$|OhubIx=h64*_==t(GUs_yoH68yTnjy1eNELRS5Ad0KDYOm zvL=-RMy^q0@Jf+)AY0{i?#=??`?jjIGX9!$@yY^%f_^Vt8(SRr+-aP@8xA@>0Jx^_ zba#2o~fH1eyz0P$#ovQYF_t=ys=)i%YAXFS-n64L!NkX_4*`( zssI^gO7tXEd3u}SVjsS(`jlxcqThr}|K77VfJJP!Fi-f9A}8&=GKI;9u~}>ZzTK7! zQ6SuJMBS{I_yHA5dln2O(B+Gu%L7BCe|H)B^;Zfeh8r5d_>OOb<(CZDeQTO9=UViBba7c3THkF!;{gb6aqoSD z;7qDC?fxfhmDed&1tk)s-resx($@_qubJoiX#db|FXZ@TxPQM?j31LgXycP35Nyo? z#jLYHMMn{8a0@jU0!jc_ECH>+u1Q+A+3>-WgJytTjs-o3D$7Bo%1r2F80vD#raf7d zTmmqw&Lj!wGcI!maGe|zpmNY4+=`D>+q$HDchqWltyKQMi|4O4TxZ9a5H4)&lPD*z zYLeV=`+7yvSkbf*6c_Iw(zZ(kgp5QWXdA6N&?`xsSUF-E|AzJM29W(#wD%nUnS`A( zF)^`}lOEC@5*(WXpxb8;7Y5hfIiQ+6s#}Q*^NvTq4$V9^SgTu$(YJK5`}FYh zCS~`GcmdljsQ4DZgLXaa6(`^~ZteQ|D)eXDI-YMSE{<3S?~Oz8e~vy*&~0CT-NeCD z-XlZS=!V>`#}r+K z*|(ry!@Vbjj=s22M=lB4 z8zTk?DOGW(uGB|@(yjgUUvZ4H;Xl!fzV=uDz*BT{!JUfp{t@G-g5@he_muy< zRS<%h$fpu~^DkHYr$@e5aJGYfP-PQcF9!YH6xb??|MT#?-{$W=jk`@;&NaIq?dTFH z<~K9CLCwN2mj{5~~7uq@O*E?eQzQA#ePUoR?fABva zJb&@^M2zT*hQ^N~sHwXWt<-E}(F6n|PyFsv~HsE;C(R0@SF>&X65V~(t5(abV z#z2xYx4`EHfnc#(Fir>>bi}oPbOjwFUF&?FkEY&sR?H3OJirLPx@h_2@iP7Y_89d8 zL6W}T1O)S86RWYNC_7i%Dzxn9jqb~?&Lhh{@r|Mo9C#Wk<=K`S5m zyvz(k?+0}4-{1T*DSwSMK1Q(<-0Hx@ac{f4PX6-M>xJMadIeq_f(Ys#qZIWpCB5&^ zHB*Jq0?%{?jlSx(^x)#|I|SlMP?hxn%VqNy!G0!Oo!pc>`-4 z(7EROHF*D8xGB2lroApJ8M$836>s^13lTAJ*Eb7ie?5~0D7&7|70xBlUCyAFA}EJL ze6fcZEMM3a;DT|;fmxeezv%VL8h)Ro&aDCB1i#sF5AX0_-N6ac0;Zi5vsYg`$7UJ! zX^{ks*JMe*!@rw`iGjco8|yX2&2jSm^bWH|!Z9ZD{CxJ0LFe@C4mm(}+^%1b6Yjq= z`mRs2t!ln;j~$&0;{upb?8;5tf0zxJBLfYLXkgNVhjYnaD9Erw?auwN=Sq$HFT=eJ z441XB`7gtTc_pAzUqo49B`i4sofC48Q_dB@4Uz}b^_@yx9Z2c?;|I#6z;ayaZ{_74 zxgp<0WL^i%elyGR-?Q2rE+(1Tx0P=r9?8}yarvrwl$d(@Z56_QybIM>Aw0)Bt)i^C z5dy5h`waC$V-iu$uZw2vt#y&#iNM$e+dGo{vHIqOPe5BP0o)(&4h#-P1JU>1niWA1 zul1XWAY-kk0NN^9WYb@9z$>ZIZ86oa52&TYHWoGz>>MxS8X2e8)(TR4ZZxsz`KbUk zgXD5Rvm%&NYC)Mc8fw{yL9&9}0)Gxh2A=prI|Hgq?9hQMsn2e-E4XJbgFib9WN-Ms zH-i$r&cKlX0MAu!pltm{M1SA_Kk=S~GfK61$z$qhGOXAXP`=1~2sArVU)7a#0rDF$MRmL0yM7qB^1%0~)j&egKu9`*u=i`3~Od54mi{GD-z==}+ zfv|%-P{P=yMK(1{S@=$yKxDgn_bHaEOlw-Ve$Ho(QX_!n8)U5pXE+tedGLZeUyP<{ z>}XZx`Ht}^2kb^6?S}mpe~c5F(KZO?8Is!M=~LN04zC|i=?7ly_ogJ&5#Y|fxC_GUsY~NREu}sM;#adKhuLWh-tC$TxldoM{b1Tl2Jwwba z z!n9Kl-5B_6Yqy!5zyZ|>?zQyB5njx`wx0fYEuPyc4y~LvGbzd2nb%ESaMXH>F*ix9 z9}@z*b^23|zwOkw8GqcdUw+9?^w{ta=#3|X)WUT6cAsoEs46IEQ#IkdRF-_Qdc#h$ zVY8XC4Ew?aINmM+4vhkc(-gQ&J6jt>PB%ld!Rd(b`%-#{z-?hVhFzkYKAG#@d$la%>!6IuYLhIb^@j)Z6<;6%!HoLd(fs+v!!Fd>8px6` zkC77^YZ<;F={q}ypJnKuZoxEK8?7{r$7)G=8&kZosGp#K07VK{V*rRrc#fu$a|BOa9?|#6fW^dG zQSrw|8N9{-@HWrGMM=@PyEThxH54}%J-}d;8X&Z%4%H4fL!pFS5J=Y;{gfAyKY7Ox zB6W68%`kvT3}b0VyKmr;bezcdHr`-`L|n?Wtwx22Mt@jJZH^e=qcGKXj?# zXf0oG8{l*mL0|>%2O=!-^?<4w=ajr)*RY+*o(YhVB}MFwM-%HRo{wp7XiyhCN^U$m zIjGu@psNNR$J8&`Z9aC%4m#4_8E_r-@YX3fip5U3kf<+Q0=Y00&48O6ML^?`LaY^p z_QFB9EnE*?rXafCMA!$&r#dcjb^bnFc7n_PGa*3Ay=ErK^{0P91CFMm$ieKLuSJ3> zzIIz9$;u`+-zn(rJ!!vcI5|7>e)zWebP%a1G~}UuK)B3a<8pa4dT4}O`0Q)K_D?S4 zmxcR3GR>f51-`S@F$=#7jVh3*uM5Y_c9rP%UA6L<~=b+i|B<>>oIwVkr<-6lrN6`>wt>3X%4scGhsx4-YQ(oMUA23^UDAL2nHDoqnd#45Xx`0-q)fzU-msHub!|-nXeB+anUAwl5!zo6r#)1v z0fi;{K#--2_)#}0W()#-DUfP&sATuf4c_Cxzh`><^-_0!af3ko#KANnbbrEm%w}mt z|IB^oxPkJDd&5qlojgbS^F*hiwnI`yZ2%%&DdW}iKvY<(nQb$gYC*efn;mv~Q0HCu zb&sj|E{b6NI-pNJdWR^8$mQmWYIElROwXMK6yV|x5x8+bYB*eug@af+sVgF2hIi6+X4N0LIU>ebbO4;F0?1)#jDmo(%;APc@X84? z`7B|+J275luNs19;yBD8O{0#Fn3o9xaRi_7Zl0Y?otZpv)cDp}uupRz@J0H?2*WHT zrhv$-iyjhLGd4-4?U`GCR3a+ZQMlDS(0?>#ftlo@bJD+@*^JoGZaDi?zKwSQUQZi^6 z&&DhaHppQ_^k%Okd)Hm+S9C`%?P=T9gOwP)KO=vfipbYLv4qqGXT3ICs?Ejed+vgT zSvy-H`}g6&;2p+hpc#$cH^efBHrLCRI?s8oWa;cAcS~Mb=Fg2EPlRm4;T_V!3qlVQRmbuMZ@VW4pM62`$YE z)Cj2z7^u>|Uh8H+O9aYQhm(5^5Z|ZXll2h&T~Tm~Fs79-IH)N0V#8;)3Ju9bBe?&= zhZxmSVgDO@*1G3{;Fm*9 zk=3^eVg^?*g0(v#!z-5$-v8NpeW&ps-Fhmh&52Pyy1`YJ~998lRR#1}=*JZo0&d}yuie~-%+7Yf~4QiR-7H@J4$L6HEKhcgvJTC9^< z`o}-XSWMp#Jb9L7HJ+qT>D^p%<%2{@Vlj*Yeam3&!k}o>2}e3BJI5IB5G{&iUFIaC z$Uhfl-E1Y|z-D?yGK{(`H2N{V(9yzKe(g8m?)Q}sP6YC=#k`MF(=n6KHE~#bFUneL z3oYE6>hzk++^(hFKILmLum}vn5;hbWy6SZpJ`DT#-5|TZKaZ=D+;p8v&)N6mj<8I@ z1eMj5;TVlj{!@*sj4{X8ZT)#dX9z`03Z##%{gdPIe4RmNW{j~|_s=OyoamiC-OK|a zWf9=gO;ZQOvcFk(z9JB~bp)9>_%p_|5(FZRVhp2jcGa{jGOcSHp=T0C<1eWhZDH%R z4Q2>_d)YlP6Tqu1^d=^hX_xK9>$=d(qMDTf9l)SAQLROu&w~;oo<^xySR?jx(dwqj z%_Rx zN&Q-eh@I`}QdKFB1acz2as6yAzPOP98MEa2UPVY=gkAP-RT^{S!gZrXp40CUXL2-) zp$b=CO8AR5e2ZX##Mw%s+V9q`MieU7gJY{{*|-tQl{aFwAQnSJ4B!28Fr&wLL>CQn zeFryS9I_46&LQNV4O=ek*rZFU`?~Nd+7(R!fmp0q*cE0^5)xJNf6m2T!G$wJ_VXD1 zvAwqEc#??HdAc-vjBwknY_~<+{~6ohtxc%Iy}(95!(TJPjXYO9K{QNL&!uui?Okagl{?OR|&QV{btV*)kZRPC@KmV zEW=?Bn|}buD4QtXzTYd2gVJ%$>cdnnDAmu=D=LUtT)q%HBm2FFBpDnrh~E94w?Y!N z`|p2U%T0S_P=~TKxsL1QCm)Oa=vyt_gwn$aH4Zx0r3}lOTjN3%*~>$M>ew|L1w?A@ zje&i-Usa4q^=pal4)2YeyLWq&f|j#Ln&Knp;vlR;NqKN@aj!ft#9p>_&d>enI`~NtW=HY=YJTucM@=$`yy7!pT@G-+ zTZVg81+vJC$;+l^O?xX@SR8tw^_6p)oU=!?16ESbBmU` zbn5sA#Et&Av^3eyY>|L&ctt89^YbT}cm6Z&oWg8pa!a;|cabIsneLcraoJre53D}) zJjV4=s#Wp!Q%>V@_L_-gHK4NrYzS>WWPX%Vf4P!JPWTJ0qlKcHxPfDVr;F&05k6ZRI?e>|ve{jcq)4b;! z$_`qbj>)ImXJ-#ZPt1_wFiDj$h8=`Omk^HE)tj0Mu#-~`DcncjEbb8xK*qVIbu#G# zC+W9!w#ZomD46KkPyPvp9=Arl%PN=~OblD&TR&;wx)AIZaC~fCU!T_ttq2nCnE8pz zT0)u`4Q94;6h&UBi1Se^R;_XMJFQsr4}l=SGQIAf>*7M!2f2-9jtL^va=Hx4^~{gO4riZV^04F)N{U^falIMGV>cjt}jgl8`V>pBOI3 z2b6}1XgCC#YMkd4LJu+K|6)2zKi0s1-Tnex@?j z*IdUnrL<-PaUtWvYFuKoV4xCBs$?D72tEXQRgz5~XggdIEvt10jdF~4|H|=Pk(w14 zM9Pc9D8R{Hf7n#w+sTeFfEad87qBo|G2`I82BhKBe0#QKm3@89S9G9ew47lN(9T0J zDplC4Sx>_BdCBnG`=KaZSuny~jNWHyHW~*xAKFN?&)_V2zDDiufdj=p5S`;Xzi^q~ z_FldD<}ZO-{GW&Qtr=FIvp2_Y^NurPAVN+O8^(-|lD|xOyr(-G%;n_lfIz zp0u)!RxCaU$avWJ*dMMJrNg-?C}<00rEE~R!+p|e%Y9hYlmzF-%rZ8WX( zAA&fVQXlM8qW02q{kNU0~r92gtfS0S8t2@L(xql!p&{h3Je4;K!V?20r(!HD1>2 zV=*wYnN6_eJvc3cpQ#>g5RbITnmDndA{#$=WHWg<7%{3`%tj`N_^xwJGyJNHKrkIjhCb*9w`EQ0g4bp~=lK{&!fM(gc2jvP1I8SJ zBvW-+TP`%rYTmHwX8<`F!HiL)<63WOhCNo2$W})TYe<^tnKm7#9{bJ)O1HhORb)(s z{TNWiYmG-2&ozveOq<0_Qb#9CxXU&wuppDS%IT**LB6A|2~Qi(R@07!SurEJv`JuHxk*?>VjDZGNSCR?0}iBN+4v3z-(AthD^U!LL=$6xqy_c#NQ zxtBV_F7|czZeO-y#4e{7$;oI#*MO<$yu*sq3~=M^nVFm1xR|%d?B${;=W;EcBFJ1~ zhlh};PP1S4R4kv-#+z3CVi_SE^5uO(6S`1}nwE#VlK0rgYu^Og=jEmegp!U;+Rv@A z8~BDk|{(RaRoMRs=L z{kE@Uv6)ZN9V07U+nHGXnW1_%m%0Q;_+(g~G?X5FPHCY(VJ{kzh*v!@j z`2NXd5kk_(NLR}!oo;HJWZslT&aQ#|t_;16e0P~8uVON4H!5(?fRqlqG|~%S3__u{ zGxr!k)9~ppct4>Nw(3v7#hXrz#nN8PqGq5*7jfQsh42o`>QU zQ(gd&^6pJwQYhP*grkC7;Yc2~%xD_&zaDr#UjI@l?-PX9*^{OD#L2|kJX%e~t+_wH zxg?rM>Fw@W@a|HL2SlxgJR`9Pcok9~13OI9ko(#LnCwnBNZ&1#ME zE_4}H6*sdYZNLLLbkDpi&XxYDyYp-Ps3%~`DSi|6?dz4nr8CaJtAqOXpikDe**K`n z$zb_a?aOCmAZ?jFb@UWEdfrv1r=d*Zuu_oz1XjlHHF5j=vM}NC<3Y2PUvfU5VtvTF z4bBdE1CJE(w_#RseRNbg(Hs{ljS1~QLNa%{2`<3yyiJifQyymZWJ;oXPnxW}ni7%4 z22W|t3-tVpQZvUZS{u@8ddx!HPp)yLREFH0G=29&(keV}cwVeX?e!Bq;~FaZsOeKk zPj{TX2H7JqUB(8|%>iuPh0v{RDz`qdo;f{nsA2w7lSfy%lWUukIbx)7x7(MFl$gt3;VaB8n5$$T`{c^l^K{B(LSmQ}^>aq4C@we>H z3oz{}sHSlrjGn&oxbGin#T+R-ZACP9)CYL5zF*mhkYneVw=SAiZMR?3^)C+Fb03kU ziwkp!LmGZ#F*_?hRB9j9>QX^+#@(cYc^5l*&==vWz7alWwgK+_!AWM+D~U(;O0Qo# zBAQ(?oYv7$Q=(7vS(n$?cLE=`NFOlHkj>z6NWz#&3AX&Ay)d&>vz=5rS*KqYN7j75 zsgG`+ox9rkSWU?}H#NjxG5v+?fYuw}Yb=o@M$;W^><|_7v#HEN(hk}VgE(z>G%;XA zt);8YHavE=WvunhdHT53QGbfr!imcL{43EkwXF3t+w)Xe&-#P+cOhI|B{AKrcQke! z*+`zU_czxIb)0(ge2oDltoFLm5f5O!)%5!LQ1PQDulPw{Pv8+4V?OIX!*=jv?2V#9nL1n2kyQjGv!v3P${Lv(PM!syE#z_EhpvE5~?f z`YP`noGk=&KVNNUTl>RI;1QK0v7WRkk6R!oa}9Z^H%ji~N>XIWc|5Se57|MZ(jZHf zgh-i`R(vsi)5v2^gM*fTOL!qF!G^jbY9o)Iu1Te3B<4xsMj4z5lN7R9XX7bs+6+LEYuV6Vwf=~%iTi8ppwq$+hyQtdslVtc+0M|T|( zqh;)58H3n(p&x7hFKPOAV+dyuNW*a*FhgF85%6tZj$^m>Pd!1liJ2WG15xinCX037$VK@&KrRyY&iqcEYGeUqebp_+y$tfy zdJYCC29Np1E&6fe^3{adt1{m(=r~thP*EL;eaOzan?ci)(_g;~9$H%2z9aF;LXLl1 zZ4wGg9Jyb60n=t)jB5L}=->B%0|5g2cNB#dcecA&n!zm0X}4G&l@H1hlZ3TX!wu;# zi)=I2E=Xw=YFno|(|w7?FyQmi_T~_(UNNud%wepxVBg@5CE5^bkTFi~8#YU$UKzs0 zpdSe6m@SQyj{^-;9B_D)ApaV;HUBf+!;qybHTjcF(f>VEp^y4@>ibIVS;Y7TH+}Ru z^e24<#e)iZO;_G!tEzTi(9lad1=)L-&dDQi0i(vgpeujrj0&MC3bFJU=x#~7#eRpMnr(_3afR#7I32ICe!r0O06 z*XT&_Jn@HI0x#It*VnS?ppBqLD&;iMvr3R9Q5T+agWbUFk5e&d?Mf5u-+ zo4{`55eR}(2|ML42&-IP;;eyk;ez<266&bpw@nysgkiF*kZ%%O8Z~OuUqDdcdE~UT zu!v+FEUZ87A(YwrwfC1KUkgQ+?sXSgCwgnu`IY`VKSuoC133R!fGae)UK{u$d5*x; z7|N*?Hi&k~)dMgm^r+5N6)&+_qMLp&tG{<4%2MHv7gV+LrR0A?a!D=23^(x@7Ed8V z*ji((s3*X&CS;={WE30T)EFLt4je_)k(;J-6}+-?q}+FH;b7HEgqXH&ms6dER1dq!Cd;ZOvB7MDXj^34m;S zJUaLGKXy^bx}f4u8qXRe8L=o5kdf2P0szTQTCql1GzMbUL*)~ASMbX+)8aZr53_^n zuz0xIEQR}zk;1%PX6+@;NHQo9!_OMpizX^)k7wZKoG9cFdY(Mmie`Nct0L3lF!-mE z$Hf25GI=XGJ*#BLNGs4dJ_h*lZ*d5RVn^ue{df3S>|m)Q9e0zVg4pjz{^4CgK;zlN z;A>c9F9Gm=OHL_`;o^Hh@P*e2;6|Psc|JLi>o2W~+(X#OZe+>G_?HD&xS(!SIe0j=|M+I(CRCiBurc7b` zR)%-$aGnN`z4ftEW@cu{eRZF&ffPotYzRPp?^X1`eVPsgkV%`bB=KeEca<`a`MI38 zZf7HQx@B;#o}mGip8&ZfVHrH2`cvVhK42>w@>`$XYhwjvt?X1G8-#Jv(bP^X*|6gA zK=^iJ_DnU+FeWmT-Egv2PBpq8wz#x$3gTkX*kUs5J(O#$&od}HPP@Y+TcGxwO|!0_=UY{L%Zg zZff`Fm1Lv1SjF26$Z+85+3zHDG~lJzJ)E9im~bPm?$OghhT>EaBMW zz!o!-%FI7XYywI`NzWSW5}*P?@VFjP>dfeYya)doEB`(|0dW=%A&)2y}XI>BqwTd3&!ZehG&b{2C(u!AtT>uB` z6Yb4>Ay@CI9WJ$adD(K!cMZ>?JV^uwQ0VXHAtOG}SY2V3!2+ zFF%9O)WOWp#s^0D2ExZ?xU>V!6G;es!z{qY(0tDb#~F>|GvGRxaJvzH1{SR4J%U>^ z{2UG}Zc8G8=MCJr5~N@iA*c;5b3x&~AsnfN#wS4jP7vf@LE~>UK7xEJw2+iu>eE}e z{M9|S%U}syCRDwIUoGI$jxfpS#uMo0zkU99p8u7NfA`?O=f=P1^S?Uf|66wptXxG$ Z*w5rgFyGz0g$e%YX&pL{weR?a{{o&+-^Ty| delta 82222 zcmb5WcQ{<{+BVEY7eNGxE+HaX^d3a?1c~T9qPHN7zKBGMZZO&yf~cb#(Iuky-b?h} z8GXLxx1YWDv&->*@B7aj!#ZZI`>OYKU)OoAp-#d9G+`0XTOeljS_|?yABFdJ?Y(cU zrai$QOG#5c*M@x3qnsEk7L_W#m-6-fw`@JxS@m1^Twgy2K@#VuK# z^rm!d?2QMo?dK|a=r?oH{f4clF4JkZ7Bg+=&^hN67ll9=Po^u8Z}Y3Ds5EKFT)gYu zI~-Gg_Dm>(QS2>X)@w2PS@s8{#-cHHz9yGqmeG)UEk^$i?sRc%-@FkX#b;XF2osx5 zC9g_ZJPO&=>w(Kiz$BQQnRmRuQO45B>MLbUOgBQ(2YPWfgN-wEs%)CTH;E8GS%57b z3}}c(oLpdLy5f+M9vhEFkUm@DcSIcGF}^eu*`WcHMsT3CTSCqM_|K=*F`JGBT~oiI zYK`i$_0p3LzIx9J_x0iDp)6SkTT_NQC!Zo6BW+RkN3;%hW40xAdpx} zaM5?9yTi~p+oLuiS|*nGo-H~$k^7XgEGBqF0nHtVB?<6*cEj&4REv{tJ;IoB+W|$V zs@40eK!0Wkw`rY`iT{aZ<~Oq^ zE)l$k+^JM_j4r<^1)Qg)_Ko~R2(RBZ^Ke0myM6cF?-Jb`(lu34R1D^EX3UjCI@&7} zOsFHN5n^pN#1Ee*3hj=y--Y}k-3cd~2^UP~0~#bnPM2cz#IN+dKkoqQTQJbmcCFq0 zZU7GC5;FdL%BI%V<3yO?6W?^DTNakpvcP$)By*HVmRi(qu(h-CxaZSJ&Qz#|*q;xt zEf~HF^jLF1sHq>lkDs>Q3=^HsURiBw)=O*hJ3Bh`aIZ)iBOs%@MQ<@lLvQg>M@$1y zmrWG{pPkuOR9i)|6OJl?27BXE%`(zI`{+o-5+J(@DZ>dX4G@Kg;6MoRwyM^7y;jmb zb$$seOY+GsHK8wjRoaB$H~A)7`jM125w$1bv0<|Wi;G*DYJotaD0*3`6GfeQ#l^*! z78J+V?hU>PC0;``sP*tS69^lAT?0NSjg!?WgwUi}HMtOd;0D$4=HZQe0?;4xem+kn zp}q;)mX-I0A`ULz2b>4wxDcY>DA@_Lqk?sZ)cHgJ~htNrx=HgeBzyB4z$NXYc47!X1%x=GK zlRa3KAP%%4&etpcW%w$)XM?M-<~5qexX3Iplu>fn=P@af7*B)bu&w!CMA_AyL!yx| z@R5i?c4yGk+PDBDwbhAh|fWy?V8!Mzj#37|zI)6yB*Scfp9hT^T<^z(2-)Zs$*9ZMFM%&zq3Go4vy= zI*CHI(dBweSsoDs`g6joZ^_y4zOjRrG3bw>;skFS70rgUA=(KK3l^{hT_m5F0w=6o zvi%VJ-U`V|@y^AJt`}6RzjICBi&l4yNltcPLYp@R@7=JLflyM3EyY`~2?`E$I-ei! z0s_|E$&q_o)3w!J$?CotR1cykKF3|fv5J;D8=hy2;*%(;Z=B%66USpd;y{X7WO0OO zw(gFA(P2&;dYk_V7@nPOG8YG0mDtO(jxK9Zwr_%V_9X7Z$F72lICxJevULknCOzlF zmZDCt-=Gsx9xgOB-Ad3|U-H--2fVIjFQ)v)C3-3g>J5*g?XDeYo^-|Ma+Fsd!Kz#$ zhZFdNk_CrOv?7sI0S6bry|7sR^NIP$iAmfwBa}|Ky_(74CzL_KUaHrurI-wt1J9}U zP^ZsnGg0W+t6znqtv!}YjmK`jIc#sYg*oZWT}l(w^DXe^&mG&p0w1ot;z6`h*_Etn ziLMl~EIL;?s~xc*0e|tRc6({h~p+UBcqU>3X3n7_RM|34+(!M`oABmy_=qq8*~>gzY|~ zZ?Upc0FIzN`)IOi+Ch7wLYbCzK7tF%)y#T^NAWf-NKUqz z%3gN=aKseDG*x6SFMhUB+ND3^p5J3T@I4`rLpoIr&ijpu!}84=&j@SWKpT?I6zPc# zh;x|y`mX0fc6xa9&mcmz&GB-}wrKXfjZ&be9ql;LX?L+KgN^Fot>#YD&fvzY0c}uw zi~GmZ0BF9N)2$&PFrn|Uuz)sC1AUFG>q|PQnj3gRIJq+!%S%gf-sKl(82E^ZPuMUr zlce~b0fg>c3VXi47{XWgpu9Cv*uGt`M@oLD^cAbeQ!UWYLgk!!CE!E20It@PJ}`9X z9bMYxad0b$*`}@8Dzp<%uhOqN(ZSk~8UdKz(TM{xzP{Je#;1&{>+K)4C5p9uk%;7g zq>!wfZtuIM48&b{ZfgUTSG6w45hPxaS_MkMIq0i{HRi0@p9muC_#Z^(1r6 z5;r4Iu=(G&0&oBbFKsG*c84>|kP;5AHJ(tU5V)(16l(Jwxr2s!T|N8vvqk?qhchXcfRFq?Skm=p(>Z(&c z+ZuB^q!YrNE*jSXccH_jx|4jf*F0WbVwFwsc=gqtg%H^c&{-8#qkUmfN>mJD6!F6L zaSR@#=6m4&{rl&T>z987ENoYpicyTa$bN+Pj>Nlc@xG-J%Z?9~HyH>x&OTpdklQd> zL283yuD9Su+p5M_%O){49LXxb09G;~Qz47(uT=wZY@SW2^Ah}R>Hl0g=&FilJt@+A z02TuX_?^DqZ>`PZtJ3HF6*lGWfp+^fw z_Bu3vJ>*Vc&r)G5ejfC9=iyt{VxHOE&x75nu{Xj>GZYP69Zb1(X{jKU_-AbOMapX# z>-%<64QzP1#Q`3YlPMLt0?}0LUB|4fnp60D7;07Z%f~sesJl!TN)TdPS|Xv`goiwK zpQfw$0Z~y6^|Ka*688?8mcoMUB25uzBc58H!x*%q!Fc=nRCR@s|H&fSG0KSb_>Ri+ zy|3TF+-+m0MFWYY@HE?ljM<;Ns6S~SgG?C?Mb(Qs^3r=Ohpdnj8*!Tnp$j!1K6r3)a|o8q`9
    z$JVIq!I>@mJSjzNiz>1W?2vK&?DtqnGLN>{ z7tjB<>10_%@T62$qhIBPrPFG~$G;|!EwvkggK1CPfF|J{1=p1WazUfl;rwu1T0$If zow#fcfBiwMca@m`pRU1*38|Ey(6&w^XitAqO1MI|Qiv9xiV*#-1NA&w-(tiH0jnNx z$N%St*aJ;Hy#=$apQXGHAKs`7LJ_f0uteS}`bY-*BYKOEm2@yi+X7q6L>V#Q*cB=e z8lx>}0VHT+Tt>}5skrb5j^KuL>%h{`80b0TUuuQ@IO-n2I_1B~)LHSy?`-=1OO zxnO-&)BuYUKDc9SqLzRCBd-6h;RX~$L7}s5YuHDrHk}XLJHcY}^S_99B`7~zk<&an z+qA3OSo{Sa3_|%fPd3_+OTG3#?uQe&#-dqo5~m?7^|+&($0{RB9IhxXeviTe%q$ig z!LH)}OLKqIA%$^Z%)&x6M4D-rjZMPfM~hZ}f~c@1kPsfHxJ2|j7{_?lV*ar!HKn z&Sy5U$zkrF4;q!r?X9->VB}!ErU5>|>&b=^|KG#opMmn^){bW~S%7y28_;tpYx8+% z?h)RcLBrTg`;YI7bCjSH(h%o@Hq$v|O!Wn=O$);fU?SG$k;N&?-Nspj^E2R5T?Z8z zaHUnobasm?k1)-)13g09kMIHWe+Xs7Q{=b2l|3Tk12q}!$dJXGd?<@Eh=PWD-}_f* z21~SZ3T0K*P;cp*R79&!84(a;uE>S)L@gIsA1M_r9noNZl*vod2&LciZ?l`UZXS1b zD(4ZvvIE$NiR3|vZ`&ll|4Z}h0WA2}!3CMaC<4Wa9)a6Ha9}ehj9-L`P@Ciy4Yjr< z%s!AAaDv#2Iz)KKmOvL3YM+rj1szhw2_<%TRnY%aVSjzg4U!!Qc0bZ>yf?@1pZx4> zzc3+T=?&*XnC!Ew{UK+ft?8dp`9I%e%kXdBRDGX9m-I~eExHtC4d4igi}R1eRlJCP z=roKG%&3l(XuA9@qVTUG>Md<0!PTF{M4*a<9+1*g!qNn3h-t2hu3yFrbg&l$G(;@f zV7R7oJAmOu`#_HZ3^r0|l>I*v;olb?_>ih~LW|dSCPr^rJd(fsW06+ZCd?tC>@>P(#1$gFd)IT1v}k`zyFU|)#je%f1euuJ^CEM z+tjDV;)eH)TbB0tL68;$n6+S8K~w}!uUzv0)(~`u0KA4H)DD{oqK5Sf;t}>yfucy; z!{5^3-Uc3!f?5?2YhwL(7r}wVQa>ngAV`joWw_FOdQw;BRXy}Cp$*m^BF+=kC(kxx1u^rINq`27x3n`4;U+O zaAa{*p31%YuMZOFf2k^P==mEq>yX_9i?oT0b&9=n(!r{P_gX!GCY2wmR>^W@Q^F$9 ziUv1*VUe}81f5q)uY~>iz=BH!!xf_?RtL4xIrWm42!=~YL&yy}LmVFR%3+mngGsW< zJXyO^>u8HB--sYN?EftIu=DhtzXXS>`->|Tfu%8bAe()l2JDaEt$SZdh3xfwuqyv+ zL%8P$B=~}%=LM>#gX*;mEU=Xd2o7k`x*0Qw#fUNDOquGu{x%ReEWjIYO!VVfzzhw1 zs_1R5&o5#QW$@&X>9xOvn}t~&XqgU>xq7;Ram**+4V)=31a>_*S$6l zdTjB2t6PsW?YwUL;(syg7U;=q;b*UM44kH%sJnpT==O+4km>on=QEtdc7wzpDAYb3V~U)f!NN1ToX>y$gO zC6gwtABd?Is@4nK&MY+nSfCh{^wL$3P4b{d@on>(CMS(l)YWf@xq!vkwBxYn{b{l@ z+IEQ;Lb=;)#^vmU*15@UNGsgDdF>kJ z&+A(o;}0ijyW*1F)jE;2)fH-XQ}?6$vQa|fuOl;t$Qk4-YDz9W_iv1q&*^TP6HXm` zXBz>s*Nv@N(Rd`7O&GRT#w4y2$p!G3ut=hv;&d3Rl_z+m7eB+gV?3sAd>zH_xo{2c zJaH&-(5rnSQlM|^`z7}7bAvlQxw~$Y4f`4LJ=2SEnWp2Nr^nI>qGxVnMQeOxWmd6o z20Nphc_x%Yjt~oFA;HQ)-+9`*K3$vSDslzd?w*_uU@R+lYYNUSq6YJ|l*T<8vGy1k*WO$G0k`bh%Ulvns^&7JE4Q&cFN>28e=HaN*e_E21~g;s z+2bKKmc(s6KL8N9)Yh+C(cPD>cAK!;_ZuDbY+2iGQCT+T<+Qc6gzSd%z&Ln4_QHX% zkcm!(y8!_qcv4M@t}py@>~2E{nSG1S!Gb?JCWg%ivqdKUNp#v%p11bk<-R9k(0a7U zoE1smLAX}fSw3RI#aDOyO|SkjLJ}HBr6Khq#qUUezyDAx;>(*gpY2ziU(;uXdG@g! zTv<{aO&@Gdy5f;FA#~6 zgm=YeI?vcoP-)=yrl8i_zYvO+8Ux&g8Tg2RHg$lWb6QpTXBo-ZGJ&9jeH zRHP1m6ZSmoIZF|^cAUxd`-U#b({;u?vF*B@56`^TsX0eWvtH$$K2N-}eQ>teYjH*> z9Sn8QjQcU(Dk&o31pUh2c&ZOb%VRMcei5|~Ff;yPY(oyb35{kkazTOTDmg@ooN#uF}l~nxa4G>d_CxQil%n;Kg>|R7V~z$Rbze#C(7qq4FL z+*TzfFSZrtN@2S}3Nj0OYe2qNlxJJlJ-IPD1Qrsg%0QFNuB%9 z*s^owB8}HNs^b0diwKm(yGnX#?=mk_-=6X~!!jFI`{8_kjD15HfW6H0klY@~itySt zsBm&bFzxn$du^#OYIr`y2iS_Lzu0uPx>JW4{z6l_+Yb=}&InF6n}Ab%8R?hl3c77G z#d1e<#zTHu2qz`A>Qblusat!lM|)l+Z)z?ue5{<>c4ZrOcuz~ioG`%nUcaAonA3)e zZ323;j?23{ni$hsm)roo&LiuW`P$lRLmt|?tT+Old6j2!&33#BK%BI=6!#oA5P$iy zw=yyC%P#F`rJT0y+sTPZf7;v z8r>K!qc1{dymnz&e4l=1Q0C$fch1O<9A^>k>Q}PWQ4z;2yXe_0f7_8_UFI=|jeT-Y zrc2Dg;;B*!`MTXbV7JpQ)%Li-!trpZ-G+1s-1FMWk-+_G4~NpBLhP*CLdgVEk)uUZ zEu_!8<#M+p!WA1E6uK=%9=cjkPrq4(i7q>MrhhO{@$$_%eXppkT5rq!?ntnWTASq6 z;>|hd8VGf8m^%83#_?OL9C`EG9j^9LhqP@&sP{$adz2zT+PgTA@qDDN#P<(0mUn%t zdOHr>etH>H6@y@O=gFeZDV_2Bjmhgz=-H*g@c{KhCG6wjV1x0u%GmERhl5(|SB+=4 zpYQRl=22HI%dozOG5m!0D89=Yh2iG@lywxgDW6=Qs3>HBPP%-Xk+pfW2i6r)juYs% zETsQK0B+>^=>q?E4ZjYHhP-kufoFv+P(rRv@4-KrUx)e{Y=-4B)eNtxsf zVw&M%b3|?3-6H$PU<*Go+m$<3*qaW0^=Y2%!=SDVb%OBK{qYc#& zfQ*b0XLF~SlVy#PK_V-I92?WR4WrZa!F8AX^0qU5qgM~;>%2^dyXc(HcguRJ5g$is zdgD|~pZT|VM73b&A(iJxgkk)E9aUGr9D)|XD?Y~F$hE@UgRB!*44yVEcJ8lmp1w_S zJ{YiAM3}>?_HFBY`#ap*T#v7asdZBP=RrFnV{R zjCu+|o3hkp;p|^`r}bi@?OM%*-rK^4vI^5ue5wIK`_@*``)d1W#x5D)ZL`cp>N7@} zVV}Smb?npb>a8P^2txWCYaPSU6X-zwax5mJgOwdiR(CKQC2$&UzBk`#J1v1-!zcto z2}uj6J`YRknAfH31@au{Sjv&c!DC|!U}|@Q^~@g zsGN#c9|;Tr>1#EtpcTYO{&Bo=vz#nEw@DE8dj%W5ec=6Zt)-2m(8LY;w37!Bb`{^q8ypzq31?zlS-KVJK zb+7SqL%{3M@Vetgj!H!RpJ4jV#r{i;ASB&`A=+?)=jH0-FI^YgW39}FB{U4FPC0ng z)>ApTxH|IJGgM%m@J>I>6o-Ud1Oe6IM;J_Nuce6k_^_t~Fb?`9QI|;&hZUInnTz?UJHKG58!XY)(T4M8UlI{TD$8vxJ2!FJglAzq$su> za5_vr8~lCd_(C=)i)6e0*$fXA=)+Z@lgM8pJ+EiwsHxO3VPxC=8@OwRF)@Y=-2OB$ zFM1M%6-QrJ)gP<1Dqe6lwx5{DpJ|$&&DsL8hqfr%Z2G6A?z4(_#S`Ux>|3mRvqCaM zC3+#G?eAm$Isamg>^-vXkxS8eV`q;8&-`SGVTeaumpK4%lSA25T(>`^iU2K@fp)TwbZY?n1Zc(qcax$vz*I`Bg z8_gvCYd0KUjJMg4tt4GOw8+<-nb#|rhMb09Tmo7Su~JK+Zk@APPJ;mR#r#@Q=dP(A z`6+nCPgmX`d{D4@F5)A4&j7l$MdrU2&Ne^u=7P}JK~j-}FrB_8>~}dAn6@AG{$gGl zO+J{i!IyDYLRB4F&c=>Rg!fU;V1Rquqc|x)#=&`9ffMwZasH9SP)1lzSw+5cKfxPo zgnL}&Dp+e_rnHwzE*%TWO5gqXDLUgjKz}|6JJ|NvbkD6-q2)7iXqIj73`Gzd93j)2`69?;9AP*B&YSqg1qNhn0bM7?Yp)Dr z>=LECO2!@zV#Vx@vAw-6Jk=hXc8&rv!VS5BKVxEUfvHdBl#oxDD4T~Hyfgo(!^XI_ zY5j|wuNizLXWR8N+y+UN_5m+V-(IdVV~;=GIUd~fmX`l?gY1XSjGItHe%Ch2EUR3@ zu=UhqVr2GVn;*mnT|m47%$bzBYfOA(`)(CK^~X=tiVxca70$3k>`sZYe;c#^(|JO#`>RJzo^gGeXDHUm#%}mKxb^=4u(p0-I%#z(hRJqn zQe3vFf>T*}-h>sI8_&z468EUMBJ`^rnHeFm@8fI=E$`}1|K(I!(?x%*mMfwt|GmSRi{bI7@ zPXj!s#=_5X&N%^qI-F1K>-~5gCa4EU!f7OXS+egl)@k^-21l1cv+gruw0s;-IS|)Z zI|X}$MD1&WKFA8;&E3a2y_$WdjH@bM3(jO`oNxsvmagDDe!Qw0I2yBF;uN+Y7P|H} z6H)k6jWMfVwNCDpebV7rp{Yz?T-48K0xeWScK2k#43MIyk`!vl$9bW+xRzIl_Q3bG zE7H7%>Qk2}h6&Ht5Jb=m+r6@*F3ZgE*dNK@Q7q5%u~n(QfX9_3)y3wURg`yUN#Y&d zh7q1*l~n}_@_ipFCvS|b@XEy6E*301={ZT>Pj@j?d1NqjlC=Sb03e}JWRZfbNC0|zrT|n^m? zv1Z$<)5R{B+dsD6yWhQQ&r5+|lhBj6Jf83B@;Tbz|AZp-2*rwQ8t3J>Eq7m5*Q^ge zW8IN>#|#1Twl%0+n!aeMF5x;<8MG&`hidZymhRs@Jb2<-@Or+(M)6*L#VA_|E zk6}}Jv~4Loc>o*AgcR)f{Jg!tFQ$^?HH+!<_*VY9$zBHH6QImhacZ|-z4mmi}6%u?K@q)+v{ z;o{*cG}^J|DZ*S5YzbU?^e2>9oxHRd#)z^zV|dr!+5$FZo@+kmoYU?!VAIA)Vb)Ez zC;oSn%!y2k1b=imPg2~AQ@ppAUV7`XepZGaD-Mr?l7?e^`(mB(WRSJ<2CG0fyW+^j z4FT(MrMk05VtGrIe$B!2E3W8TZtWWqB6HOy+{^~=W3Wm|*Kp{c# zMaPCwygOAdu#`6J2$nY0t|VrgF$>3i?4ZRrfDs3ko#2e|MEG**^~0U?c~sQ)N#bZH zaWj(ZbGoP91P8;w<52l(ydIwp{d&CONH}aFL#?aqf5~X6RJ~j6}Fk; z_g9;pXW+xu*Wu%grz=YVgGD{7{X4x{Y~|{(06&M%#<*leCj`QWs(vl~2eo^*ECi~6 znkhm^s`&Vnz%Lb!xp*+KttEM1ppe_SmA->H)@}W!fWS3Gm7I0&@E7TW78&;yug}+k zJxjM_1*~6wwT2m{yJ4AkRFK{wNqvD@9jA;tQxiExsSy!NSU|Ud*u1b#r_R1S&WbK} zVUF>?MsT+Mdg>!&Xp_;<=AglQKL>~sX$z&(?+5!qXvZ20M+TSjUw>_w<&~q!#uJun zgE{C^8;frxbT>|z91ewsSeNouzWQPeJI3YR8-vMMp@AbzDH6_K)b)%EO78OxGZoN1 z;K?)BweQHfhj`{Ck8+3rlpEt(y`s=V5%$lNs=6CtmIr*hwIK8Kd{->fI1<22o%3QE?naCr@asg&fBoGJc)jek%)5Vu} zcZ;{MoqOS75ocRIvGnLsdcEVVkONn9CzVk1fi?DT(!73vH)hLtsfMo3y}0>ca5~2F z0_Av!F1xRH)OByPL|DA+#uNrTYZ5T*fz7TjHjA8!*Vt?S|}^ffIksq=cs z$9z<6m}=u(w+WF_2}H>p2H@4}&$=gJE2~gtui~AVWx6q9h<79Z*%`BN zo^>d{K{x^W<)Jik@tE?#qh`51jqzA1J=HzZ2ezWqfO4j`8UQE!lx)9_F7PrHI%l$Y z5;5EOvc{#6F?fv-Co1(rhz$+tH^|3a0WPQseyH-w*z3Z&ohRl|gC2!`p z9*3hN)Lwt>z0SA}%hU3#Y#F>h6mp~pYF^WAa+4bwD((>#BuJ~0SMq5R` zOwDLv&e9vEw5t7M3-12Z&n1WtubO*E-@Pwm2JXi@f4;$3F{^H=P)3xD{ENl-b}(7d zdTyc%>a!&}j;{bd!UMY^e$`sgmENC#PdC^Ls_Dq&Sy_--A4Um6?H~1s7WQ}Nv>t&6 zCLeOsa=Tb)W!cAmgI94LzHI2c1QD;13@d5xg9n0!rXyfeGscTs(n7ZZW&^NO(^3=E zj?T?BY+f!CNV{Ma9rfwBha^|>hrouM9EfO1(2CZ$->SW z`R1ATGV-8a@Mm;)^H$P(#Co;N;9@-uU*xC4O4;-3DMqh$*;St>J1&yI@wuSaR`

    ES**UAOLqf^RO}ukt^{-#Erja!G zV<`(GU&SXw@>T@%=R0bw3vZzh+}LC4Y2$;;(m(Fs?PyJ;0y6K0#om7w&mvKJ z4kFJMn`ljifc$ge2i{ZeBW*L6N1}o{@;HOhe+E==*F^{01ahUE5eddE6GnBul{O8U zgxf>AY-(!O$u7aZm?`MD`Il>adEv;knbegEruvH`!yF55gOgSM;dn$^c&#~tzTLXQ zY3gZpoc($`h>(f}^qEonNZ|60DwuXh;8I!f25?VfHd&>LsyoUrb?trR}pY15-x;|#naJmtS;wIkk z5iO#qfN&~OG|Oy~?m4XSU-BfpY>a=_T+dB*DkXIQX`2EZx1HQ~qFUO9FppP(AC(1x zY4Gj(CiM2GU{6%umjC1gwRgv9`Cb6!LD_voAMxA!Qu{@dJENrNeLWXs0KpBvfyeWt zGiEs~GDYrg28xPB@?2#+$<`igABxel*4|cRm6#z+rgjVObcqauq`|b`p9aCgb*1EY z!Tc&Om9_zX?jIVdxiiJ5&L{A#T79tHP)Ei$c1-{JcXsid%t;i}Z^>arYA9EY#r#%W z%Wr%`TusN{i5#AtQ8xU|#`T<#-Z^0;F`3^!nTV=E84)A!SVAdL3|A>+G9ukdik zlgqOsfat{fXs40fN0f{~thbsW2Op(X?|;#`alrn%R_y&^q5C01h;x)b({bN%+GJdG zD-+=Zg2R2F(x)rw(*Z@$^q%(w%H`rU_x=8?kw)D@LIO*F(9#Q*5~4L9j?*41`8-iw z0f1ye(m1;!!AeC0&M`PYa-2AFFt+ereQWkoU_xDIPL)MZHoM{YJ&)eFq1jA|NgIxh zZD7XZ76e-(l1^(aLt*9_R*c0=VyAt*_NNBQ$In7%ht#QNz~hcY^*LC+IZRxc379pm zXPdG<|2dC5BJc_q-qB0AwIz5qgP8`(sPhCJ&OrZJ4&W*fM7x(WidxLkquq5TIW%Lp z^)BdV7ta*&@MOvtmp6ODVOxF?*GbwXMINY4IxH~Ki&jW{=N%*oSW`IZ7M}Bok_b&D zP!@Jy8*GE>R(6a>ZoDVuePjr9$MefGd9Sbv#-&+A8tw-(c(0@_E;WKsR^Z9ivHG4; z_$^1`T$O9CnfUlu&F?lrgI1e<7tZ{HK`jma5Z4w)XzO^neeHT3DP3zRBxAh&037q| zPcarRg6J!+{VH%CV>$4>LFU<^VItVLpYSnKpcN1#4>iBZjV;-e-0u`zTuF}*A@9Rk zhG{U_B>eG7@$YzdaR`hgq4eCVMLv?dGu)i<)e9#Jj2RDu@MiQnqdlY>00wo(zJz3b zuGz_c5gT|+%s63ibnpDlfg4iWb<3H@_zEFmv+L8w6Tal;;_CSCzHb<~uTK>CW4brd z?#6w^R2SD$$&xBa#CQkEI%qHG$`+TSnm4vYA0K_AJb%V1oCi?XPZ&O`SnZdm6S2SZ z))zkQ=Qtsxi9RZ{L_6k#p&%6tu#!{oi5hd+0WD97{d{?AC8tZ^){?x^8Vpiy-l*MXBu{F zxWq^gV5De(0=;E5+rjJw=c3htEW%ngM5{8aSrlxtQXkyLhS5z9wndaFIohrwpws+N zMz1$Je6gyM5t~FGxF|$%H8-PC7SC+@op!DUtIv|fUR=%T(dy^(5i$$?YJ9P@CJ$EU zMrwVal_yI{wFRIVlau|9#+FV&KxL=tx#$ewJHVjz%23HJavMG2L`8PK%KSC)uX)Y{ z8db@#iXCu0P=XB`*9-S03?57*Y-~8VF+|GW5P#~Ay?`3Rr@sz1dQ+~6_ekxBPP&g6 ze+!(=RD%b{bd~lzwhv!UBh2f%8mP&f?*xhC1*&@=tY#q-4==$HaM$^}{9z6t#q*cZ z5_V3D1!YniKsr;}b=gwL6j?uXiT~95f(Cavwt}@bO&UkLNN`6Duge07{h-pEf1q+F zjnJ?okML}Yr#GYgg0bHG@M{3+c+py6fMeH$2hk`^dvKV&4gs0}5wq_XBf-7Z)cL+E z6*MxX4)Na#V&9xwMeezJoB__Srz|1Q4J$~qU5czdE>zD1C-R@41?)eW(oC@S7zuCS zOInNB>$EHC6y0oV+cmZv%Y^j#w>%t3(g?eSowY`JlP#naZ>xM_B7}3cNur;n#Ls|T z*d=F|PLtH3A9f&cU9Rv&b&uP4eF%AjXe@pa|1OT9kZ!OY(pAB&P)frbId!q-DpwT7;3TtKO^M;G_RN8N2jq8 z1>U;j*S3A%K6?3sz}BvSECQKwaSTYMaYj1`k-Dv5!8(THiUN>u2F}nd8v<~@&0E{h zo(!la^9iL_v;!>8mKY*beSwM8U0An>To{P@Ht< zcKwRq&>^GlsUMSQo^;!$AQfRlkJ807!N%o$^6S9X8KI%px=`GQj5iaN`M(Qgle(oH zXzTmq#i#1kik;+|<9a%%1ibz(+ZoaNncg#67Eje=pWvxJ&6;qQvDgI0bqz zes_?B$Y5T7g>QvSbkYo+eSrAgQpMoN^w_qa&5gqJg1x+Zr0uiNYJ39SU=oO?2Kh?T z!?xjmR+g4+gE=6yoM9*xb$=6`F5`_F|K`58&=yU(^c(;5=Ac;?nzuZdyjK9A;WNo+ zW@Yy)UM1a1cI)Q9hV+hJ%yz!v)i`qq5Nui!4VI*QUwvnFkbu-#Bx zckSuu+WF$b7j)`H;Y@cl&-%z{_^QvZoU2`7-WNHBKLaqneP)RS^bs${fSR5N6Q7^i zg?%-gs)9hw2s|O>*j~qdF)le=;VuRC zxaL~w{$ zZhWX=oeGRhf3s)_FBeHbN}#iauczD)9If$k>oWr6rW|(E6pJ zD=(8z`X(A$yQbe}GdYZQJRyweKz}TCm=H+vh#~ltd^hFyo3ZnOTVu8373?ueRYv@( zz(M?8e|V-$YYB9#g3+V*xidb(rp-e$H2=WCSXW(2?5%Rp?MkU_OPja5OQok@oJHpj z=!7$|sUew_edDf2gXC{6MUSiAz5NAh;+a@(=l}S$TWI??Z_;o< zG<}bh!PhXp9}g3myv_u-9h$03p{-$A%z&l6LZZ!i#gm|H37?v9-P4ndil-cVJI^l& zlDXu!ye6Sk9lX@N1+{$Bf|ELDyr0gI2a;|@Ck$9jtxHLVbv7Pj;n8YaDdf`d;+PEH zR#Pa0SxOJgO$tjWM%@CT7X2L>9cRrE&iyO{`npprWQk}B$DgN1D@vgf%qzTwfV`NL zh)b}L?N~ELD9-o+1qzzqU-NFJOJIE-^DzHiEB}S%?*aCn$J?VM`%RO?;7NHs?mOH4 zR{g)=J)eD)I<)RNM`q;5y)0d4QQ{{-pYw`$d{;)$NLxjijDz zgLz=9LMuZ-Hr3dA=)4kx%hOT)^^;Q`B0`mT~RA$hvHVYAJVUO*2sG>m*YA1K?yJVF(T$fl(K8$U3SE;y7OZlzMfvBU+i6j_vtVi8ZBoCKu zo(7}D+mpS;l3%yKY3vU+TR-zPilaMAzA4uGrak9oD=i3x&JGA5XqQaJ2D~Yk%;+qA zcOMhSK^{BKc!O*IM~TOMdchiK?cNM4zHoVxU+FwACzs?=Q%(Rnxn}hXoLyKt{aN7A zluz4pjqu$YNZ{qtPk+hfuFGG}YR7rfTY{2pNkT;)s-Kdgub~b zKRYGnNM@1H_$2Zn*P%ZpXM6$;Vr{I_AHGM%QZJsr1>x|)TrFNiEVJ>lFAcuG6We}l z>$+g3(w#Kg{!6mbqkvMg~ zV^&r$IEK2=DivW|NYR}ZM&e@q((;u~6S@SN0aM?(T5Nk}Mfi0CQxvSb?!1t;rK9%@ zNs`xO_h6)nUz=lgK~ksC4(#nH9n66gu9yM$gxzrWDW3FSAiIPE;v!98lRP>9qwX1F z#8Ycz#8WyDth7>Lt{ppPdA--H?nF{lWvePPIws08bbd_zrdEfOEiHG1W(D=eq(e7A zP-vMU>Ts_uq5O$-?7S}?_#;iE+7a(s=Lna->4OOo=SUsKs(7A}C$^5OT`cXSm5Ojb z#JNRdZao_(yLMu^Enk>n8O^4*>R0kWaGU+=!BuG9>4BoMx4b<+sxy*7CG9j-G|W)= zu4U@7RbNc<1c=g8_|v8%QbA><7I>LB8#xuhH*bS@-c zTv;p!LE<+9#jdzI%j7ulh<`)SgD`t4W#xlcyg`q;kDk`IKC)}$Bzi@TTh*W&aVE?2 zcap(BcoGen5d8#OE0!*4#aI2XmH&alYu~#!lxgsaK5&J!7xEXq=Zc;|^x{f3cAKwo zRtnr(q_Q~RWe^TuC(eF_jVFo$f!K<~iLbB74#X^{pq%N>>R+0>E)M=5zCrhDo{uA-C@7Y~9T4G}HI zzfz|cl#}wK?~ot2v`mqmhKLxI?a=5qTewf%_6=EVRZdpxdou2o*J_7@5&My zYWxbcb+-68Kd*EN$h&S#wteg6F~%HhL9`Dm+q5lp5a}g&*2> z=;SF@H7-UY;K>LW^IwwP z;=$`5hz2Pu3mLz&;VOYdKeSoeU=4n>_uVO4mDMkz zXa^);b*Q3kk>rIVLVco8L-cl>M|sl;H2E`{wjOvIs44EDo-w?iBCCkSi}DrUt#pfRZO% z0py9&5bJ{E5%<6U%;@hw`*+3?g=~X{8Pp_jhGtpUZ670HO2p1WaSzml#e$;ioqwUr zT*bboFa+O+G~HyU?$_MEYdYJID?NjbDMI;uV_MvYXtaZVLe9{i{+w%&0*bfC75&r% zuWrI#h)~gn1`xd&Q;u z@7MJHaqT|>kAB00^$N2d9RE{Q~??dIwNBT>-$J`MQI|k1UZ5{ z*FG8lE!>?z18>yoEh0v;*o2R{7?Ni z@BP+%Bixi9u3yBixW+FZ7^kY#(h+)iKcp3sihN_><%-R6AoAX3C}yj9;C!QR%^2C|2ct-n5!E`s1s(S4aeAE+Fbi{A%m zfzFJGW;gw()nF5(#AbA)HgG7Jza5>ZQ?ilgpO4ey>hX`{cq<3?v zv0b-E!&`suDtPQQ`2R6>)^Sn2UAvc*5JeG?Mi5X^y2AjZyGx}Rx_g7DsH6;?N_U4e zqA+xKD=~C;o_qW~&s)BqbIzZ9hKarRz3;u&wbu1rD$?!cmV*y~DO<#Jq3tu5`}WHd zPzdO*)5_O=g)PtNAapF)zwf@YN%W~fnfL$=9<^#>- zem`6sA0*utV_=Xh4-!GO%z(!);AOzSa}zu*KOJ zVq|{JWGVdCEV4hMXBWlMcs}4Y0XjGtg~YNmp9EmbbjS17pMmsE295EFqiRgu?GIlh z95DCjSOEW{59!kS=0H{ZCw#4(_e5H1?$|cu0YgqYi>8U=1JZw8K>M<$nPw;zPeWXo zzws$TYiX{zEIs--}ZjIHN+B#46xiNOpRymms>xmgLwXsc@ca18?4KYzAS> z#5>EwrJ-s%a1Qj_q-{OWq}(?5?_87Rd_TnadOOv>BAh7(<$j1<9lmwpqTj0w=gma1 zJTTC68a$SL6#zk|A)(Wtr=w%Ha$CRAw|+TzOILj3w83VQZ@dDgRqbJa4gK?-FIX1o z58AJz1!A@8JAC`W03N`fiDFL>QXcb@#GgCE6AA6LX(6bH^QOj~S$yFx`NMy9>2FfB zf?r=>v1e=0ch~WSA?SgsAzj4HS>_)Hof%%3tklK5iiwKWR7LJ`6~DUKXZYp3GAUjePui$Hnj*7HLZX~eG1o%%^KAAoBF0u{h$`LnuZ zhUJ^$zfQ;F4TqzLKI^ zQyT<^7!zOYvqqF8kWg`8>A|KPkkkuMNWel{p8A>+wHZJCrZ0g&S`TG^_8=b2Qo7~I zJD91!U6qxYDO+S%|Au`mnmomk#&Ih3ww`n|2GZT{!iqX%XHd%sATM-}r#}+9? zfCUr_vyndlaa)E;F=>M8odJQCHUxY*4<%hP)@+IVRy977D& zG`N_NtS5gPWA$?E*hok+c(*!{QhHjQi)>_#^}S<#lKS}yl}O(mkTBt1*5tzTR#YaMDuj_X$d$4)o@2rNCEj;pDV?crQPxwpdJSCvrBHuDIa-q#-RTPvA zC83c5J*FTMwg^Ur_}C{ZN$1(+JeC9L1FmtCU`J;GyH6A-NC5~^i6FH_47)*BlT{p3 zDG}Orun`OS$wvPbR{8l&V#f2?FrN~&rMmCK;r)Rn;Q9T>vm14gxY>^2b2WVNv;U+_I06iS5Ag4|KKzfj{^hUt zOTen!m>@b>-Rc185GoNur3(>6AYDj*uJkX|K==aOuaWej5ROSEQ^N4}J8>otGwtv7 zV=bhJvftybvV%>gX7>In{8zxAkWcv)NM#8ULW%yinWKG9{k_bx&cgc1_iN+_KRK`) z)a|{~jt%#{^6ivOfe9 zyjWD8R#p{_kHXWCV4{B!!9Z6%P^YZ0{gY&jpitJ0UU3%sme1?$cm1{R zf-eqoi-JAZhUfGE(H$H0$_5UVc zz-!`WtlULrB+VsB4v|LO^eu7c?Zp1~P?sryKT7WJ%alvF{ekKGI8L@X^p>~(mt*eJ zFs|G7vVQE_itP~`fAT@~YH*ayjace`_P~EXu}}wz8wKv6f?YictUnS^(<9U)_+M{( z_#bckpGyM`bBeWCxQKRjxOcpaj~G#5DTs*LL0={tXg#Vx z8PPc%qx7-=B(~5$>8%FK+>ZOt{*>H$5n1V`C~$oZJSMYw&`VgtQ&Ko6_-%P0M@L6% z(i#xWZ#4ig!K4_>^l*KApw`VU4meGKLyfHCWG2Aa;uVnb9e;mh>afg8VkBae52u07 zRb1D{?kDisL=-h%>b3@x(Fam{W>4AwAoy(uQ9rw(kN4kmY>7m?J}fasJCT?W306Gs z^BsiaQk;3~?~Hdo3gr7#k6-$+Y8N+krGXwCGnPJ3zWapz89H5I;(l0TH!DU4gkZ9J zKJy{`&Fo=+?SR>~s5^3!>Z=$~8*-aUAOtlFh|al=Xq@|=pdoB#8>VX&Y7>vMc6*M* z#Yje^gYW$!@V&g)g7^cHKsBktU;@AZC66pSCmrTn+d*FkyLpOF<;FeD|M^_B&jX2g z)r{aOKxN>STMf(ZGfe>;!whTi-43GwSe{sf*C8(GvL~| zbB%YH`4EzPcV^Na12^3(H8NfMG|OOz)XW)74H$w=>rCu#a2!m!7HGdG7SQdyT>s9} z?N~6V*Z|Atu^64!B{*i~N)8imaG%ZILLaZR%30VLV|CGW6 zA0-Eow2!7VaCDM#ghxTF5%H~FsWzj@17D@>1ReS$BO(*;l;9_yE#8RR8#fs*_Izip zU#1!nQkPI}1-qo=-xgWvD|4$hwb8AW6{9*T&DUvA=gU8pb96J6OCu==P8o4*>t|}~ z&$`8h&|~4JziFQ#@pY%?boOlCwEet#H)aRw+8nFV(REp;v#H&Wq*oMDn{dHDf6X(6 zKU^nHah1zzw4Zm>~uSe7dP!;Iim3IWvF~saLOLf_oQZDTp+6GNr}i6mng zpRDSjX*_>u;!Knx<(p`=hQC5E9b8cU$F3?d<&B64>)~Pis&1fUS~i ze%fId#Zw57s|?U~sBh}A(Ys8kKf4x|x=JqKtBPBH1hLoS;n8WU7`CwQ)-${MY~ZE( zrQ*Dy&HjK@o0|H;@6*dC@pR0RgqY7Cz1g49qYcMw5F|2-i(?yqP@7=A+t>?Vj}&&> zJU1&F7xfCNJ3PiHJrDB;h3V`dy!%sf=Eb5Xy$<*>jzVvYYoDAKeC1jhJX(z9d>brk zqf@K9FAW8p3^r1lCffOx*KFVMBIzv^+nQaOeKt~X+*zhT8k$M@xn8)cbN5|iyAn}r z$FQ3S?Umb=ai~cC!YwpI<2{5=@cTRg-`~%9)Zs#O;Y^sv>To7oj3^1(;-<@sGgAO} zK=EcjDq5rwjSUqV=&gh$2&xA~im2YpnYVleLxke?W$fUMlmRWKR% z!d|a<40I_-+||EZ;J{H4RHb_&JToc-oo(=!C3i-%!HHQkUjQDY0tYW>O1fmy$W`BQ zi3Q|HJ+QNzZc+%?e~IKUG6Xr&^*Aw~8V_IKh4}G4bDXVe>b0WSLfIEWYZ3c7Kf#lJ zAmc0m=h+SI*Vfkg%g0|J*l1xCK#hEw$$I(@4@=~I5*hlVz{AdTG`F^2qQd797WmyB zDRmD6k2WKtN2iheMy{?FutD+0i^t2)j%*BJBS-GKR!zpyN)#+!BUkhrG-S*l-vwcJ zb@nXZ>67?vUPN$Hdu3RTv|Wwg=zf5-Qim(ye$$L7SNT~|sn=vQLo|x3D@LTh=NF}E?EZ%9xi2dQ}fg-{J-%n$97qzFX`I@O=5RMrm9WT zjBC9l8@=yc&E%<;Le_%x_Ff5I{A((#O)%@bvvri3J_y5Ec==NCm|q}`2<7pGep7(Y z`cQ9D5GhAA5CG)L%({UwmoYVks5E|m&)CXp+hwpDVxm=tUYgi50b%zgKUkw=%Q&ezf#aOV;0`6gs#(#zEs!Cw2BOO!Bgu~V~q5M zVj&TPz2f7*v?1mT+Q<_Uc$>HjV`8t$QE?nMD%ttr>)hBGDJP3XJ8rS(u0x(9Hz=x82Jl-08>{&qLsiINdEN6^(i7DuouS>U5N#Y-$ z^<6S4qq|`I*2A%0cuY)hfsx@SQpL7TjT6w12%05p*G(OMXt6BN=WO3XvXyPd*f#zo|w08$WMVk4ztuY}*^2a(In!6V*x%HIM zdkxo$N<6NpO*7tqbSQB&*voxF!?iwLGO%~auKG<@Eys0S?FmgxQtG9ShWiz| z;z7_-E6}Epq#_hl*1<6%`VSRo-7W_Pe5&%*5jp5 zkWBRG?N!O{xI3++T72B8NNt0jPAVn(d+4tq5S#J{*u2y-)7lFt%)gZ+sr6~P-ol}pXbUE*< zde4IepiuK)0ksz5?L0?j&MRDAo^1_e$dQ(1q}2uk*i8arO`Kl=1YA*hMGd-&!W%p~ zz^)F2#P4;XM!qL4J{Fm4ryJ>&Zw;y)GtL1=-OSfUBX=~t-3X@Ol(nH%v-4eQ&E%qN z1UjxMEMVf5$SyFBNj7EoxIs$(Z`UVZRPmxkauENHku zuPT1_6A*RMQVSehAX`QPKRvYBvL44uv-WP!#RDhUTJLb9_gc1DJ$?}{#`_Yf;=Q?0 z1+JCx+68ULR*H5a#GqttOp2465rT8vKxcngSS%prvO|yC?UUB_-kU&dxo%C;xIoIO zYDvh~`$R>s*rq<~LzK7=LkPzPT?>o)p0)9PPpr(~leXLPL(vhvz9K5QNt#B^UCsSP zHzKyA!?yLmxCU+wRJcVWn1@CZH|Mgj{iw;0;{{z)8)hu{=e|yFMMVB+Pu{!Wc@<25 zEfInq%YatsTvn+6ByyvFU^_S% z8SnaohTHqYxb8T!62?B%onVU1UtlA*BouI&RaL7@O>vm~fTZIdx4ItlQT3uMOKmzt zF3P^|;oD<6>RDMrSdU)UDz#OQFWEgm&#f!ldhcD!&(*ERneas>Mw3MDEb{@{EIzUR z-Ut-4cYIGlJWU~iuiIL_%mB7zm8q#o1TxR!5@j9kli zeK(qJfFNbpU)FeAnOpuQ1!%=wwVF35_~cPFtuj_osoI#yMu&2A$kFg!HJ^p@F}RKR z0-8qHZDevM_oXo^zdxubd6NSA6S*+TCJc-WP7l_Y)H3Bc&A)P}q`uXeLM4CLAs0~v zPYeh=t@!T{aIJ|BMQ4!TsI1vqE3vEFdEZD!S85PYa%xk==uw`PS$0(5jLUl!8q>3hGKI#_T@{%E|9p*as}+GJ>WA!Ec7LO4mns1$Z6 zYi$>2lT8h>hl_K-7q0EJURBBIPY1fBGHkq;|K#g`!p$hGYtTxz#ajp!X8|fj-XLzV zkaI(e!&m$ePNBilzKsE){Jgo%_zqAVwPWaZtsGw-LJeoVE^K$SjNa z@z6`dt(p4zR>}SnRUSf11}df@S@QD1&8her2!O{X2Kti9JNJ%v=DSHUYV3ah{AiY^ z>?Q^?2*?m(%qCthuY|2hdXYL4+?{}jdgY+$DhTzYkEk(Cm1LnH<}Ur5Pn_s2W46#R zuw%1EYF5O3smzCyoa@vHysZXE(u6j18YK#_iBKxF<NYp>rDK|m3csXRs1u$C8Jnq7?SlHkrH4m8N9`)~q$#Lbvz zQf@9R70Y+^^QJK-`LqLL!z5(|1LQ_}eiJ9o8(8{m%&+U5xCgWQD_mX9Yy9fqt;6^f z1rumI-*HLQdQwO*p}M(YiN7%Dan+%x2E|Zwq@0%aB=fe!ruQa0i~QV;%Jbl}JbLA& zd)#>|!ttj^*E;m5htr*f>_Dj$t8^VvMrnYaaA~lTZEs40W?w(MS#B9}a1hcQ2RYA>N6_h@*!`v%KZZp*P4F9Xmdo^p>)#7Z{9|6|$)(Brl%dap)NeK627z zwE1;@rqsjn3C3?_wz0(d3Cp`)bC;Z=g}>ZS-GJhib_k^9{DH$rCg4lFvn_^=$mhZI z5GUu|3E-FYB#&Mwmk*Vuy$6)P?+-s;2b^?DA?LQikUBu?8Tj%9qmj4oT8^vv)@);= z#i*l#P*7DUL9PtqqRU#sGReD5c0v~f{ns)Y)idSEdJ941?h)HC&uL#qS&G?{PNIZ@DM#{eVDT7_ukj?VSO#T^c2=s;uk(XQ=Ax~NQ71X9Ac$#Je<9>GjfaNvt^=@fG#3$%3 zF+uq$Ug}!-jZOB=3UVHY#9WuN(9nY%W9j%_HJ^snZ{L1+iL*-Iq}pHTN4g~haUA8B z{X0-#Y2zK0Nbead%mI~lu}Ayd0hxepqx1teg)Cq!+lhJkXL-Eyy<$2^#frLHzwJwD z!|BnGOr?~>E#E;GDKlE$8dMXFv{Ox$K$iB-6&Q4K%l(_>a7i>|ht6AUrSvXXQ_x>m za^ia7$St`B_@a0gvZEIPJwLfyVs*Ct8s2jpmAGkGi zjZ~bSC*DfgGYL|X=W1Q|o;da**gWe})vzLClU8yW75O$lRGOu1{5=^l;k(}_W=!{B z>t$R;3Zjm~-lDO;do*g@m`JzQ+h$P4jOX#)j&wMJ^>Hh#Jc|TbUR4y|DQZM^A7iWf zJNM)njT8%{eBqE4>>zbr<_u1mg~)@tX)P}%pqZx zv+??p%8z81$lS8pQJW=7*@ZfmLbwLR9=T!^%(X{7RfR|tkzrDH%<}B+jYJrFj#74x z)Q24`6V37Roo+K|?BMnFvMy)u!LHTpMLolsd^q`u?1+g>!{Oox zUs5_Gi#P&tme7xjscI)&U$go5=_eMVWaeGu>nttRy6Tyklsi#ugEo6o^$(;ElRSL{ z=1&95?nbg|3=5J>#|y$4y7_(uYrZ@>jVkp@a+BL$9}yf3qI>oebE-g~qP8$pV_gxq zbxePAL2RU6Ja3I6q_CUgsHWE;wsNy}_3%1GIq96?nS{N&`!g=}d#c3B*C$0Cl1Vj~ z9nB37?`$6=%JGQ#oyp=Bjv5~I;WdP|uU6i|7H8OOuXldRxFNQsDD+Ef^t&CV_iQ?w z^jQP{2qnv?kH}j8qdDImu+b$QaWYkiMK4)nM!-wCgb$BPf{!8q(3KD4Be{IJFFFB~ zvcx`0{$5TTW93`YWyH{EZ@MC;S^5&*b#;7P z@J)&`G*78QA3&MF#<9edJ7->ua6l71QLd~lc^xH%w|;59AA^_a>#kW!t@Y<;9^jj` z=}ua&N0C=VV3WG^sB%tdh>5MJYo$d7cr7BdXy6xmqY2!Dc6vL3=jT#(DW<12hACv@ zvF2EzHRxNL<1+=H5KQjZqHqtZf#5v{K!CQ4l*l7t@{7%%u>FXbhKZlM_dwRl1#lp7fD>Q3StW2r3%3yA7m`b^b&o+ zw){W@OtRrbp&bmP8;wg#>s2#kcfp4y1X_<3KRI_P>br3X>8hD6dXKl~nSrJ~-dh?# z+d@}*@zRaWsKMJ+k*wd?IdKfQY?bqW;W{6ljv7JJ-pP{XbSKLu4Q7Vo4kFV?;^fDd z(HrffC5IQAi}cqI+w5o>=f?A&bh)}6;2uj2!^bO?)15u~ap5Gv9>ZlWwYT@rPamaq z7T@%G<=qDz#hh2z9>k<brm55OINLTj&9vsx>q#P@%ReS zzAKetJCb{nZ9L9r(XShEi#lAg%*_4GRo>gIQgG|~ z+bczw{@0;)j-B8t^}_z2j=JTeJ2IxZ36r@)q6RZw2SeL0YE+P6ICs0sNuy0}+`l6~ zz!3oPte6k2KcKqP9e+=&_B))g4b^&4{k2E%cH46abF(Gg1(dv!!gFOW57=UHvILsg;Z;V0jz2GCHl0PAB9J8Uf{mT0E>-2$s^Lx&bN5 z+>cF{z=#T<`KEJ|MriI~!(ugQyH9W3!kAfs_tiL#^mdtj|LfhZ>+>0A7gCxeagSer z^S&I(nRlr-K;&giSRROaXnzp%e?GdWQAEo&d<4s{t+4n?7_61I6*S|K8F(&aS%<;# z-CcZSZ}Il;Xs5Dp+b0wP?$TW_<@yhKX zP{_{~O10o#`^cMD*XUG2fId1nQ`unp<$y;Z^b{J8E`VpP~y`k>ZbEtR{M8WKqJtiw4wH zBq5e>+h2dT&&%Jvj5ep3(;8#DneLBBd;op#9U*mzM7*y@y4|@4gXqZCri6Y}iXP`o z7F`%=8S5Fd1f2-jOd(KNqf(Pr-$iT~GSg5ecTFy{S@4G6>RFE4+0j0puT^L7Oc6w< zFqybU%017ky$G$q4vmuQhX~=rqpiqCQ8NqcXp@!6VJ~otO=ah zQ-*dK=$iVh9jYzZY#i`x1MjktpP#*KBz}DzeVkC>Acu+Kx)TmM&ReB9AZ`->1B*C} z8`?167|x5K-(!PqsNo$SCG8hdfUWx+@z0v@%_qc+5WmlGPC!b*G9UDe%1pg zHN#jTeM1NS5#Pq4ZAjiJDBZN#R85m?s*%L*&Idc(fy+iZ=oi&TWmf!1WSbf0D%U;< zwwo*v@uv=3pIjVE@>~dKJ>9iI^!~=(65L(x=m5j;Jl4^aS&Sa0`b z{oPi(Y+WgZO}v-ataJ0SdD6FblxD0}`p|lIRUo9uc6S0V^GWAOsJBr2okulT&aFog zrX7Op&EQT&WVu|IU{TM5Rs`5tSO7F4W~;%OQnTD$W@WGEp5@f)5NT%Ujn%k}~^Ku94!MUcTiae@G?T=!h2AH;$tTL~Lrd(0Hglk{BEbx__b~t@{ ztt=im^qI=c(=HoF3@smEI?91wXWMosLrkvQeeC0)jK}L_$q%^`(1ud*D^}Xg+Eyy+ z12K7--@>jws^2shDAy>#Tq#Fry!+~6_9|{|qM#1YbGWRA9y-aZ*ihv>X<$lrf6B9+ zWmm^rfBZYh5dgg7I=`N0T&+9z!SdcM(T9o9-4Z`97G%%%?b9xK{Z0fWD$jq{o)#>1 z@;x&VD}q3sM&x3D5D+9lxe!Yu(MMjsqJNWY582BfNZTh9)^N5dY&o18)s-kX$XW!3 zd~!K2_kHQ;36V#kpUzKVQk_q1E(F0$wBGgb5!TO_)EjV-MXBku#Y6^XLGdx_sZCL{ zt?wHY_3T;dQ4@xfj<9ESw@n_p=^E3=CU8QvStHMwBuld*qG064)0|eJ(+|jF@`iHf zH8yN0Mm=mMYkUig^@b;0sMd6J#%-PV4u&@_A}DNQuBZ64An$$BJV8}VNJPUjLL7jA zO%Oe3TWru z5G;Dfw6L-H*1D7L*Dv!!r?L|vGfANjXdLZci*{gw0IHNi%0&S!$$Rwl2B5sCr!b?0 z&?=ZQ*@b-EFptdy&tABY%j%1Z(;G@~4e{MTML#9(uVv{EFrTu2DUfgxA)(@y#7LQ# zP=;6DsAzaMcz{RD+y;7K7Kf9dITT_fAYr1}3AB?Yk3c&+JIR{f$KD(6}}D_C?8PzfeA?9h!BHuZIR0M-f3i(l#nE1_BE3Xe*yIR|A-YnfKLLEnV{v)L(-) zP68inP2UeZx>t}M!Xe&MU)4CjkGM$_p&(}_SgB$8+qz(3tX*~p$AW+7-u%Jk+a1rB z3>)z?#ip&{EF)0m!4+#6NjE}Bf$P1xgY}Kql88G#6{8oeKRL2mi-v!I8I4FH3~sis zsr{dnn2bup&c0b5$mUEJ=Zu!7bZ&^87N=AqRG(qZD4-z?-5Z3-;0#CZvGzc+)1fCLQLTKs;y{f6^h zIqnamRgj%g*_uDKoy*4&k~mamqH|b{Jf-gKFfdabsR(MQj30uEH6Qon24O#}h|1pV zwH=T^f_f{;i4x5q7ww)H&0$3C{8?AJqVHw#hzdrICRIf^Xm;-e)dDt9EsLbPAA9$~ z@+VXJ`BKv!S}DFybK&xsxhW0}##~?1+95EKVN1Nj(uRXKY^bqHLmZlanDP38M0IS~ zMQm|SZ2TC}8|uy5Q>Tna%<_nsqB645O>A)tY1pq*i}cw|HFg?gv4#I8=Bg1vFqoC| z34Jq$cTp@_SLkmw!z9;CGE4_sU(~=v>L+Z3b!$C>U;Dpikdt|4%(%J;b?xc5KwTf_ z7X~Ocw=p9x6VJ-lCu-Wuj7TS(zcrT>vY0j4=!s0aV6U0k^z7xYy{SysL*gl}{Nxm3 zOm($8iun}v(6aKzFl|iek!AO8;Ys7=G8?J{8)~};=0KW{RA!O8I3eY*y{NQ%swML9 zIX<_imb4^ApyW|o-_>W~^Rp7OuS z*q|6~uk97HE%2na^T|)S%m3bQ`zTT2h5TJu@YEnV{}E*tzEFRE{uWbPb&Q*%_0=KV zjdJZ{D{B{VuAtXl0j}aZ#HgxLi%n{T80P-o0+^U+0y@{Y?B^uRaW3 ziNco8s4k{5_LMCG$lV4;SFrF&ViR!vYFxC#$|Zt#-gYgYk*{z&iD449mHLH>=NOEF zNu>n$tVrMF2aD$**3tz)pIl_FMqWH3hQmm#^8lv>4K$5#0WDeX2QU)F7HB_HDb#3H zF;2xyx+WXY6veFJ2+2#lTx?%gF`@?B4=f(W(g&nRSg$toMYkBG_a&d1=`afP+W+aR;5A)}-Mg+vvk8k)w2;W?^wUM*#F_KH>d zYm|bXlamAvN{=@>f_rZuxpX88S?|TO>xVK&c|A_$o}t_`;ud|4 zAPe~jz!>u}DW4RhO_Vl@>Qy#3kmq+g8&VdRZv#Ecb5r9nuR?H8)LOh20b^dAdDFI1 z#ua^)?B`sCjMgZn>puFfT-4g4(v{}|`nf#y5_}vz(*0b8b+=CDT^P)e5{ge^7 z4nk5X33Ez2uxKjr2j^X$(wzk($Jto}udj;BT;OUAa9YLlb{dije7RrFbF!Nh z(@o>%YcT5Y)Xsya&i{@P(p1kK#0vyru=2=wLK=#l+}4)Xw+o3vG0%Zfs(a?$M5-FZ z6UCy{vtDoBt?(IW6A;sG29|~9K^u*gf7{i&_osMZgf^}DW{fYz|73MES>mnzu z4+ZO|-=P&r=oL(ldxa-D%(S5GolI6+j3`~Ynp)P5q@4xXL2?AN7ze2eK!wuSzfna&Z|y3w9_oPUQO6ON<#UL|Q;x0Hn1)d-gy zN}w-%HWNAGCsLQ2OrBg2{ud7plt>__EtXeLQg=m=4?jN7J8g%-W=({1B%Jy_xruHX) z^|@yC!bkp$7deib)A>jYG=fSIG4FJJ7zz*p6-?(?ZFmlBUGy$Stl2qj;aU02ZYGsF z1#R7W`h<9{L%sPPZ8@y38WkP<+|@Cb-$y%7T)z{eY{EVi>0Ln zVv2=~I6_i>@#4l6om&U7qY{cWUllz&e`bx`7lAjao-_Fj8@fEe+zq;B?6(vn-0p%p z8HC@!9ES(f_x4z`WSAURf$vK(sQqvU3veWiBvx{yO^;5Y1u#TaHP_SNkxGfDGMm45 z6!E4_j&xX?f5F;f(_u)2MX-S`RH;c1%Bw5nO(nwATpI=)OH9F;b9-PLNu%z_t z51g&94HXa}Wb1Brlh&cQ)pp36kBlm61idPKxDx7g-ByXo{f(v7Or}3In)deCsFoSh zm$}$KE?|4+beJFGIJR6`M8s+{Sv+z)kP{AyWLnG`MW%}TQ3nG8bmUkN2(Mdfqg9ak zp=hTQfkrQmt_(ONb9v$MT7hmBK^J*uozW{lGQR6jnY7fc^asgz14pIb9?K9;a;(}P z=Yl)F_rb7Ms1CB;3A2jfd#q?&q%gVz?z{A-(}^MC6glxs_ofd!N31L^pRXn+GHu~T z=ILtyOob|FKomTIJ}de?XO2bx_%0&-a>ZP#yJd|OuY!V?8%oU_Q0GYpkM8%8@MhEv zX$>0kyZNZDqx@uFNsoaISCoipA>BEtB>C-M{vkR{2@evKS;ynWTxNo8$FqvB%>UCD z^}O*KT#r`rN_64BjPU(e8#gUKR&ZPoebPXf^kw4b2C1GviEQA2dM_IR29W=cuKzC* zc-E6{0>4LnmS0%?>}|)r{Xr9mc>QqyE@IUMT>noQpPtuIOTzCBKmNU(_}B{T81`{O zPrBtVcMdHF&d=7LvBQ0{oF!zPzjC#@+MQYu{h4g~t&s^=n8&bo=qQo}#n>RlhJ!wK z59U8krW;^A0+&p@ogEwLdz5<#6TEcu|MJowi>)Bx3&H5Osdzl5k|llu#QC4ctV3i0kjoPMFeL6xInwg8xq!NHeo?rMs%s2`3#V=Pw%= z36XK)&lfHDOlFDjSkl)naN4&{zS ze?|Pq-L)GAPRh9xNA(?)h5tGVt>9n$>5mcZL>e6GWp#``F3PZC{0%I^8VPO5zfXH9 zo&EXo$9$7pGBf4-^Yv%wPi-%HIJUHE3^+X8C{i-|^IuPSH*aE0!C za4RT7eJv4vIF`RG1HSn}C?|FE^pJe-tMu3Jh`30OL$0C60pn746dnUDSC_hNd5aAph-SOD^lmjp3CB?goDx_mAX*zucCx965 zt_bSStz@T`fByWxy$b#3d;jN6OiOsve<8Bm{w_sv7d?Uw>hU9f&4liu4bB>MgP=OW zU(0j1L1JQu=ys9G>CT6g2qHSd3iKTnJtyY>#e=wj<{u$>hs)}=l(ERNlE7g5ZV*$F zh`B(oDV}L4T0t8D}ToGKi3$48M^gkrCuOPL5A8 zQ#HW3D4!98fAuF)^LYA2z5nU$B5Bpv`v2Oh{EsdE_i@9!ZzP_2OP!U*#@_oF5`qhh z3Ra7K#F>1}Li_6@5hGjx(=kE~n@6qS>+9g4Hpp537`NdPe3&E+s9#uQ%S0N0RmfHS zZx1bMN#nSJBX%W-{%5SP5m@ZqT2=6%KRz@(u(txIqjD{FCu(`UAi)|K2sPjNkZ3L* z9i6=Q^dXu*IvsG>`Au1{6a95*o9T=BPqf|fZYo>JGNbif6Ip-Sbf0ib+xQP3th?QJul=F;6tbOI zMDdeEEaoyg4Wss_e_W+gz*QQ5cH~F?XT!z!122Q;1a9=}e~#}6Y%@!jqF=E8x!R22 z*k8TLK=D;b%-Vd=%=~*Vyd((zdZy5R!n#Q*6XL!B56gVFpLlk(6$6MGoj_pD08=Ih z(q)J~hID(>Ru2uSpMY^arsJiipyEJWwhj=t@5?vuzLzjV7latiU9k`UTodR-o0gBA z@wXN!MYCE1&?;2{QlWag8UK%v0X|j^K*~+PB+~JQ!u@^cmFe25yBvo4U>2rj^kjpN z7wCD&1+s^9D79$(MyVj7e?A_#{&}w#r2pQDtkGNW7)EwH{_7X@yj=r7P~w2j@u?@9_N zb#%(HlPx*#WWB6+0ZdMc0W8WFpa(b|)JKzj;@AMnS=elwo0C(;tT&krbXx9Ga=reE zQ;kgadQM{pMsf;n)$NG@m;|jpy0ibS7}mwVoa`XH(EigaHWGIT_#3q=nZgMu)B2#K z$6hyoH{zu)m9VS4p~oUI_Znypox|~6ZLQs`kpuHBj5kp&WnHN9Itt11izL(gj~_4$ zH&MXhX{LPK>ycG}xH!-eCV2CQBC`BI=6gCA;x&}`A_u6@0|0Ma^vwBn#?Cwg8K>VJ$&RiTbU9%%=bAZ=D0qrqA_WNAbMteP085<5$O$;#E~uO@Ot;@$v+D)@4pT zD*zi|b*cc}N{8K{i(uI2bV&~$`Y#gD8~*Lxqd%Y!sLdE1OgPyc$_9FQvgz5=lGiQi z%3q%@F%raL!%<^R%?4?+D!GPBp1X7@CFxuO#sVB@Z=L=a*WWz)B!cr+9&j)2r=3p4wAgrtTF(g zIevN*EGXj>(0pFTQxc(p9fo6qdc2~V-cffx?H37sdHJS`_2;xjxcTEAFK*Jvb;5>I z#c+=AW~+03yq2v*#z*G|f%uc?g>Z6(ouu2k_?;TKx$3~@jkPXH;gN!HU~O#stfvQ~ z7Ib|6`i;A|ZODoaxSD<^E{)g{a5$U_xm#6gImqaw6-F)kTrWfnxQ3SA%IX3;k<&s> zR&oSbopD60p#H939`y*<-yUV~Ef6-8wSAYim@AhkglqB`Z%blg*oubsQ zg+sUf=@ITWi#p)naNg?Cit)jWmSZ?R_bW{0YmM9lV~!t06`4`1l|JRIKDjD%rG&sF zM9dpz6WmM;SfdFNwI#4mK|y6Sn@)nFqSJ$_&R`(jE~qtkfD*><_Q`J!#{D}|2|Si` zhMR; zf84*y%9yVMES^|a^~H7buDmMIxsg24iEAdvexH6!x5AJJb=L$4{xnYpxmWBYn3mve z#>KZeUWS+=&Q7Rv$;Xy)5$QL0RO>1OrbA``>(U5&uQ9lDyQ1?@s1?8xC%SgfR!pb} z=r4F#?52OVJ_!%yCOp5)%*+k|puQM+%z~!Omabb7Op*&4{rTQ}Z}_2Tp6rq7(Ev3)Ve)i6v3t3cPOnIxKN(qN^khD)ZIzwfpn?0i)Pw#2Q$1 zmt2$dQMa<#-Zy?e>#@Em9^fZCV(&c#@^SYfDCnNrFs!zBp`+gilqPcY=bzvu*NLQ3 z6`JBDeSxn{EbkH&CF3ShpqY;!KcYTG3{3P)WfKOf5DT|?kMkCZ4cRH+c*IWj3L+{0 zwLJjq_E*_B)USF;P3jIRpBeVQnz_6Vt(~M?yZQ+*D6!!lrs z6^fu8>Lb+NkMfk+u1!^!q5N`XMu;kD4yY7e6sJUWq;x7Cg5E> z;BExqbWrdVa77hWN}X3!bpY2FlzY^B#Llb%5aT16g#nn=Oy)d1X9YYZl!%fy5~%q2 zy2mpWpNEN4S4! z3tk)!XY1Ct$G709shP4mllX27sty6wc5QgB z16%PT82?R_v~Mw4X){pc0_&#*iPj5n|A&+dApc6V{JmKSiPB#6ZD-jhO6u9JDm{dI zT4=e|)CQ_Xm6kFSE^RTC>Qe2}$8b@i-9f?fu)h+4Wh4}1M~+5@C&CDB-LPrG+yGQa z_J4Va^PhLg$^PxHM}LS4w|*9#_Vo{aNP&Y;x&TChY4FFL#2Rfx=@rjk#Pk1;2o9we{I_2xaTkAG z#GU7Ng}65>+i&E;Tb$>465i$Tt2ZS29#6hJ-Fcd9_4DUX_vjxbg!E*KqzbY)(mRnK zfmQ}$LJm6?(s%#k15pH~;y20UXTc!^cIK0@W5L%frrHrN4Cs+hXygA6V{aK&<+lBQ z!y-hG5&@-=l2$?iNhPF9NeQL9JC;}|-AH$rbcdpJcZ1R_LSoVJpErB!KIga3b6w9H zFBdE4J?9+v7-Np_XMmtr{Z=>E5alIvAypTQZfN!q$0eKK6!^(Vre^2M7~%sIG$_My zs!^baSAQ`9#zNAy$xDwDz6J%hXY32G-z)NuQwx$BK@_t7hpVbV8B(Y5@dQ4m5M86d zCFpPa{2${NO(b|CFG98Lo}R=<_;(yZBdo=$Q2#%cd>xHa$&=J9GP@Vn=ENlA;dZ$% z-(jt~_GyaJ5TTxHaiP8pM}0>`Nf00^Cl>%BDb=b@e3NNra-XT|Si1+;tDQ9CNue=g zUNZ7Zp#fv>6o8o6)5D_rE;L4T8k7zP=EJ3Ae$w_@Ph-Be<} z{%7pJ-Bj$Lro_imbl%17QiHd>6%;;*mO&tnk%vE<*l8f6lSRcihZLS-A z?x>_HRmYa1Rr(CV9=*?-57hm`Y59n;eXawVPO|j_(!U;lfgb+QNFjNm zAdT-78WZKh#|w67nTnw(O)!*lNPFvF{r36BEJ~XGew47I!0U3pd#hNG*pm>Nx z6(H~3Jca?C01J>*c|PNN{Eyx83RW8yj5$Rel+fNf@Ku@3=ynJvn5o5QcFNFg8U7Ys zhE#elo>}sxVCN5w&mElc1f)SfnTdQnnZal3&8Tr;|7%&_W*{K-D0yQg7w}E1^M!z$ zoTcO|Ug02x`WLq|$2{*e>B$P$@SIpI9!Lr<_tIoO8t)T4taA9iFbr_l=u2~q1X zNB>t2;>QgTDeyTllV3X2vp7&p2a0Prwzp?4R8&<|y1e{s4}ZPj^j}yoyagOd$a?@W zD@oB=4CR)SsPK3)JP~4zLG>>XSi?ybAmkAeW!4yJ3YqYP>j8TpKB{z!Yf(ED{hl zgaU$~8tBe+@hKQY3Gkr(&dl@wyblmaLQ(s-3{%MB_`GJf{qc^p0~8CG-Hrz}M!Gkl zW}ubdZyHD($$&v1KAFz$gP7C(brPwLW2omDa#ST><6f^3a;3ld+6VJIP28~k&1vl! zuLGT4Z<7BS4=SK{!j&qq5PyCB+b78&!buewm**D}u3hdfm{*ABqXm77cog;0pAb5j zqImoQ4c(N`psEy4;5?TA0E&ZP-;ZWr*_21`y5zX!^7j3zk;YQ07i>aN`-({tb zCIZtR)f*2~JvRC1uZY#~nQcihF_Dod=_qNPtH%nSVQ~znAr5rVUTyQKFPZaq$scAD zB4JTe>8bhgg^lv@QXP9pVicoD05$ssn*p<>*%WJa(>#ZxnRjiyS>S*yF1?>o{JXQK^T zZuab#8Eu&NQ{(c;U)V?3HJ}cxeU)9;6zoJ?XmV`rZP6msP8Y{Baimm(z!BA?(b3o0 zg{-Rvi)#<5LN03I7x$#7)OLKEdS|_9sRa)r(y1PYs;xD*LE?rHCSs$N`korz`ZPN$=>-ROBNY?r^E18b#r_7Bd_!iB`@Mi= zUQa*keHxTLe$KSggPW)JNu;s&)9pdw3-Qe@k*;fV9(acEFSH@C&QrxyA0w5L94^k! zN?f;Jj@P-B@>q<>#qyYirQO=Vr{JRzs8ddUTIXsfEq8626z0?Oz2P;8N6YkQ;e}Md z_+a@r)w%5&e)~XJV*BIK_~#dIR>;XIo3}yQCwz0dR`%m{23X<&D8}U0zF^m`kW&La zfj~h|#1fD#$B76~83F_8NmZni>h#`LUZ7eDaE%s0$!-V;GCP4*q`QRdzU3XL10R%S z8pwUdnzT6f;2db=FT-Y#kFmyF2J}};-B$Sr(}oSqWy?;S2!1I4;IIpNlh|M)9T!2F zH8+#Qz|{yIN&&|wn;?hdQ0)yrJ6`hE*>ED-S(r`x~_g>#vU&A9Eh%_FCCb zCkJPk#$coF8api)ma`i6f5R_vzj=g{-SoxFV=*VLCMJV?WI9xvHdkCRy`l| zZ~;&W>|dECE6$#p-rWIVTR155RDv*?QdBzqk@A4D4Cv?ZQW*ky&0_t*OgMl>PJ~(kn&}R}!o;G&AA6-`p=!c&6qAIxfKXjZ zXHVm-U%)i^uF`;jHnRU!dWHGW9tg{0rYj2!oGM&clY&li0Fd-Ns4{4Lz*ccd-=_h0z__^^S$bD@$+uMJnWN@s~r1gGp!&!?GPErGxQVp)#U@1E*jt$Za{*Co8 z2TnfG?cK~`U`nK}zt>yc*cMSt-EP!EfTuQ~f4-t#uQ!hX(2#5Q`Zbm=_%!i0J8u>} z4<3KvSVVo&f@2RVCJS!6mR*hD27YIAux3Ln+Hc0UTC8SGncM=Yz8W$7z9WvvcMslLKS0zOkKqBvctr%ka zIIy%Z$!#enb=s~>;wMN%b&vQ}&ba6C6CHxaY`l5Bd*DkCjXy-fMI zSh26SxQ*KYh^Bj!FIf0E+f-lWpf;S!CX$+*oE1VFXp5~?{?vMpd6V};Ttqvc?WVnx zj<%Nhs1@!V^FEEvzNTVED^xPeftWsPQEJr!*{VF%e^_1d-FAX9_ zZ!H5*2s~Y^7b1l~xf0~sF!xqQkAetDO7AA7-h+I^(s8x;0f+!5(u`T5c+wIMZ{eVln0srg5kJ`KX$smv!|(I1>`SqRbuQ z@rQ_^+TyvevR}x~kaXsn`t)Hz!;!A(9_yQhlYTlu((LIUwT>GJi`)9;yvt$yJpl4FeVWUoH7_CfC{3oP?B0 zIuf~z{BEHP-(v^zm<>^1*Ns7&pIg<-R1US6ZhC+s-~S3;C}cM3>@}E+s?=R{AgbfZ zupwL|y_Ht+b+~X#ZJ-IVHJIrOsH}oWAQTjpb-c#F=^FVmv8ad0DzHVm ziDN5>CZ$;{_u3j-+nz@n|5AO>a?|IWyQo$w>g?_eA-xW_&HH%o3>SY3DKW8pOYLfn z^V9QE8o0f&A@kF+xAArxIZ-kVc!Jh~OpR+ECpkKjvUzxt{=;0uYw;5;Dnr;^N%zM4 z*SLBPYa7qMB!VyClwmSi?K>hlI!;q%Um+ngaLl|_w7cDcc&RopyKC~W1mT*h1|o(C zHdO_I>FI7ll;I7FKs#aHU+L1djfk7U`{|1s@Gpwfg$280Hw#e9zvo1jU4tb@N$v@+ zS*+JlGxBrHz0NCo0Dy(cjdB}+@WFNFuN;&-Uq{^#&1zHIYxmrkwUdOe@xzcbV$*$x zzjxs|8HL4EZb6wjJg1B6^@-YqWq^Ulx~bnXe>EU|jBxB0t2JjPg446U#x&UgTC#OG zs#3yDxg><+8<#!_q~HFBf%Im|aqx}ujivyzQPh2~IFV-x5cxIy*@j*mrsE|U5v=MQ z*#>QIMfotkPThSN|18V$GE}R(ovc%}QNX~vVv6srg&ZoJo_M4jULiTefrxdS>VSY5 zoxp961Sp3dVDulBOXOe*!+Fj8m2tg~a0J(`;BJ^(0~MPc8QaiYKJQo!=E0PaDS=Yw zC;-!oNc|d|GCduzUY@uPnIrCU)~ZbF=^8gsPUm4}(4vV;HkuPocnyMen&v6NrE5D7 zpOV2j;nFZk$$koq$CkX^5nw7ifDd5dJU=>@m3b~n5IYAq8QJ`LUVxy~nL@6x1SZVS zF_JJd5(jSn+AljQI@%w$WX~cCZ#>;ofS+YH2@2mA=kB_A5x`Jm(wsv}9(%gnsB*UF zJf2H;Wl_kd@wu3^m4rHw>qtpx$4|}_C;m4iqXG3pD5x`ZwV!yCG}7WXY#uuee_il@ zDH#wX19vKH2jT==2-H*|i0?xjj}(FDkIqPXMyE4&y`2`W0*uE*G{AVA{*>8OFVJbd zHc%UE<-o;fV>PNaFyRiECu5aaEG_xln+u4eT)j`BJHD?Pu2js79(m|lAS?P04klPx z$I{_CE4&grvb*{S$%fa=O|0r)>b9Qqg7}_51(kAg5L?ddvkVj2PIQ^mkH`2zh7~ic zfT!Ax_P4k0)=#B6A;daF1~yg5Er5 zohUz_3t~?!C^+vu*<#6?KI&?ZXrNkUR%Nm<_c}vMbRtRB(;sYPr6g4zm$J=N!8K8Y zqYCK{#=%;V!bpGWrYej+(QOi5F$`EuRWWZ;#l1)|WGlJ0nB08Lgdko>XD>JF<^0 zOdh2?f~z#x^IAM}{DAp4r-OdpU((`aotX$C>Rc>^S?ZDSIhR~_f(gvj?cW?2N_l)G z3E2p|>h6u>#ICiGP3|;{v|-qgyGUUmj-#_1C7J(QQERM)o~(+gDwUp(P~wi@z7IHy z9#`E5JpUEV{qkhAJd4U*z-N-_hR5=Dv#H4{>GYrZg!LR-Z3S72jCjZQ2vFY{H@P*s zEnARo1N2qS^2dSY&WZ=o;22K7$9l^KUR8nM`y>#R%Sd?uI!|1SC_AZ%Jat<=7d1Yw z%eQYvW}`m$GANv{FwaA?Vz->WfK;;-c@Z?Gw#O$G`MVt&`IZ@Sm8DiJBQ}eX+;QqOd_%hS$oK&RF*6I9DUx`ufAi~qaW0$^UYr1CMjX+=%BR4; zKo@X8y2#NlW3h7BUWC?%x_+m?*0Roc$Fq8lwll$zkD=@Hj_#BM`fq~XY1ep4sP#x@gyb&D#8OZUK zbZWG>Dw^K|C)Xfm^B$u}iORbg`3$u@ocw20yFRZl-wq6_P^!PAmK|)M>*u|tQ{rrf zZzuD>{Yn7Yem*?yX#!BMbXH{T7FH=ZPARgkUYihb8)%MBV(%64%ws!si!GO3;}ucG z^yCxnoR4(J)~8jxN-?o@R=10cN0xWTwy{Q#rV}}`-);F;{-4~9@3N8GmKV*Y7`eUA zyn5Cg4w`1BmxRw!e+#44W zd>cX;%wrtC$WF`LT~-gzKpP3tjiZd^CkhJpnTUpGBt{6d8hYCAqd#H~DRoIoDa2j8 z%$SR3ysCIorZSv6Q$U&Tbxu`a6OQaw2S@y8M!S}kS16J4fX^3BYQ!{NlpzC;WQieB z1_!EgyKe>FEM_cXdU$ur%{s2VeX^M@HjwWu<{VAf4pJC!9lsmxH;WWodCRX9(rE^p z3PcArn<-A%w~g{y+ezkWyL4FN^o`T`cKoQ%(xskoTxDcnF+Ah_s*}Vp8)$E7?$(_E zseJXC)cvczg|mlDsYmN~jlr!IH?14@y~cIU^?#}UOmvlGG6bBp&GhhoN6J&lrqQQX z{1XTD#Y=oaoAa8%3a9||<17A)S5DzkDo`1!BZTi~!N>!D(CKb2QkaWtC+a%Y9^9oJ zM(~70GxhMG?fCSZ^d+k=rYh*bjF^9&;AQePmR)6POQDSG`gw7_0cYxOLEEQ3D;XP& zo^Dt(z~Cv*QcOPZQB$PxqP`_p(sD6WH*u`mdOmuPbo?C&%osCh{lMEgxdn=*G>Ggl zRXN-U+CG9agZLFYHYhLsidny$V~7|L_*Bg_|G_hX^QDx>zFa~>R82U0R!*51M{WAl zCIf5OTXb|Q`AH*KPvLt$AvPGdhoqWFVJMx=i6c1JN2K6ct>ISd-;93iu*}W`->}}G9@oz z!CcH{)7B9C4LJ?w}$_9eI0~-OL?$z{oAp~@_Uv5ZKq1&?& zB+XyHb9vgA+=MgM8Z=eLJWp;0zjKT!b49GOm#2`sMSGsF5FBz4j_o`PS}IlnnZ1JV zA%bg$57j0>nh;*Ta?r1b^R)5GTYX=5&35Y-Rt_TOr0MM1T#U*6uZrs*);_ecF0}ad z94!WL*YCCNUmIf3LzvA{Tl&A9s%w}McFFH2Yfp5%udmlGT$oudl2=+mJ>Do~>~;5> z!1L>2er*^Bf|Pq#xO#nrNu~-!;`F@3gP&XaIs3R$vxg$#aNnW@Z;lgzkx{!Ct3mvw zEqu>M#@<3bq0dbG8{tiNTQ&UD-XDo5L_IimELH24qBJQM^B)QG{Pl`0mbaPclNNHB znro$jJmbPwPx^=@2EY2}orp+5raAAKjI5LM*LzjL8{36yUE}pt<(<(Tb60FgTwFl) z@fN+6fNL->JWr43+J<*@zHaP2?aIVTC6yOuGn+7G)s(R6v)RVzPs9Cn26}L@AS(Xb z6UQ+NpqWVsAw@voy@pTydN1T8WDKuq_vKGZl-RbUTPC%yZ5YbGT_~clSU=Jyr-E8X z?X~I+=n4XDB+oK>-$=?ta?|;E)2yAP6b%zB^u!N-o2h(3yV27Ihl(Go`YKA5`*qxAO=8_aI+tSl8u7)*Dg@?{VrIzfbDBP^8(fF9MepX%mr8eH;e$l<2uKry;#Wo1{ z@iNCmrZ`Mvi-h*EB&rb4vJg}B7gNjtN8vCstyLCsFH~6}yulc31AL&8>H$3HPj-_+ zN1=&|Xpgt>Ru=h???#_5W>}r9Th~-}dg*Svp7xv_%)eA!TPolT?6E zOp2GtPo6dHBK_sJgw004!<>>Rrpkb+p*(G56LIakaGlk}eTCWF4=*!Q<(}+MlX6%* zbG++BuXqlQc4HeinbhAvMrr0&3!MEdo32ae zk%u;_$#3pzBH#iaP646Vx^T@kSu0r7?AM$sb>Ty&wHtRV3O5JurRMWk$CrMOET48C z60NDYF`RBQv%&6CASft+QTO_jB+5lqFfGM^ssjm%&P8iciT4lA#Be4MK?$G^{Zo2 zu08YUC<>1+DwxnaQU1;eIDN_L|>g>mgO!yB9 z#cIGlgR7K!(@8xm{ELRsT46RX?3`bEy&jsWq9$RyhBLo1-yq%;WspK#ueC3;ynu;_ zTnTV5o{Ga! ztVQ9yT0w*#CSO5R%96q)=b3Shb|u_8kt7&8f*p??iejMm#~0cQ$%+^DI&KT1oS~LO z#{FvWI&t4IG7-b&S#`=dyt&1p6w~}3cXxe6v+se;%(`YJr|IW0?2t9NEu9+2;;kOl zPOFJBLW}DKv+xU@_xRMCrQQ?rh@gj?zXb8!_f~#zQe^+f2D$nm#g0FCh|&@c-)|GX%5q1IP;W`@=}E62tCYHvm?_ce`*$Y+$FEU*1m zYVu!hrQT67bRs53S6lz!_YN*$Cq|gB(S(hC|EV7~){`xH_k^9rdhQ1cj4?5WVR6SI zz2Pz1JwnSWS5cpRiGXq5GAq@B`4`7f~j>PtqAm;N|XpngrB?hvdn1Z7Oq+C ze*i$l?B}44duKja#1GHEAHx%rAU-y5-2)BY0wq@nj$gwHB;Vg{FcnqPi#Qw zg<&^tB!i5rZj%^MBM$tjYU#g{U*O=7yPKUCLgp#y^CY@K8XNo z5nae4(KR--KUIZ$HRm|w=VoCF@*d{oSwl6gl3U}+IKj#-+`WQ>D@QS#x+oH8%@c#q z)UOV|xB%RgNMM~@NgM|N2HVTufI4l`->jS%!RxH8IeTn%;_y)~&bd(M1;1KGiqlHk zqTvk}N2JTv_s25og&Yi>i|jDM@5V~cZeS<~=Df?OfR}GZ4tn|CK-}9(zo0hl?fjX( zHT8JRq~XOGL9W~p_D++(jUR{M)2F$EpFc8m9*dy$hF7@-UQG3ulWP=}-}t8R2y5l8 zo$tv(feRNY!H#f+b2ai}`hB+|QkZhDqKDylFGPFV-$K)4!{XO&O$wdg>9j`owCY5q z%gXym9}l>*_0CaD0db4(HL+QmTWO4aAlEqWHS+gqc9eT-Ao+5PyNrm&fjsD87y+`F zH@CR!dv!pcyq5T5fyH4!2r_mwz1vGI=&Do*9Gv-PORwHF?W~yKEi7@ZCgN2X^QFj{ z#tDOmCPN6{=~K=Ar>RDGqe&0VGc))|ehh@9*{N(iW2Xtj+=Uw2%^w4r##oi^8-3$} zxtYKRGp2jkL&J0%asm`mlgX%jR;74v9*|xiNI{St4jn1jR4+!oeDGMrZWf%|*X?w$ zmN+5Q(kM0-vxRh>o~S5G@5&_!Pp|OiG;cfkq2jv68c(EENCi{b_X-`rZx%{Kl5#F1 z1FL$Pb48>;fK}bkw#5u{{B*j~pSGBrSLheb$Rz$yCW3=$I4+STuHIew{N4!~|Kb$< zL-nK6_upqb&nZV}Lvbfg@VOuCY)t*E-WqsjF0-LJL4LxkXw7aiZJIFO7k&@^7%j$u zN!Xl!YAw|w-Ez2&HofN>{DR6}`&`4YU2jnwA?OqpzE1y6bdqDagGrUqbyGsUV&sTR zQADJuGUOb7uvly3-h*h7&MH437btq{1&_6&oESe}-_3BW-bu8mtnr9$;5()2ZK$>z zZhE28A|$BUi4`%x4dR%_VzJVwD7ML|`rRe()4~<(sTs4D^z9jVZP*Vyl&1w7B2Q<+ zJ|?jFe24~4J6@3S1Ep(-xjn7Au7CB7R~>R{hos4%N&PGH{_Ld#`2cGqAn9rQ@- z1kG?IvgilW(`xIM*B`JBD=P~Os{obG;+HGbJovJ;&Km*vKatI6z?M8)0%g>7l@Awm z**1X%zn^J6a4=D`Vl-WIda>+d(5!~)dqQn?c=6KG>Nfn{rC5MI_0_q1|EqXRaU^a~ z4-r|%sj&MQIIRvuIGL91!-i0D_l*Ob2VNZ3{H&^w@|{7z#3wSJFm6!hIryx_i8@&K zfiykB>ATBDajw``&`R;@Ny>8^ub@j-r<2BcoH>Z zRji6|QnFZ~^B1zU=2d7>@(=asJt_;M8It5H^JjD`%-L6WzfF4A#~KR@E7g(G4@8dj zQXYQFnELAM7}R@pH1S2d{dV(3WNx5BF~#o8k>`4S$(k>xpzp-NDpLVR%}8mI4Edr= zrYz!E0{$dyg^_=ARS;8W;8~q7VKkeLWKaD%+&pJv(~AFK-o>!L(t)kgZSoh>Vm{7J z=9PPjD-<6u8g^g~qlf274@|U3oNrz9(ZuY2;gL<$cb;$*R>u@U`h_}O%uI95edT5s z+KJzsB_<0Qh^%sI7~ssj=+$6ruU#)By->BLy5Dj_FnN_Fd622@u4QrtlBx1YWOX7h z7{2XrI$kDS(rb~6o=~)(bFg$+UwO-6)Tv}HZvJ9+)fe?mV%G`(;MYp4=S*ns2@KVn z7`2b&(`FCFd6aU^j{OzeCpV*A21L&p9!xVlGatG4I?X14a>IIe8!Sa^U#To_7T&eQ z6)#N{YZJb~g|Mjd*A0~ya-UcZW)2l|yy5oR{2p`MVtVTF0T$k;&l|js#LNa(pC1_{ zBGD3mMr4heE@PY#xDRH|e{rFFmm$^Uf-jB3*SKvzLvfv z5U;1_%y70KI0eC*=Le5OllpwMo%0Jv&G7cT!&Vbt8s#AG3#a+6ynH;VcD`iRvXs&I zp#hemppL(Ttv;gP@+ z#UUBWFn&_`{G&9ZHK#xlMyBm;k^KDhFlUw^eNDvA1zwt%Z22?BJO8 zw%1a7D09h@2yaNk(^R$Er-K`xF4Xz@eDglBD>O~!x}Iay+A}ap*>{ccP&q}0cU8na zK7vK{{`F_7k`!1g4vj8T2z&D^j&~K39+-LuMN;X=m~+*D8}2X8+c6v(u^F{P-O4D@ zTIy(q->i4Ljl7EnXtf6n=@(74Fdx^@HP7-cyM}k&vMbA)@pX# z*VjoYLpfrGbUC)eE?ghbYF(_OH{wox z$*$O$HL)2@J4^}R-{7v^PyB&8nMb;ceQ*G;zpH8Oarp5iQUh1x-WAASyn}Q1ifP1` z96YF~7sbbru#Z0poV*8qnCby2E(s5cZ!8%d6v;lG2U}yeC)L>_ySNt9?dus+QB_nt zanhlu5?>d3)*Fv+l~|mRyFa>GEi&eDnf;3BDPpz+3n&Ev%KH0eYx^^0FQp4{G_%J^%u zPWQh1CeB-eiiIz*kzlwIPEYGCXrRz&n{FpN@S73vTsH$J&Y^eSbNSIK)z~}rlV0@^ zCunP9qm@^%3dfS?oyh)D2m-tpb=Ncdd3aJ`krqSOy%-)196fQ`rt=EDylg}uk7B79 zaUyp4$uu%B0Zq^y97(4Vr%XYf2)VZ zKofEsR(PXY@qc_4!Dk7Y=@?QJr>feRYD+x1O#WS1(k+GKvhkg}%sk8(hofN;!I5zg zpo8QJDv;K6^rBuW2<6MTRMG`ht`(@Bho@VV8dm#M|OZ*A= zS#N2WFBt>qKNXvZqm!Sak|}oBk_9^PiN~95a^BD6DfpHkles)>X7KDrpy&YTVk;jj z)gJiI^8WjOqWKsUN2oHj6E7ULoq6Nz2|whZJf1V?P+YuRb^2L=JMa()7s65bNTWrc zp;BFS216-VjV`&=Rbgd9HXnL5&}i7K!1$|E^qPsW(sRL;c*h4=yB8!TJg;4Ot?! z`X3y_RSHN(IPd#-@UojN4aDB~ab(B)SMTLwFoZvE*8TnFza+c=A-ECs!!cZ3kkp}J zeR%97dB&JS06K+%k0qf0{_l%zj!hvLONk^GiAo?s3V@f26WcPFH=991Ff{^vn2{_6 zeo=Y`zok`JbEvv^&b6!pq!J&jo!<6T&y;JL_~5@(C4ZSWaZS<3;bzbr=I(z}w<>Hc z+*jp*6%*)Vd9n@V5dEc^19%7?maeC-FHsZ4P<9K)X$|U=jg<|SMj$qKg-EWxg>XKfgM;4xnLsRM@bmny{}hBaT=cXC>J>Cc9*WBm`&|dhs|bSS z%VF@m0G5JrJuU6Wzs3PBdUY5%(s*G(vau(7mY(=F0WGTf+m4i*m*4>gjD4yOzhf`z z_KuIH2|&^k`VNP*qa~0$1ELMoxj9J7`dcxth((_!Iv#d()=6<2519YxDptEJxoGZ}5rBmn3dFZEpn3geU%C<_**iHJ#i@eZ z--rnF=0Ro%BJVW}C@bjAv^Bwnfru`bC^( zW^PXWyNwwJk%t}+ zkRod~cvc_KRuF!O1$!V{S(WgA%$x{JU_s|iQY?re>D6`JGP3O00f08tqOe+0GMI`T zB;TT(HzwSA-qnCq}u)q4ihrl2?9Q@n$ zz9YViV+3bf|8l&8LQ{(ipiJ|512Gn8t*6{1_>R&OFd!B|CYOZ05q@$0&S;SWRsos7 z$mp-Y3;a(n0CHyzi6EkJmNF=P+_&_wiQv%D;^O8Ge4N6eUuHSZWk8s*v)I}B?ITg3 zL{wd;F`V4T6#Fxl3Qu_sR6g?e=lR^laBTbey1aJ5RL%JlgB?GtpU_mTG(LhkhA^Qc z8v|GvU~}F+9VlM!+wpx`rMUFQ!-ZVRJLV*V$7W{Az5+CWYk4Ud^#)@dC?Q7q=IyJ~ z0hIz{;d+iz%4^WtkRhG+S8EVucmv$)qyrqg8-9IzRRyFbV4ch8uztuq;2N#{{xMY~ z*{+Y?TTVILDV%&VPjD9lkGeB7ZCrDSW0vrPhff)g)Kq0C=IuGwx$}JHbVr_t-*6tW zOvfH6d86$8@;NIpY`@;w4m1}Zy_J@1Z=7xQ#1EFf{O^wijv@qFux1o~b*A$C zC3dtJt+$YLahmnq`)Kjw z^G>%i>{{RzM;QPg{SK46+w)jQ-3Xu2O!Qg!V2n$!gbq-S&ECYJGSha%Q@Re@rU53D zDiLP&`?7q(UZLja?T8F8BhRMr(Z|56dLvB_O`#7TvF$oWD6%&+g;<3?|p7pdqd>%Pep*vi67Vxk4+2wimk2Xh}Lv_B;M zV8$p(KR}8K4O6sSDhu-oFR#_a(4p>FbRhWyvMi_Xhkyh@ z`UwfZ`!~GIZlh??5Ha0`7(lN--S}Tt?6qiyOe8=?m_u327+#C0Dd$Nm@9soUV{gaC z&8-TaS#+5>5UsSHPWL<_-)Zm_I?+nRlvWf%J~f-HC|Kqr#~K6Lq#^5QF(OoAv7btF z0I@(ySt0DUv+xSFk6AVIX2Ir6eXXIP;SIMdKCiQ9NCB4GO%kOH!DlN>Y@4ln@ct>y89lsXHhfZMS1 zs~P_!VIBpx%?JiLlZdtP_agg`efah|pDq2vwGE+oV($cjYr1GjH6?ZX=)*5IAIy4| zLfw#$54LRbmjI{XKo)GMGx+9KTCi{!KxarofSX1K3Q?LF^qVkh8^yQ#|MmwST{Wyp zoo(!jT>9`Kl;hD5XtUfulk88z&ZVnm3o4NUld~}&Nn_F106Q!We_clX&u{zte=k|) z?3Xa;ka^Ws>TlwYw2Ma^9q3@hKq4T!R}P3Ont?sj)m1&lwVZ@TT4{BwFx>oFr@-Ysl%|NK1L~S5yMUA+{LR zRM1T~n&Sm3Cb}>g*(Gle{sUM^T=fvdzpNxp28O*u8vJCuIGg9VQaTVz9Ew!LZ!)9* zYj`Tp^e;m5vuBTs^W(Eef%8kB`-)YWT%q<~9|~*(!5!_}oOM5KAGB;H8f#tm%>=hq zn>R(Er+qs3mAC)G>oK6y&IKuC@^=v!5ghr4eh<}am`2`6GkgA8bP9Xy?2Gu+3?J^} zONV0tf>@pQetHNTC1M;JBQWyJ^|a*{LVz5(f&IiAxQp!<0o9MgUtkeqtEho7bO^g! zQ-l+~Cg%O~Uh)5Xi$EpE2r^tA({P_l4K_BpBK$=eo!}P{;ND4n1s9Qsn8wv-YNb_# z;lbfjD3MPDscl9b3LqNXCz8JjV;I~$j^>{-N?#=<4<8R_A6zUeCh?P`sRlcpaddaV z`u82=&xNFU55tQ?OMdBcbgwMU9DV)zT@k*|d7d2-m!Z~6Xz3g8Jq>{T!mED{!@^Sl zkjxnJ>?IW;5FH7YK9Jxl)&7k-68%snYHgh^n=Xj{Csw=m=No^IHT6`*j9UbVrS zXkcEhhN**j*%8*>1kohW=SM(ux+4NZx%79V6^ww-IBstO21#s?UI&&*#04SB?;-qi zHNO(J^>pJd@4VEBOCUNE@E=yjPUuRH5 zooJI#K^q=gPl8;(MPo3$uT#{(KPJXzhQ#;psD&$o zAXLHqf8O}}8vh=~s{)vBX^9oPh|&+KQhONX(oGl^I^-0fyn75S#9NJXU(uE7tE@Nt6F8oV?CQ%UuKqm)vh!sJjeL z{+2?mMMs`*5f=R$F!zS8ep6_F`??(eSnF?cG~J=3QA0&Nr&hIO;ZfA6yojOw~0gn62PUaycIxOoro+vIm@7x#Z} zE}ugPx6;`AKI#w0f{>Wce&K`Vbj`3UC7*3m(K-Oe-CPEO%Ow{D{H}zD>ROf7>|6B* zx=p(hI@G^X>4nD{;c%!I_KO;igH8GdRcnB_{F`y~!3O&62fHawzb~KYb1bO2%>Zad z2wEZB=J{DrITHp>>lq%Y{u88z?176OeopA{=Fi(~VL(P&M*+@{2L=bvI1WK{A=ru? zJDL9(Aw-(h^j?9mDdeC{cqM+FP4oDBExg3(`^)hL_Z*6edR3MYi`&l>1f|KG1r>Ae zTat`pd&y|;4pJQL<K|E!=w7;P53Va3%`k1XM5}!BEr|7*XxaIw0{?uA$rh)bAL*n<&AG%Spm2k~%BJXPo@(z|E~izG zfK*S-FD~Yli}Tw!OweW^xB@Q3{2Xl``tPMeLxtW0BHqujgqz40&w!qNd1y4qqTaZ%$NIS z8W#{()BM=1=eKr^7VBapmT)3cwB||BMbW;{uN3ohXSTl$J!mgZlM-~XwY|_C2d!5; zA6UHie274+J}D;6csab5xceQ*rI=cy?SuX-QrR_=^q*z3(E zmRL-Z%&LlQZJk*}3*Mg-Xq+=ihAnyCl~1hS!t2L3^&HH9pcm<)cMAbNyE{v4r1i?o z9njYIKgft)AQxLzG+q30Q!t<@>8w2_evK9f>Kpos3ABguiMTSDDsBWNd|;~ zNZ4(0Z?^7W#v{<3Ng>`ee&}5^$lowQB4fKo5c)(xMSP%!hVEEOoo5hgJOH{2ap%?n z;^hJWtpyk(r8P%SD|JKAYV-Z=BiYoY&EuyggeXRjv%0?G>xSzfE&{Tf{i>7NVvkDc zw2YI#aw*&)U%A@~zb+^08>;B>tR6`&(s!f)ik7iZr*xVkEyPf;g8ZMI{ zE}|D%o`nwftTRUDj_$FVZ@cJqc0JfVpXE{99II1i6cHMjLVDT^{p4A7Hq&EPZAe}I zWNqnyUj%oUN{cdBry#ZEr)+geEn9TF@RVQ)yv5z(x|Q4rj{{HyRzPv!nIRI|ev32_ zh&awvOyZD_<=L()8aR*w0Y*5Lkb7HC!o$bResx%H*}(zF=c~-W-Jk2v=Ad8(G*wse zR9_kS(P?b+4u_*k)KM%hj6wEzI+F_2(tL(A8TJYud@Bf$wFw76PbFRVRR+2~6L|Z= z3{=Nf2SiXUFW9Aar-3Lo%98Bna0$Dah9 zT^3XucUd~fUZ)4pt#Dr_iK=Qc(vTdS+(W(q{=7Cs8^HyPd-jv*>XGKo+X1fA@`H5eH;j^sHSLN*x2YAqzdO(gP@M0a zUxcynB3y$oO1DS%3rbcd`J?hTY7;msgIy)LWFqdmyK|f!dAQ*BBNl&sLHmD~p7 z?;k&IhRO;lyR?)Ul1)OkvUezZWRHWJ6xkGxP1)IdJ4z^f%RKhx*vIy}PM_-i8Qp`ecjT7<9qcBZyu9F5S_)|>U4*&CC5&bGMl6iL{MvrPZ29~e z;;Gs8vWJ8^UC80wD2c1YGE(=8+bX+tyd%YNc1g0r%-R;Y4+Bax)$1Ckp%|zC⪚k`QCiBq48DM^nqP#~ zxow+(nGiDukbm|-P$?)0t%xa{fIHDETZ%MHu?o>DghbmWCxl93Uf2nrb zFz?0sz48J5N{p@GBI6?(k3sz$i~XO*_yoy3&|)?^S6HfSD0A;_wrci!@|XdE;^N{L zr^g3U-g^KlexX0@9mCNO;`AosWH0`OiAgJnDA=!Fy-Lm|N_9a;`0QR!N=9sNf}%NW zI8SOrB`hjWT}N?!P^WYuGj{M;NZNj^OQeJ9G@0K)pXQo0u)rOw@EYy>RPePo83~y{ zL|%grmw5QMt05|9qJ>M3#US|c#({M7pmY5qBltej<znNLo(jr&6`v;0JEphoqCnnX^@glOhx zoTwK%p)cg{cH%T}8mw=tW16Uf>p|8;-ph~VK-t8%CT*5X6X;ZuANSY90O+Ek6=N*U`21+t>AZCjMu60+%`PVwK zWnhoS9QzcS%kyJS7gNJ%zR1jool+CeE-?}VSjRB{3&P3Y?R3u6<%!D@M_QzmX_d^} z)3%&cfjL_+seT|q?-6@j6v%Re-%2sVT7N^KxItM^CyP-w5qLZ*Z!$SF#<5s)?(J^6 z#3a;#m9=-H?qqNDLiLTwP0XH_G-(R-W>%?rXY^=Y7bWx0blFp&;~SNA%n zhfk;c8H26YB0ix}TjyQPnbctG7G$daqrkwyli~;&-ntQ1r==|Fc(H^JeR{OjtQ~5R-2Z9*;;?TD$ zX^TdR+qXhDBd1zfS@djS53hgGR{Ci5O&-u}MNj5;tS0hP(}mMz;Ssm;Mp{v#o!fM0 zyZ*M1NW7PokAP==Ym%Pe7~`ecCjgirTNM~3{4$P~eyr|M9c~K$I_alUv>8-3vK!1= zKh*D7K!=KVq;f^hzB-$@aWl+2lpUk@dRq=m=_P_DaBkOc3hGFg)gCVAxL1BpV-_Hx z7r4P|(a-rha{^TW8X{~Mxcw$f8y2%HlzONi{Kk76=QwhhP5$8bG$3=q5l9%jc*TNK z3C84#;2%1-q6;Mz)Y^MweR3hiCpK9wv{t~V0U3bb)W}l)a%DbUSe6%%%wbXm%Eq@F zs6Dp}wzb^}M0w-SY3`B|l}LX)hpzR73k2)q0=x81W$R2A@X-~rSH$cB`fLXb1TAvlo7WbvLp{M~rvMM{P>Rpd1FYB2D-pzj&Q*vb#R+w+R(|jDXuw z@jDw!4nw_Nz2pFrA4B@pY2OKt%M7E^Q|t#$V9M~kby3G&w1C_-@zurdLsX{F%nvO9 zuHgS6EI0bur@HGd%)HBo1eBKjWnLQ;ivkeh zwFk*g{idSZRA2HeBznx!)fLC#nP5msFMvDNs`EPUY;Hv3`s_u z>XI-P>vP-A!=i--#@9HW7Dvuq5d;t#L`^#~Lg4VaaxYWXtrKr`X)C-Bjuv#D1!;(& zUC&_K-iEAOo*Wrsr(m&=6i$(IfW(YIm-XzY5^Le^P{#p1&N1DW1 zM;c^?7nh^jwNmJ+M|bD447GEwvt_U7J3z}-f6iSxluNQblqN9@c`l%L>XkEz@7!79 z^yH5HV&fqC9=W?9;Xkz!eyTApZ0v!%6p?!`sIK!pol*ogjM<`cTK?jxe!_Gg*#tzIKJQhaC?zOsd{n5E6Mu7B?GYqEBNFQt|1;c+`?&)sDYGRj~U$n7h2VZ9C^LP+*^&jNrQ zw4VbFkmhhR>okzh0CgRowSni+C>*isI}v+_eQDcX%e5cg-lPw-zO|MI&Q+|JhB6wN zd>_!u^ChNT4@hYp!rh&5tju&AMGDMrqmBlfsLsoc$nAo?iF%FqvrGXmwboW$K4;16 zS`Vm=(9{4oX{1EG4UQLC{0lzPN<7VqLZ?zM|W{yKbiSdt|DA=z?(Z z***bYy<5=8kYr>ztLQ0#SBVW=Hcw@wtr?lyxEXY}<*0VA@Nj7jUPmfZ1C7ncw)NY; zPJBc46Jc6mFxpY|piT)-!1+0ff#vBk9)z+ov_0Ev)IBiY%Z#?yPMjDhE)WDuQBrEG zUwUt?^QdkwsD?VL-a}T-{-a2EUG%zBqqdXXQFX!#TiT z5fr~8921Ju6TuhPzVs$ghhgX%SspWIggFyW34Z)A`F zsA&+=T^wkKxz@7RQ{Gu(G=95#gYEl0WvwRua*Khx#)r6ddno9vhD+MYu5~J|0jDLD zZcA{A^|@hosU!Neia&|#7Q`sH3gs=lxJFy^;Fx7QGXy<*MRH#QtquYhpC@x!({_KFD!(jr+yOD+6 zjY~js1qtEq7T*B^wlHKEf2O)PHNoh_>wJ!NyFU4|hP$6bwEbOD)HNzeBY6tx z?u66bu}uqGI!+VN7y)Fi;K_3KjZ#$SizuTCOg?M$bhC8;^uR5M*@?OYay)DlOhVoj4r07eX~ycVUPW%gJl|xtDaD z1LVG(74UUy#SH*Tk7B0N4z?HCs0}=(s#G%xdyPSQo8J#Qk`4*Pc&k8}8Nfte{sy;* zjS;z$YVjfW{d=FFJ`R|O7^JSvyLXB|M@3cJ~R9qcsl{x|Hw<6Bs{@DlaSixNicRa9g17foHJmRbz~1UnWh% zu&|LS874hcxwz?LbSrjxMZ*0%Z8a1x7Di?mZ4&JD8MEE&SCoS`+Yq3Z))szPtY#^5 zQm%KrWyiFMuee9Lw`}K>^f?%01g9{be7+(R=-qA9skY#y6o=q+fPp=BQ%lSLF z?^8MS3un$4g0XPF)o4PfmMm3vY|k zw&2UYoPDc7w=Q9R*wn1rB@E?c_}%7(>*OPvpY;J-0&O$SK0$|oOFo&GH#<{D1>7Z7 z?5I`xkuXd=Uj(J3cAmK*J<$Q|PqMdTx)KG4w!UOOB{v`|6bHp`;w;Kvq935Zw0SS3 zad4Noc5iyL&`=_RcpwQj(U=X4GgqsC(kXA!+Up07n~Zulq(7Y7u5G)!7^fV%GK~$W zyft<|0Al1kvG?Wv`1_HEfgdzF9QWgI@md1^kdIK4YmIvLv-IxT1bIbljf^wKIeslS z2@y}!!<9wywvUc3Shpif*Sc;NPx@*I!y`d*F@g?|=cqu6CeE{H@X4aX)N z1R(}&e?IoRc=g%p=k^Ks{G%uMR4+O`^~XqW(gI%r7HyDpUYGd=`aHU%XJ!v-L`mCL zOwZw%?$iqw>ObMeGckX?CCDRQz(4ieMW<&_V$kOl#~w-QAAo1aA;1A(243bndwqsR zBeaiT+>|_-nOaHnTnnogGcr@$^a_z>ng0B_@+%hO)sy;`pjL3P{O7y*Z4w$wrqV2| zMNlIW-W@fZXQ_>$y`Avt^%$sYNb2VaOSXh{#j6wpD;oPBH{n;HA=fTD?HAWr;=V+g z6W?Ny5XVkmJ+9i2|E%nv*L8w2+gz8_&eo?mupW#zv1@i&`>~X`XFYo7!nSXQ8VwKk zW!Wu}f>HkUr0kW;bGwuex_B4EE8>c)y3n)a-*>dWJ%RXMREc73uU&O-KM`!>+G31hDM$b?hcm@btufhYVya~z9iVmmZcE6VRC_PD;Jl>L{j(aasNRZ=)@M z#xO(H*O6lhyKbZ$qY-Dk6K@DC3X%YvreH1{27OOH$K^zhr9k7T3v$1rtcZM^{Gdt7 zNRW8X@ekCv*mgOmJ-|fGZXi1knPz~F?Tl?&#kO6S7-xXl>pkbPiN=XiwAGz#Jqvn&HyAs)&5 zi>>Q7+tec^VB6{$gQFtFP#AK2<+XG@wlcj5gRH>D5Q*I0fT$PC;a%3X;bKLvgu~j1 zxybdii&wD5%cnK?-6!{jf?s|fq}0O-h)lpyv*_QeDLR3jz6TkQb|}h$aa*za5H@^1 zs+RO^X`RK7X>Z|QywtJ1ZVKrAJi{j0oi$L)1rEZ)_IJBE-A+7vcS@b3v9j2UAIp+p zb7OsWNp%P$IPbl@BLE!Q`$7L?2F7VXQj@9BjaE!Qzc@7pKxU~|5@n^f&k53r##I0C z6=U)#6(ivP+atYE(g17{Bw?>Ddc70Svda_byvA0b37IK|=;}A|wOz9eAmR}l zFYbu~<`h6VQAYZ8B;G` zQ~~7Vk(G|BGFvaudD!)g{vJZ-;d`4*lM_ELbm_|z8iaBDP<}aGj*}5{o7Wvd9ll)R zE5||=UA(=v`zrY7M9V@wYr>$)=Yv{bgi)p`kyG3^o2HD|-%w;(g*K^+El2|2_qd*S zV71{}wYSG|hs+#`WPl6+dwZXupOk}l1Je5G|8}0|G0mGfm%0L`BWD12D3ge~|GS69 zl#lQD#2G}xArF>0f#E$sp{bSIj9cuecPGJy^kI2i;s}#ox6kK@X=bZ3K?FP6x1!=7 zwZe!8>O|FCnOS?G{Y*%p5GjcZjch;bi{Uc<1|wlrGBmU?LnVM61{yBm`~vugH(kc; z{JA)?_G&djHS||C3Y5tGdC@5c1n(c9V@;E|{JbrhT$=)Gqs)S;iHY0e>YfbDW6*Gn z&~Z=9YcUvvsfq!`QJq*}Sou>`Pqc1cnDCzF(PmCUMtUxa8^n zJMX@WQ6Q$yypdf;%xm7I?2rXt*T~Zz(5xG+AxL8w`q*(euMcxw)(yjuZy7Gb z+g|26+pJb17L{!iCknpl zc|T=U+w|lWhzk5USolDbisN>f#cet0;EMxPf{eeruv<0AO|USHn*Sc!F__d&-?_B( z!`22vZM^T2F55BE`>8QD$#dNGBw(s|KCBjhqn%5~m}hW(xIeB@y(N3TIyE-1em+uU z^cY?I?Fy`^X=`#k zErEpUnSUGpq+R^|(4RcSCN^6>POkpmBa9wn9{HexdlyJFBC}Q)@i8@?NDhMu)T!go zpF2x^Axuf9MJcvhyN9{Q3>G7!MF93CHH;CLM(#$4q)6Rb<;@ za&Q>+bUuGm&o{yZtN&t__Wk|URVPven#4OvIXf<)zYMMf<P=oa@lg9bGu)~$HU}4iX3re+!|##X}dpOnf@z0h5)>HN9p%NAwmP0Qo z-*ADouzLS)QYNQ-$HWy6Op&^%dLtR_F=3*|=(!USVL;8hMdEEi+FfMa(h53=R<&pQ zKteYSC?GVKJ`$*?CV_^~GI5z%Z^=8V_d~@hAyD^kq?&dA^XScdf;j3mcha|&mb;x2 zNmX(EKqrdGLh)|mkcxqOUH;d4uM_udpto%22~@_)W(ocRUW91Wd?jO~uhTZb*(t2#td-Ik-#)vgqs#qB%=NGr*#yHp_XOZ%)b}#U`h3&elcv zu0aU&q&SrBZXvcB?E|>O`k_+9Hp692u{TO|Kh24~!)Mag|%ojnV$$7~eWKz124sx`V!s48gqZPCu21PM!P$(9H!KyIdNGAf=R5 zp#=>2Qr<9I4$j`}->6e=#nsyXHe#o-6XTuX3|(CS+$1m-Y`K(ye<3p<-HIT4Smdd} zqEQ#>A#51As#&LMU+Wi}0T&Xdsae2Va~%+~tL|lfs`t3{CyD~AqdnP{=lWZfC%P8^ zw^Dum>d|L;(9=CwuHysN(reLsg91fGV?wJ>&Fw98+BToNIM=j|)|-14Lc>Ar8s(2} zx>=2wZZ(an%#IUv9hPZa=%!jPpcfdse-ZKhwTh5L*0z;@>l_(z$;bysHD^!B`%7lz zD01cq*7#yFLiL6zhG>dx{k|`QG9rHtPE3FHPDVqZ_0jBR7o2K)`CGj_^}Ug*JyQNc z%No_((eLzE@@V;>IuQz;!HGe>95uzyQEcY6H|?|!Lpx|C{iERQ247-eC8|sJ?M9wO z(4jrv3szY1%-QH)4Pnqm0BJsdS2!^EZU3)#Tbr9_Gy7DzEA(I%Xve&sk9%JBgpFK(ey6x}R24IDi ze0dl6iLY>`RKgwfuHaP1D3>*}HQw=&BY&4KHb+zT*dQR>k;Fvr_=y1qP0cinfl-Z8 z%z6F63t{8ooX5CSMyoC0nw+10?(Tkkv5e$X)r-II-Den&sMa%x5DPr)e>Diw{2D9^ z>Mlp94Ldct|1x~}^*>_M7-Cd{sifQ=U=udpSF+0KzQ0G|%n5CI#JX8+5L`IhJ17eO7| zZ#!SE7s5M}iB?0=M#EEx=y%zrGTHl9;jm zV*ure=DDASb!f@R06???n4E_$8HMxRUlz1*`3UT*A2=VLBa+9!G~Xo3{)@1N8J^US z^7Q|LFJt_QxePO(VOV(Z?r>}xi5qzhMXL!}NF8-&oBkC5sJjG^ie{fg_(%S1r~h1x z-L=VA-%_p?WpSUYkclFE{HTE{;L#mi%x&FJ^KmrQKf6!-X5My0Dk2&g#U>m=m}UK) zGQkmWb`+|mTZw;JT@e8F(EiBDe{lwHB!VZB21^_JaP2KMK@7N51)}=Dbf@#Zpb+|h z_#`$Dg%A@?6;ZEZT!5Gm2Gs4JDMtIxdini|$q=7}9+0(0ONxnnr05gRQ>ZsY6|!Yc zVKAMHbcB(w_=WRcxt6aEbGEsp$Lj3$OJoz?_zhKrAHK(?r88gu*&oOwtv|`#G4F0= zA<}f6i~@tM`+po=O(~gs;fWt;cVTS-;)zcYGIt*{Jpwn$j3>kO->6&{C!LC+#K z0k)AqFV*V<;F!I^(7)^3+PBvZ`5Q`!&`{36>DDuEf`>ZKck!Ky-b7#QuZ5B|Z){4>%xM6>4vzHmo) z{s*c0`<2fm9tNli=XzaLY)|MT75?GVdtLDT??s>6{L-KKj}NR;y%M0?QzuK=DP7dK znnFeQNSKKWPdA`PUi(k+@ZSdRTPEN_(qwbVz&j#-lpg0>nIvwqJjx+^#{q!1%|qKM zy#Jc@;$VEpzop&5+}(NbpExO-yFzBf7%(c)V`XqLcw8*!F^K;2({kWn9>S#BzVN3E z*wh4|P)s8(ABZ9IQV4OLLAA znh^8FC)_C*f4R%&o`VUl91kajW79$ncQ=ZlbEW_B1b#i-OPWlAe0-ED&m_PL z1xQ_W_2^v9qOYzP*P6Ho)^|a48v*R6CC7u#tLT7Xv=D{XF&;esIYwV8zyApE!s6k3 zjSY1Js@(VQ5zH?ECW0XV_+kFQKyiSK*B0G{m&S)XOZOcYm9@$&qUMT0gwu*K#{_zD zWSM>D{4FFn&R};a`wy4TKMVgR-u(nO;hXg9HBK1f05`%z!z*!0JzcyL<2^t};t!O` zRJ6+>E0znFS;K*yvMMl~W(BUUkxx^l-l}#0P!|&>lg{S{5@Z;M*=L^~6CJGApk4rj z-uurRuzsNm0Rr?ww(fmqWnD0#nxbX|{`*#eZh$lZJvwJRiO!dp06JE{u2jZ;E=t2a zPUK}PWt9YQ1APHJTzf%j(0nLAaY5IR78GA2fkQHl7LXffm~lsa+@ZK7A5Lpq+mof# zRtJz}7w&~R00Ie9?GL)^y?Ey_AZkj5cjewBoP@B3FbV9Kx*myPgh_- z9S1DfKsowL8b-AHyE7nD2DMSSPTB$3U5F6S(YEB2d{5w02q5i#3Qw>xyDVfx1Po$z zLE`+Me8YTU0ctjFO3pBpIK4o@y?RlgG*zD|bOeT)w|u2PdxpRH9L`_j^?yzv@Q{5T zQvPBcs{J?XkcN}^!P0B#>JeB?&|EU3irM61`Zgv^5 zR~r6|yzso$gwqq}bx{l9?+iG0?8*XtH{liew2t?}z?PZFHx`_uacoGr*;reEnT65Q ztw5a(RF~dASO3Jfume_~(Z_?`+H+#S3F~=w&OC2Go;{BDF1S)s=BzIBN`b+IMFpG@IV6xw2aQJ3q&GC!*p z7>?OB$wsBW3{qirK>{0?N4~KWu}(4WoIh6U)*Jf=U6)#U z8TiR#LS(6O;LWAUg9Wr%;=sVCX5%R6VW#0jI>H$2Zs2DA1N1r5rg3TIRdnl zJ*G+Gg7myCFqB0CLx2n%XRuNHZ-ekxC?aZi=Q*GdnpfY4rC?=sQhWyWWRRHg))O4P z9Qdqe54wCmTBGiT6eHbN*p`3{Um^Shsn5IfZw503KdGr)fltw4Q;MZcK)+7*NqKmg7i=z1Q!}pr zqGx;IVU#>mxd0!UFJ%+Yd z$XvLw;y!#}1_l0t2o2b4z&ecb8ggUowj3%jKzY8i#DN%}*v$^j%5ql-sZqmHH45}U{a4aXEqeEk0+6{wMUt2fB_-d$n^pj4W34dx z$p5AeL@bkJ{%QQzsbem_NpboaWign`a(Mk(xD}2w#2I0wk26O4FFs8;2%GWTHF>Ch zasMAtfG8!7y6RWJPvnD1AiEnsW&bPo!BE<}Yrp@B<#>odT)^YQa9QDZN9&){84$U@ zcm4=)PX(Dy>Td-RJTZ#hA3bm%QxFK5~xhhrRqAGGR0 za2&W=OVldG)QVYk7Zv|9PyQxH;k7s>-o5#Pr8OQkBMMc7-=2X)5bkA)P4jc=77rJsY33x4 zw|b&+^@Gl*UkMbU7!!z;?}!qCk)|dn1^518uLnN$44dOaAwC{F=@}rSnN@LNqE0-F z49qY64oN@2CA>mSz1=&esjA&xZhsZ0=rVYIE4Rf?{zQEL3|lwOFvtmPp^HZOPZ81} zeRzhG|Agu>|F0D0?)fyq#H7*kn#@>c#I4N6u`Sh(bAWXSnA@dTS4e(sUNj#GAC8It z;EBimVE$;5JJR%#KAvjBI_WKCssCj}LVp>N|90d~o{iR!a>@m|+cxl)w6Pn?v|Wm3 z<@bI3luSlLRXHNX2&-=)iYbuk++^Q{yPqbh#G6+9ad4(@>fhHVk5coI*0fh8i;fDu z@k0=K3|ngCFZlrLOo-#R74BLE?A)5qhj}lf4-}lhK8Hs_NX8`QE!KvIm%^4h#0kGP zfK5r*ahc5ffsW=0d};*v$d}5CVU65(C#C>z!~!4r^$GA|Mtm}gzaaGRW&6@H&F%x< zC(m=w6N_=d!67uKd>p!`)sl^VbPh+=WCCb6P}`m68NuoKb0PujG>BoXadWh`8z0o zzVag8d_1R%mzK1P!E{b+dM=`*L~zW)#B%u-rwAD98Dly<{QnxI?z=vrLtnqI!#i&w zEp;9Y%$jK5%+|wEAm2dB!NoJMuMm=e+oO{bXM}$l$EGAQ16Ntfl6&_i{NpEZvrGio z*k&ZDDWt!i8eBIS`wIM32H>cL`+8&X_&@CiB&+|p@#pR59i5y)+>r~x?m;v+IGyqF zQjA?zcM|^Vv3wUxp~5!1H`4B{O-*=F?)h|Kr~Re>x^^&yu5o^JSU)ZfU3ck7GQVv7 zIJWz#O@}8+i3lupwx&ZHna6ExHI{Cq6I7DKuOE)%w({bAL7zP*5#HK!{ zeU}N#ngy<(iOo%z4sJYp%Q68j-iu8M&&vdfRCV0bYyqNRC2Z`)eL4*dP=x)DOA3D9 zkmDdF8e%2>AvG=gDh@LTm@{rFy#M_~t#5P>_i!z1jK01leL@4~a2npNP*ut0YyUad z@d!&(#TG0NOr@8fPzb~c4neskH6eHM2=zwotn{|`$;iAz!|WWX{u$xs#0C<<)o>di z$jW0k{SBkf;I|^Ul3UX2L%^W9D!BEXxi3>>H`}kPjSoHUVxSF9)wYVVp)W}Wf`$oJ zpxbP_CrwXiP2l}fg*Z&!a=Ra|5Jy&%4fCWi16s>6#U^v9Zpi~PYxaOIc zRe(VXrdM|Xn3M(#jQ=*y;i(Oy6MN;3fQCV>H~OlPcwuRpLP_Z9>dN8_%QbKzN^il1 zfE%{dz_7j2Csr*89Bo;PzhB?Awf?o~_;9g+r}v>TqwnwdLUnI257dgPx@UxWVVL zjb22OLYL6}d*iT+wFy zkdJ4_T9eGq$$=*-Obi4*i1(2Be7hUs0js^wObrx8|NQVQKJY<}+v!U_b3IGUV4|Yh|7A8J*8jH(NY>D?|-}9 z3yl0mO*0Y4a~zqslRfTR7d2g8GG;u|bYCrM>E(b9+ICt*Pa&?8B_-%CgL>&^Gi@dx z;-F)RJNW=(6Z|aMsfhrXTUs7=ogH`m8VMyg1JgvOz?Gk%K>iHR@g&L&6{(`gLXr(V zI}j(JR~4dUVMw8z11(jBvs2WWzEwd{GYyMKSntq5lfM)64L90rzmYNH6~MfW1`%-N z)$NTyZ-)Q^J|5_SR8*(Z@|m{BL?{Zbl^U4}g`0ysT>k6`aTf1&vSL8J<_`(IxYI6j zJVlxbI$EVgugSfrNqyaCBcubh`G#M7AkWuxy7p@OW@h?Ir@%*pVCge=x$4O?FP^$1 z%gnU58SDe7B8d}DlZ4LWIAnmf%$HJ-bWl?_h*J|X6#j~Ujw&EIkY%|vegp*xrNe;&uP8;3^G`d0Vu;Ott;aU_ z3olsBNQ=cb#?uXdN`oGMLiMA2JqsTc<6swT@!o@>)wA30^4cre>4Hv6oGpab>V((_ zPgrrg>pP)>#GiHKLw11VoJqjon&Rcs)3Ue_a62gmXpX~oJ|rLCU@K|APu%h?p73E? zd1}(b^}H+Vcl4 zBbWrirrAebzwvgY{w}RFy_0I7!8y;(AOlKobsxv|LU3L;8ae!PwA-sn>?@5-8)y#@p7r<@pvk6I@4mp=2ZHLG z`UVT?50^iMV3e?82B52+0kP4-a>pAwgN!9X&S1ka8Z~qCs2H`&_#)!ZiWG17sQbXCdD?(*RlQDtLeu_6i)t< zZ|J$=_W@)aM*i4et-U=EfqR5PK+7xG&pH-IU9+~!?|BWp8a?vZ=7@9_na(udoKu(4 z8$U~wnsi&4GaY2-wed$fhI#gVmD>|usdnCAdT?sr8P+GtSnEP2ow3d z2;CNf_*V*=tdHin>=Z^l4p-)M3hIa){lWfWYl`RFRr!XHpE)7e3$dbyx@%I21cy!& z;@)Ok{J!KUX0MvackVu#hCB55JxWmno4EQ9OMvhE9iB8W;ZGBE-LDNC9WX98!3%rI zztMo(XV*fn(cT-d!C@mBrZd|H*#jr&p|4#r2h`lO$PGWfNJi=cjJ((3ev7Dn`)NRI z3c0>i!-!76=$F*T2qSfKA>~X~^0aj@q%4oRE5}6+pF%ae0OX*%!y6iQJGe!@%cXXP zer{>W(%?w@aNPwFPa?e2ZIof|p)6`#=KP$Ic8cs~(klk+MWEBv3x$DGTO@Sd)8)AE zEL+L*^FAj|n=N!SD)~?ga%9CX1lf8fA?Pq%@g;nCbazt=FCA>b>VY~ZZI`|*fUQy$ z?_&`Pkjt(_VeNRLU)xOc1e53}Dq4V{3EM-jto4s~4E zzjpl`?xEj7Kz2h)8G`vjnb&UA6&xk(A`hngw*9u}Kh~IJ=&%bKcV{Z?zHPz4*?4@5aqz7BkJ|E zY}h28_`p3G1TH&K1s*p~CJ98Z^rE+}@y+7l;1WW?V@4#nJhR6C7E}7MlYLY@vHC&S zz1-MU++0~En6ykk2auv1Hi*zQ&5ErRgJeNt%a3m_%Tj|QPiV@Nj+$GRc|Z6Pw-oc^ zH4fKc9fYw$_u&S2XUK3+(PeY3>+fJ(Nk%gz+Bk1)<$A1mD;+gQ>hh zja}k^%msNFxx9LDiI*j|iGt9*8p8!LWbN_v_7bV+(y8rnSvs45^dixAQS>4OX>I+t zrjb!pQZ({;+qJ#Usj3U#@=#BW*5);XD(k)}4@Z)6866y<=D7(X}hQ(iNaneRNmHx=d{?)m8bKmXW&I)pr0J#p`+4 zm&`62x4`>#Ai`t6S%`S2WCj>tvx$VHpZmn<8AZrpEL>Cktvb^)N|U?HYrNxd<~rpa z-<~TuhrIjCP*_eeeTg9^ydC`fu0e?54b?}fHw>F1CNrmC?d0T{y0@`asMfEQ4Ng8p z?l1U4@eAjv^-Lyil3&(xf9p4sL^sQ}LVSDI)nP3b$uW3^&BgY(9>}xTp9*}_$(S*e z-Q+naX061!eTrn*js#ETQ4P}HKws2 z?pCA@C68{`x^8(G-2B>6+~36%%=oU2fU?PY(+=it@Dk9KS(9v#Lkk@vjHEXM%OKU% zZ=@?l<>t^$6Pf6jmHKvG1|~1Z)eMaN%->8E%`X7Wg!e|xru?HN_3J=uH zwOQhQh3WzyNq-VQ~<15jfP+~Scfz3gWZ=c5n*8T;4kt^{yo1B%Joy@Hu?de{T zv;x6e(`hB`wR(TjatuGe3B9k`Ihzr0^LhLu=g)0=>E~(q`UbhCX_o4Yf3{w*2S?#E zxG#RxrauvWk?E`)7E*~{&kE!L?5lL5+1U%FWwUM; zCKrDTPVteYqYzCAGr9U%qy17yIL{z>+$hO6{T;vD84a4AJ5IFK> zaU10!`06)9SZrd$u4Z$ol%9Un_WhjJLE^qS>}J}mkYKwdbe`f?M-8fJAxUPt12HMo zLQ=}yS#x*>(W?ns0m4h4da_XM5z3&04%oKrE%js^jrZG*ve;tci{vMWSB7Ram8p9q%!CUqdk zQ1Xe1(2Smo*2y2E+p)aL)U2Hd+$R~KjoQQK_mfB_bOO48BSt}m@d|ptpsl!#^=Tg^<7}(SDWKWX?3)87f_vxm>SB@tJ*mjD z)-Cyjeg_Trxu&TOtdl--qS$ zW0TH={HHaC{%2e@W-DK}yA+Db9ZrUx0Y2 z>MO$3jpyTrVCMTsuJ+GH!^A1%bI8c#ztw0utnFs$l#Aex(TQEbtv&T(VqalwwZzMO z$zuDK1Bu?F@nnG3p1Dib_%0dKY1lp(eBHSS*`|Ay+nosH*U?)V$QIP@Ji={!qi+H= zV^dTQg{*1myA^GYB_*TxM3NZJDn#}4kIW7Q;xo$1NAvc)Drrk>>)3@S>P}6yGV%-E zwe3=ehIZJ{ur-C0)QUufVGS?aV1m{rnx0*Q%@eai(r3`UGlc*Y?BwKG&T;(}#OzbT zU^P&R(4^--x%DSwHoM?+SoQ#fVqarpA)bb)kE(w>H81*_-I=(M<2k_Lp%- zp?aFn6$Zv3yoUnqr$czRc3ijIHC#9cbYlD3PCJPtbYj;JGLxThdwDtL9R`e7blIK+ zG#(NcS`d&H+Np#MC$FsUe8KMAuw&tzDs#MP)J-%F@oZbZrJ5IcZV5-P$6b%iJqz@Q z?^2sYpM(qnXF0^Fb@J%T6q=RC_wbrOLzoI#XoA<@qG0J;F;6UQguad~H3T**O406S z?PTQ1x=WD6GH%_Mu~s^4Ev@BIcC?xs#xO@^%;M_J!6;TJYdx=XKle0yu8zJh*we1# zr7M~EBgiv&Zzje*u`gne=Eus>vIZi~XxOK+Oqd~(S2#OOCGKd((^rz?eVrPBFAz)lZ(*e!+3rhJpZOt&M&e2* zBX=I5abs+17b8c8I^f5j_S~xMx(49=BB(r4yD?nu0(MdkeYlQo5RbVJ3-M7cbkzXo zB^C|$aI)KctO$o9*2wJ;(&lh5D=rd8=l@;RrO6E=mIaA>6EM{D*O|-vD31lfy6eYl zwvb#4y^HRqmJrzt_y?@$Bne%cxz|4dlEv7M!uJ9gy+n_Oy9Qquk$Ozx`qm2vKKEXbcx0|v6zg3YO3!SyF=wwd?k^XXq0 zGx)sC>^M>`AiM>OPC7E$Tfhloo!BlEK@N%T+#zvaGJv`=5XWdz+ImcJ$gLAG)9PT+ zvUAmY;$iAkZ4}>V{!|^`j8e))v6g(ryia=$PK8HCI`8Pta zW+p4Sm?Djsz>*%OFABu`O}?j_=~dI1hD#eK^wk| z{f}}bNQHTTR+0V^7-a>H`hBA!hFKsu5^0pNq^6;#mT;+}8%m+#TniY#4u_u zujzwX&XUiZ^u%Mci8@o^!SIc7Fo()w3mVNWfrv7@Vo(_eIA<5GZwLa>DFU?ta9F~f z0n2duCy2&xTbK4oEkUJyiU7UX65LDpzL6_GwN~;JWb`sQxy7w1%eGD8MT1BTdy!js zHe90j^az!_4ulYY zsL|2V`e^$~n^9Z7?pgWE(M`%De^kP7_vsdGtq81k3Kz0%4iOW2QE;s929P0WcRjeC zvsRI)*AEKh1u7h4HBfC)msUg&{VC0jNIeJOmL-sV-5Y!r93(OqsnQFCKoX3iXN-c_ zqK(Q;4!qMa8SPRksIn>bB$A zQ@`BJQ;>ox2rQ=f_QLCT(zj-?4WQ#0#>{c&_#@Rd*YoZ%`GsH4H|74^2yWp1+SS?F zxwBgU)x`)4K2K*8VZ!Mk>@${vOmlTG#H-66t#hwkx%T8(7(`9}f%*)u8n9Vjo4`CukF6z$s!d zZ5KN3_7_o&C)S>bz9Os(lFeQ!4c(_T#Py~H%PPAyUa1PHUW&wGnbF-omKiNDqKnCi5Bgw z+Jgna@=$?}HyBs4qpw++U%qC9*Mc}2NA;3R(~)Ne2VAo6IzhYwdhFz6q^be0 z_;B_Bg`IB{s%0S%iV@#8LOw#>mP!K@_AXX!0-bEC_+CK%Sfi9+W_ zm~C@G11RIg8?DV;K3Cn%wNht#Fx=%8ErPDi^(Vy>*bX=d&D68%ur#xr!Ac}LqkJt`2w;x*w8 zbq*Nau&vu2{3L3(;yLzOPlUxQ?eJ2Gu`ydW82lnaO@9^J1V(K8c1=CdhzKOWWdlJ+ zWA@#62csW8ntn_SPpBY*>AgY1M^e?t`)G99$~w#_EvF;DzNVo5;mH)@H1(qaxn}<6 zpk8kuX_sup+5DUsr+1}E|Nl=fw-QPs$PXY1mYdh^Eh?$X4zM;~ry%peh)DMakG zv&B2>E_KY-WgX>9@C5^&#ce-_N&JOlml4#5qcrq-IZObxdd8p#Q4ueEt8OV`l{y|J zL;3=(gttND{pDfIcqNIDq(9_^}np!27BbV%o zf=@u|a7_8Xb;PpR$x6tL)r}6)&>sh{EVQ^s@#kt?U1@DvJKh|~y(F5T!3I!juSV?K z*n$e6&rWmpUT+rSt;@AGqvVpC?M2mLZGwID8cFdCCo61aZLW)JvL3edmcp zVS+Y1d&1~%jl>=Du>vl{<9>+~JNGz0mZT!P;x`SYX<{EBHr{fi4XEp|HCc57_*v0@ z;A!(E7OqC&=x@q|!A_X)l2QER_OI{~`h9lzL|^#Eae2aP1Kp=4D!|%Ny`<#1mzfGX zcta?v@zX3vB<$)Msr$JUJ2}Mt$38>F-`5f6vbj&qj9B2aSALkZ67+W@2@8Vel68|F zK6zj>NA}^}BKhyivaFt;$jqWZGe>P}+*gDfSkWBF_q5g~~E6HKJJJMJq4L$0%(R$9qBe>|1<*$0HmDHeJc67}d z;3od;&xFCeKAIXRpoT{mVbg*ea=l+!N}6eKXY2*Ki_)M=sLA)+gaiFs?40L z&;wvXCWTlbla%*x!|GBwG@|5_4U`vra~x;(c-L|R@N@0eYjO>qf9s?TCIRy2h;^v6 zz-krrAc`2g@$&jJlWtLj{$a)MkabBL**#7ry&rp>m$hroG&2_r%#0P~P`mWF)^@225$$GhXy<2PkxzP7k|83@N${XL#C#Rwz_XFGMuTQ z`q}o_b%Y~Qk>|z8!-9megtiTYh76o^WD$>aNH=p2KmzK2{4Z^?uh5$YkTf&=KWXHN zTE%_a;g?V@E+I}f72mBhO;#vPIF!r0H#eP^g4t4G8%ls)bj|-9Go8_u7+jjbR zdxySZ91VM#UACv6(d{Nmt(LT7-v7SW>v%0^2a=i!sq+YN00taPc2N-k10)LNAY#+| zd$AAlfeEAyp}YL};A4p>;j*-r-9Y%2inlj~n1;%c9R+$~_^5gMIaXjSRtx#&E>?H; z7vRKW&VW2u%JHBg!wk{0bIDR4pXcNdv|CD*2dgXZl|R&aMQfX`a%cy8akd?k)&2SP zfUaBn=2K3dW5my~aqeSxR%j|g#=bk)&ILX(n87hkKP=HVaIGTET1~ckRl%#Tj$>AV z={rjt*3lb4XG(S6ItoAoa^iB3+hCo%D}i8|0Qcdj41~uP+aof!fJq9X;BnMQ{-*?4 zcku?`nNxSpUO`hD#Ta&2l2r^Lf#M$JlN&SLTKPqr`jJdytnw<@>S~$GPvxn{I}E9D zdC}T=DzSy#e9r3+B+Xz0=QIAogF!fwna^VR=zoh0$oF+wJ@8IVkI`^3S|Ub$8JhQ3 zv~c*zNcNF6o4YVv1G+1&727Tq?IO@#$(RLb2L)3HV}S;UX@DWP?ixhs-GC@@j|AE( zD>rc@1;mLR#{*vB)B%LVgA&Kx+kjG|N40w=tZTw&ty`FMi9;#rtqlr(p<#O(N3n#4 z2QI5mrTE6(W{WLTzSG=vi%GE;wn8px)6?N(ijXSEkYupL(C(`6tK8;RpS((HcpL*#Wk(ia^p*1o0^p7q4-|qopNC)? zG4gP45|(QzMkx}sq-6qVCOa^5?o3S~Gp~zsTE2FcH+B!;OH&A=5_yDepF(JfR_P>bB$2PmRw0I99UEXKGY zsqGkf$w%Ot!&;@YrIZD@7aUon{~fBKvNV7s*@FlIqZ-Z%>lZGbEcy1lTPvzyXj@Dr95+r8Dlc~v%~4Eq9tBAiBH&wp9B$^EqU(f$`<01znLZp^~XRd%1tcfY)J zo3NVmw=r1<0~NqA@qOMp=htuwNBP$E$CjNK>Fmp7#hC|}ygS}l9Q}S6Zsak8^0_qn(d@5?}<4ix|cIWAuny^|Br z;KQfvAnKYS&OnYY9Hg_TUZ=^@*k@w}+oc^c7YLCGnw^O2YWPDf=)~}~C6Mnbf=Uf( z5GSD0{&5vb2S4oMEyIO?&|YgV9u;;!GF6r&rH^5d0+%lttui&L^MAJj7aP*B-#abb z?4k)t6%@ppm8u~;C;chzExg^Qo=Y}_k`|9B;_;z2xT?f$kOt}6I=J?^X#lQYS@@S5 zGE{wIFm$N)S={gh%24tO2lUb{uU>@e3C@Gj<#uqb-N`6Y?_a!a%zpZ{K+fu+`;|&o zL*LFyd312G#fBOB(vsfj$&ArX`_B5u*V&h#S;HzZ-%>w!MtOD4*4r99p9&9|tux(g68T zoSJQ+QKC(R!|QDpi{(M%>Gd*F_GDNO6}=rXz;e+NQht|eR5fpcGo#^q?Gz{HK!3cD z8Z3P*>08oEunOh$|IxDQLQ%Q-o<&P{i|M^QoF4|c@bpThh{I-Y=Ks;*6nI}HJFi;Z zH}x#lQ7leGgpj&(yRjX?FY`clK|{YD^YEr{@u1i6p@lJ#xWFQMgU2bi$a#`ajald# zrsFpsPoB0uF~@L(WNod02VGFE<%mP;G3-kyWs~U=`rN7gTHAZT&^T$^8j0V|{gM*i z^;;L~Ti~U41IWm843Va=NNo-mig#=_9Dcfy>KZG;wpqN4y$?1S=(;t7KJcp~Z~6w2 zSffY!86}DPRJE~nYHFnZJS%!wDaf0_m6QjDL7EQvYQzhH%=VSH8et#vSWa)*@J?6{%g;+ zU)tH_SgBhRr{B~@s9$8ldH+wT^VBG%<31HM?^iNa#!OaLaSsF7hql*;{;guY8uA1FUvN zNcDhA%v{^C+$<6K-hf#oI`vpIb1kf|2!;nwL*&6ZAi)h71yN714Nv4l7lttRLKgV$ zM#Q5BTBrjsbq#Xi1^je|W%vFSwC#Sfh~rg6s|`aC?STm4GsLzTo>s%m)ktkGj6iaL z(Q=S@bPhX!RN5oU>)@vVsfQ%gL*qQ)#|8x70Yg+l2#kSfwZhXxm}2&YXE{8p60=4( o>m_FW@a)3ze=f}m`#ZkR$s&J#rR|#HC<1>wcUW)FF>ycjA9Ugtd;kCd diff --git a/docs/articles/mash_sampling_files/figure-html/unnamed-chunk-6-1.png b/docs/articles/mash_sampling_files/figure-html/unnamed-chunk-6-1.png index 528237562fd84d0753bdf3ff21abb550138d7fc2..22b234c164f442d935b41bfc34dafca73bd3b7d1 100644 GIT binary patch delta 71704 zcmb5VbySpZw>}OCNQj~cNP`GSiIURrK}DpbJCqn2=@^&?L`9?-LSg_(X{jNGROzmv zgptl+=o;ep@Sg8F>%8Y(zdz1mEtp~A-uvEh?|Wa@t{J!lN9BAV?FlJyeV`b}B)BU}y43{hq zz&5}B{P{DT?P5E$X$jbXqB1jEqz>bCG-P*yO&L3$qfJ~B!i9jQr`~duWH(;45RWg7 z>S!k9swSPGG*pRg_Gh>lXCm|-~Y!UH(?4OqE42nNVm*3Dzu^quu%)Y`I)bj2&JAstxmgsmG z0Z|jt5e8@uI0k;PArG<0X_wuPLIH})y~&)eO(%J0n-F>U0PL{ReXDMz9f7j$y0akz zZTMY#t@l%5Az$6ea_Ul=NB_=rpE$8L-5`17P02-SBcxNZS80=krre2jQ8*s1DaSh? zdz2=#*DIoJZobuR66!E`gp_>{;+SvsoJT`gP|;>M2T+MB5x5`IL@+S1k9{hFCz7$M zbkzt8p#eAjseJzng5WpYk4;nR-4SH$K|uxzsDe~lIOtc0|yoxTmKDDclD)F3fr&y$BS=xdxXT))I^yL8)DMD_18L z1Ov?O&F(-IxP}@K9%m5)?sUi~mn$)KrvV{R=7~R~?h&}Z1`3b>i3Jwpj$-r1dW;no z!NtYpz8I@)T*N4I-eYm`7B^2D+(Ke(Z2Z>d0X|)J{d@LrmxGhNj7ypST<8C9830@- zv$$AoxgX)klDz;(nZWja=53P9=%!0V+Thfki9tv?BBhHN4F7w z>@suU%(S$$MxoAARrZ;`fXQfucH*1zXYUHWfB&(II+}&rbj5M0EXlgg%4|GE9KYmw zeV-#n@G6?)^3v4Ez(qgY_p{ak7ZtDOr6eTYRSY7qll$|$l#q~k_)hKo7yg{W!fvUf zSY_5&$I?36mH=_533jP$WS%u7lI^k1O$B!H;3rn?ow-To4lFp?tj1Vjv1?m+mgqB8|FRk4=N>rYqiKo;k8<#=S&1T5k;DU{C%2+M+C z6LlA-8&;{!2PLjukrK1_OpZ+6MP=JXUtjlOo~9=I=dY^8X`Ee+w>smQ%vSBfBDMyj zVK}P{ZT2s_AfP$=fjsk z_x{m8_GmlF0#+BWX~jmfVom1Ve)Pv9`p>&xROt1O^yUxYJeo+Koi7Ud#8qt-Nd3~% zQe-!DEjMW~=-SN~o{RuMxHNzYmhRThpsl0BnUIi>CHVX$nYGE)x^29!k3pMW&+e#G z;sDd?!%bKDDWJo$Mr*c&M zxH(karEZCCvMCUyS7LFon{<_fZ7aXTe^gKdrHXj}F$Q!3s;aQ8QgD;IIVCG_Wg<%oM6 zfW(X^P3ypie>OYwu43Ln2W?5>+kR6&X0P%pP^ahPLkWo9&^$cOQ+ZP)dH`CR_&xOm zm`;`!;hIR}zP0OY5O!S@%JImDihGiVheYf8?lX2HLM$3~I0-Sqb+AZ#e{OVzui zl#HD^B8W6DQP70Dy1E+egtO|1qRL};f6lH%G&K{rHk>Q$JSCbe=9qYwj?VTWq(YH> z@9OJZUqxM=epKkBKO$GBdS&{e_>UG!p$2w-{o1O`_lUs6qZFhO z3x6U`!!|+UIp62Ew`@e?>Ts*RqN3u5#hrrz9YWjSff1%28@Sd_n&*n-tB=(>SbWlG z&u$dn*nQ5f1DZgDOGOSe2Ur~Z&tEN%seMn^@@v5%?7n$dZTeOY44=MG__z0T z285KNqN1J8>CLRwt4>Zg$|ya!h;L{dGSILSq&#^g{13R})C&LhYuDf}c>2a>T^GyhN6tk)+ zE`Cw?L_#5w#x@Bma~D|r*3u_|5fHKM?fk{R5Vvb~6>FF`r9!U6u)X~J#gCT_DH7wq zJ}%ZMDA#nRM&=4b2h4Ko3-31zB^LEi21klTR+NHk(pWQIvE7F?;>eLNmqJWupJC1U+Mp$tJoj z;-aJ(g^qOK2ntfl^zXMHo}FEr_<5hGj{misn*@Wbqt=#wObW-}+pA7E>N%nN(G*2> z0@Q*zEJ}|sY{2>_0-`zkdpdf2s;-iPYm{|sNQ>@V`izYI9Yq#HZSDJlf`WI#tXG(3 z!1Am+_&TednTCf)Wqd*ciI@FicuOYnagY7$fV2pOLt-hZ`U-=4TpW*I2EY7(W(j$) z==tCF3+A}Hx-M*%qWuz!#iEB|vnY7KPsx6C5YdsZLV1OHN|eBd7V3l|gN*{lRWJSh$$sCQVaZ&yCLs&r=5KTR zx48OOn6(fP#lCb6au517jA8M-SR0QJx!q|LoY|b%IZ2uS7Ct9IPkwoI9AC0}d<37- zQrs{BL*&;X+G8%sErF1Zycyh$v*r#{4tTj3t^P0lAcEB&jR&Sr|2>Go#Xjg|qEaWv zfBECS`y7G-*<#Y}fcgL5?4ZCNO)Ty-n|@GK;Aq$FpM3YTf*}dTFZfPBo#Ep9mWd~N zQ4Y_Tds+n(lM+aiBqcdxAXma|esY)m58YzSg6FfcW{Eg&hdrYt5|p{KhV3=cZm8^4 zQ04R>-c<^Ew0+{nR!0GULqmfh6ubxp-cisY({J*~n=8P9c!H0)Y*0=g5 z+?k6vAtD0%6Y=sv3URN!npab^=DU^uLbve^-^AhnRwrvcf$2{X-5!akh@79SMb!i3eZQ|H}2?)5V~c&)pTX z#?S8Gj}1~1jGb+PJomG|X)39wz|Kq*U`8EKg8ocbpC&beC6(WJRpLRt#23Ze!N2vWv;FpOvbCEmfY z8QJ(Y9dV*>cR(cyr=Q#Wr*DoEqg|(B5BSC;s;nE;D4www6McMB^7P=ve;UDE(Q@}# zOS?8eW%DVBSMihRWF$Px-yYB7&GG5quY-Wcok6b{DYh zBsMmlt~B>nbyP<89UM&s26Ku1$*rRl5=mBxGHl*`fpav{lWksoG+m4fJa6gA60focL)tuCi z9SUw}B}852Dp&E^fL1ml&42peXU2bvXZeFjHl#0T@AkbH;0HqC86W6a_lQhtB&_8| z)}^$Z2?!JjRGum761o>V+A;hk^X7y*d}P*D%FN1o5Lrp;^Gd~S?yvA)C*n`B1kHlK zR&6~B{!N1`Pmb{m-W{HwRl7<#Ayn1RvPkreY`l|`o0;98vEzvC2vdA&;wk@uew8 zlfyAMGXrlo3X#suu5^6I@J^iQ*HwOaTt8{b_aAxK+T3yhUEd3lQizIFcfQ8l>OxHY zgMCB zf#W7$IC4b?&7j5>;rSSkQ)2#jI)_)uFzdepD6cp#( zpM1qJ_i3bah2Ng-Mn|fax_V@FU{9jptQqH7YpBxIBYMuI0chwDl(_>u^o@MeyQDj)xP5(#ifm zbEsqASB<>|>zBKz0qs^h3gJ)L89H79q;#)^zzoW5_iVgCpM$1(I;Wg!qp_R)g(D{Gu>03QG9_5 z_v^~3(6%c|@(hm5aCwzos2kT5$0Ac|a*A>R{V&QmJ}xV7_-z!MK<>ifOH}`XV|pUG zq%7a-;xj|2%H*-^LO=G2#gu4iVC!PhL7>hQvYG!QZFa8ps3%?3tKG2d^c zbU{{y^HpZV_-jnnl+l_+CF&aE#^atqJ3q@tiw-pHU#dCzF0c(Fv5{3j_5JG%KWr^-37ub~EgC}-6ti-tJ0xhykC zi=RWB3>+JsEDjZn?~^Pa+3vp=neQKiIE_r7Mg-1`9;xq1qwx`ftOWxTM%rRr1${lEYUMGd>f6|12SCoM50;$h#Y#Iqu3iEo1bNGCTM3gQ=T2C&Ysjb z!PY~|&1~Q+JvrIU%q9@9BS1|CS2zVGKdtc<@DnlIU0qv?z3S`j!ZXgEc^ZvuPF5~` zwTg;0ajD}9;iAG7*Q+U1WG{2YrZ**4%g!K%MC%-Wr?gb=r}RXQcI7o=1WHmZ z+{X0A5AwGXc&z^fvq>nnWt-vQX7;l`h66qu$8FA)F1+f%GSCxX9bCoP>|A1(50-s; zW2M%`KVR5c>RvOXxp@v$-HlG0fBL>H;P&9HrRl_rZvWZI@zI$f0Ys!pab_f`OFPS{ z5w8H74NhaN+G~%`-WWd+-(P;~5sTydjC2CKy;IuZ<5QhOj%!Vc>M5E8o(j3!%!4-R z<*6BUTo(+eCB41ek)q|E{O0b7k8DYWM#{NfS~Dzy!#|e0s`>?i;(JV>fN033NX@Gr z(o0D31x&1Ir#J#Sknh``mz$2~5o@@J7?lAaXClZqm44!Ku1x{{=G+aqJ$rd86Y)}E zzZ|?zYeWuF=4o3CdIgBZPe|(i`m~HJ9?vV_sd1JxbY1bACm+r++@yybYRN)}$bQX$BkiXc8-d#?0edX+2 zvI%x|g*}$lPWJQV1w+|jzRkEBSm9E$W((gr5~D`{YO%1vG)WI@T;ow9XVlYwG^DXr zA6A5nDxSg4H!Mu>WsQf%3=%GBXPC?LjI)C>AOQd__Af=fVV7}FacQLAVNbrgT z#>U1HlTr}C#fh-ru64)IR$CeDpeSo*=a$U2f+X1bdCp@#d@2JzNHlu$=9`$-2uCRM zeR?y`*t_&RS*TTq277=_y|&I z3N0_lrE^n6aj*gwtzK1Wwc)-d4hX8@)e(ZjO>;)Pj5QhdRO@VH-MVt;=n#Y?|KQjZ z`vz;1wp1#Tr!?EXuIeDe=M`k{CPU?4rLgFppXvR1&K0m7s$s3>a?Eg*-YV)ApyVxi zBCRd@*0i~!Ld2BFtB4iO(gTdU=2jmpx3^*9RXPGiu(wvv>0;*l4_RD+d|%SgmGGd0 zh`@cmHHWQB29v$_a#l4xtLL9U@7*cGFdQ>D8h1<0wbT2 zrY;>d9m&>h)ox|*$TsTGpmpF@beJX?|1x5Qp!I}V~H5a18LRcL3q{e{0Y2J9q_`_-R; zNXicw4E)n`UH(|{t8qbyiisH! zE?Y&*0^+69b+GK5qZikvPo)A+J)4V#>(w`QB3%QNX7|#MtVJ<%=|(=vaQoObm*@H1 zw@0vz9aS1}rD(aVQt`D3^Ktn5Cb)l6eM-&m>+;!cdC=j9IvUw0_UqDZRd>hO zA)}t*(RUM$oF!C~yE_ti`M=$nILyGbk;qea1BJP@{uNbr;e!alwu3D1hJ8rEA1r%O zsvq)^u(9?WpTji3`WbJ@$}13C5muRNGs2&D;9;TqKD}RmvSM*Xc&WAwwQI@MK*SPh zl6#nYU6!6Y_RffQ?a1ba!y3~is;FASTL52TpX(jFa%^dk-L`d9oS0{}ukuUB5?_G< zIzTC273FrySeew+W|V_oC6F+3OSlBcx1s96N6a++xnw9(3b%_Uz=?Qoc+~VPS~VwD zpK8DULBv93(o3%fY~Q`}MDP*hiyVLm4hU>_!H)+D&;1zikGkbOhEG4>FKOMsV+UKY zfT1SP(Jf!s5tp#%_ur_sR+*qK$IkC{oGhbF8Zgm7m7bIpU)7;R>6c z#W0@qAMKSZeg~%ONL{%5Gwo@^QcY0#t^dMQ1}8kg1&T~d5r?EvvgT*fj#PiL@ynzZZHYCNUw zeSEP|!$cRm<>oKas_=^bhRvA!PnIsMVvW>=_mpatOcQxz!8SHZ1 zPwxXG>cVcSe)zL`z35aN*=bWXTloq;$33Y@o)0yZqBedwD^$7ZQE1D`VMQM}dUi7* z8I!9&;=IQ*ym&mAJ+mprSd&cm%4$=`m#SPLWR0)*SHOfy=Mv$=ue8Q;4CR0@x^?xy`f0|uD1IM#weI{adXw>wrN|cs zgDnAE++!Z{f2G<{acYyZ9)<(5U_~h2a1mehR!R;#aZP>*A0ajqHZFfkbtHnWddKQo zpO_z))s`UafEl?iliP?6;Bc)Mn&y3*p1OdUJ|a;}@d)F91ZF zr|AtHW~#aa!jOFtt-T^>Z>qhbn|=%a?I}l{@e5_*m!CAN)61vTTPj@+<|nw;@dZ|` z$p!LOK4#WU#P^OFiVob6^UO~c-W7aBANeR2-;W}3R%yh3M6s9{swuM%J@~n{xa1Sn zZYYva)q^6H7qc%@hx~fryq!~i+Zpi8-T4IGd${$AVYcJx*^JO3bo!Ku_5xwbz6a_i zZ+>Ly=F=cdPOl2>Z$+{k_Qy^{nN!ZE;0>f4Mkp4PtBz_uTM|MMrA-y%!JtT#nDteH zUE(n&%@I>6I&j#+H;lR6uth%(c7)P@H$3%YPk!o)F568I(O8hb1P=uhLXDLt{WerE zhfLjTB~#8b8jQzRcv3NYo(t*#ZL$Ea_euxCIGyfts)_9`6BP0KP6#>u#dg6C z!tLp_Rbz>nj6lQa)+J0OlzA`*BV%8rjXibYp~`XSF$?_a@NO4#dEsM;d9WsLODykv z(%ZBSt?D5bMHE>67Ue%Km>z9Wekofs;NlN0;6FAJh9yQA?YlvJHpiw2{BcBiHfN>N+Geb>0poyw>xUjWgp z&QyeU*X0kMoil^UNWw@^(d(rmcBepTCYGAanEkK!q@scR$yv2@9jT(2{bL-z#mSc| zb-iTzK>%Edz9eU+0GrdQhDDaFd1T;CO}BMwq8Y_riDu>tz7mt&A5_XJ`Gh~a>0~n@LsU2L`v$vn!vsI(6l? z{xPJ4EA+G{l&;f7r|^;F>XkY#=^S~LI6A&YaLK_{c-{;@d4bI_o8n>i)-sLJGg-pA zuivtsk4-XG{O|;yaa*D(d+%3(Kr@2S42l?kN3mL4*8gbBpVo$Ex%SbEYhqI@v&qKd zb(od@z_mw?KVZ|%r-1RjHK_NWR{Hivh#25XFp85+aglv0R6}Aco-E|dtiodx4sF-i z0e!e6C~KiALSV~xO|$gq?bxPOPC&O@X#;_$9r$?>E!HgR#=BEvXkD^KNs# zFeI#fw&J+04e&@*z(qbgxK>(t{mgj=XSRPl=0aO2_z&XRZAX4(JV!3(d%(I*80y6C z$Lnh)lNwUxaI^`WG@3{4Ys?RdDF=PjiQ{nT=G2(1*09TESe*Ar4V+~YIqpXlst_&l z*uId3w=kk8Am{tfYQaa;Yg0_kZ%K?HSTkCaZ$ZLSafwcTeT4MDXcjbSc$f zai>tL{)$8C5=l%Id$+Oxw}{}SUUj~6~riNl+Spsk9Dw}o=h6wxv%Cqpx;b=xJC(kGG)m}Z&QNhP5 zJL733$Bo=F+wLd8RJ{4f0NF7Td9;e0_$I~M|JkVHD+lvM$GR=!bn8EMB)~qyM&Q6{ zqs4i8Ig8iUoF{1r*d!D*NjmEl3+SA9zVf@inkThJ&WZ_Re1F=!VchH0m|A-IqZsP4 z<3?Nnrjm9`=%Lr!95wJA<4u24S~|h=Wa|<~bRIZzROb;_>K1Bh*<{@va4U!!Gv82L zU%4|hQNDGQ7ud?^I!KoaI6FJ*n<}o%-+Q?Bphv1Xto`jBuE2B zo)e62)t9=TtG|kWiN4vpcCj{+&m$#(G&2cvh|%pJ9BauI=O1q=l)%;(-< zG{{!;&7$T9!s=an)DQ_`l(Wl4>bA(YIR0e!bE5CFZvT{c)Ss?@EeJ66dc!43B+Jdd zc*DV=jgw^SZ++mCq4#h1m|e>aN>t3=@`o4>0v1R?di9OfQ3Le7zEgH*|6LF-WDVQ%Q=e91U?cqeYlPLGV; zrYCm9G@-qkN>`n_tNB@D5~+lYD$!6BO9!ImQyTqtU^^Wh9zBhZh*XzZfqFVHiCN;D zCJltW+k>ox{8cCRkC^&uGcI>r5_jz>i^=2Wy?=1T6Ey(?h8agrqbsf-K@0j6A7}Tt z?!&HKn!T>sePTbeYrH8#)tjWAqD&yl9fzV+r;@&+NqXqLW}-P5f7j{vV`fA*^s7hI z$ePTHr0UT1JoVJLEuu168L!zmkTQo()pMFjuQi0$I%HNge=;kF8!A{*6HVfnWIuQ3 zK?~3OfcNS0Hm`6Bmj#PUH&``N(bHIm%sp&Nj)p=`@ayIrIDIZWZ5-YkvDt-|MASc! z>JT5fUVyA;c{pkA|7Y5@84Il?O*|*+YPw$eB#p>T@CoO|(Vu)9+n@Sra(5bM)@LkU z8Pf3kMPUo00inbF{c1WeBGca*QH3e^Kuk;lbx!WVwWJ0@u8ojJOfav@$kuwlOW)LK z<=gdH-sQL4ej%xY5qdq^uR)WI8{Z*$xqLhUOp4BQtA$7%u;If3WcsnCw5W zPEFb0Ou5fsBA0vFOexyZ9C&i=i7mp;W&eULtfVrgp(QV+>03pdJbxo6)|}d8Hc-8) zBX>`khZZfw3|Be)nm(0U(L6Ba?Fy(u-!BF4B6;BGZ+K5(YuDkh1kVW>>4S0YQMpeB z#H9R>1&o8FP{q^De+ybBn>YQVW)1=}_YjH$p4Sjz-Gd65h`54iSauvX$97}9tj5-6 zJT$qoALixee{j?X$J9CXum!V2-13fSH$TySL}o zBtK#{Pg-B>t6bnQ?c7pHb2vmRp(i5xyAZ|_CBkNr8Sw6z zn`;uT-pj}uf8OY`EhqXyz$r5p;HkCy(os_Abi=! z*5$-_nMQTup_iG-z~SA#mlXI=K_kBE0a*y*$C38VAe&7flPH<7S=JdkXV z<)J+(=QjU)Vh-1^r{N?y^iLLDLRi$mZ*Q>?Zd}T!)5Y2E66sB=;*dU|J(39`v@6w1 ze1G0KEPge$r-_5(^s_deGyC1B)0%^5bH44_YyM<_?gc`Ozc`~wCNNRPY*!>9T2sZT z+l>%+-KD6WoM=btQ<}(0`i2a9*ixdPqv^~Tg>r<%jPR|O^J+&NAdFz!J80E=eVbQ* z{Ooi*U6^oaw)|_A^q+d=W?r2re0_YxY+K-c$8R*B9s5%Mp0AJ33)A2xL2yfT8ZGuj z!r2DUmGe?`nsNQ++v~|-U*GAI(k1EnK5DlBpW-H``I*I7%H`7}vbh*oG{x^` zkOj5J|E&9qPgfmfoIuUcP^%kc9xf@Hb9Fdjo^*H!!Nb zzjCOgS%~WO$?11Bk_b#_XDKmyOxpH7biHqPCc=gt)p3j&y*|G0N2 z*zS#rsXDa&)cEyNAAgfso3{dO^<$b(!(+PFuD4&5xMS~G=4R$-x3e6(6{H8oM~Dl) z^pRYX;uRKYn^x%%E%KWiKjbUpp!eTHr{Sa?;`Cq_0}ESr2xX=dWf!{SU}0Mh}3wD-`YZB|*f_2?Om&HD9T z5oGsXgqgn&VDv-i0g)Ck;G037QJ#)I_V@FC?D#mz_|m}vpszwVyfGfUl+?GL`%EtE z-~Q``@#9+X4YC{I1mC3ujjKB69b8-zuNbi8Yo3>2tJR7M*eX&~W7(_$3ELpKlx>5V zWqS7(aIExm&#U8L+=6Wy0Gp@36Gf^sy`<(pa~5xk)}D08Up#q(Ha1wvQ6ayzKRM?# zTUr9qQMt|*@NRP(7qB8-9`w#B+e-0T&Mo<~AJxxM-B-f{QH=sO`GY^7m)K+UHj3lj zx>lIHTrC0NE#?d=5a867Ie0gMovEsS*MGycepZ>ETcs=h7gwoS!Co!G&z6rZgRg=& zl;li44S(I#^FBS~%MEvNXJ@Uc2A7W_s{91&Q!&f0!0Y7;oKHQN=0|cTB*{NJ^6-DV z{rT40LqnPtKcP%zo`k)eR@l?=Bd4DBCLw7tC>Cf>FUT>2&lK>u>^?Yxq20~cFU|`& zM}Id<97r`xXFp%2>{aAuQ5{O|?lY`D4PbT}mU)4Nrtqht`-=S9>!1CsyE8d?zpU(# zJiS~Fx6lfE;&5*EWA@xfv8}W*1Hi8@m>2q8XzcZj9Qhz&bGpr^Zu}1mgSF1>y{e|h zD}n0J&yAX)pV>cJwA>6@Pk6e*=w_aLlk+>FwtA&)pZgZ$JEP{+vsGnzhw6)f`7oZv ztMQ1p3`2IJ;#zxz*Y`2nel7nz1u4@cSNC2M%xR*X>T7kz7}dLX!CrmyLqWktk%M$^ zV+yY#HYIJcypJ&eEBl513T&PJ{3sw99D6kcoS*_6>T{QVLNMCd05_LbNXjR zj^44YP<;;068c+0N&J&ByY}K<>(3PvcUMO#kr>1Y;9mmwlqt{J!K}XMC zdU<+gqa4lk)D)c?+_~E67-(~26vT<|Q=P9sOqENW)vfS1w1X$w+-hGf0liC*M|{+H1JU^s z-Y#)C<70ubYGA6K`>FxUnZeNHVlEDA(4|LB>0C2o%OU~}Jk*`-_LSJ14}h}^y!jch`f6~fpvG@TF1h| zHS$^oxL6N}846n-RXjVWVEa&inV0Gk2mVJ}l)3>Vt31oEMPl501kt{%^;*vTQX2wS z0E!rKu!jFWhyfWMpb2Be)(lI)*6vY>%>^JLL*ft7^Ra+xLg3I4qm8w9uHkn(fbB?W z-Y}}6EQ+v1-Be8Sta5e@bKiaK&6~a_HmL;xXS=p0F64LhG`;n{XTvJNUdjHi_Y$-q zm0>1+*uQc`N!c$>q}eCsmb751MZzbE#Lh!nK$xbj;jcxhRfQ0q0MQw8*M$&L1DN1n zRTb4r_+cvhIc_&}aHD1`T4#h+c!Xbxuig8?$UL0b$s*mMv7bIYroA!q;5h@?`!<{I zPsJ3zm`bUe#S$mZOq(M5A%!I-e}Xf1yP(kUD3fuBQ<0n%#`M1Qg=2cT<9SAT(|}HT zxm)_1>Ay$CoBQV|%j}KeFS3AYE#S>vH{3AG5+RT=JPm!)H6rQsm2x7osLG_?Vz^vJ zjK$}Vo63OeZZvSF)GxEAR=1iJepYdJrtHaGmMP6>PL3F=;Ny$D$%kH_y9f0%rD|~q zpILhnx$5d~+|CRR5^Ly#orMmQnQnw2yBA&)IMv~=EfACeC%hIxfVkELK*H=1A|;L< zTSa9}##J^}F>9dt6LmCY^+3mK?n2?3_`#u$3cJaa%_lFTK?as1srUJzO{N>N!E20t z2b|S=OA_J5?c-cpsuD{4E^22@cEB71#Wj|&S4JIuGr!^k6te|(N;dBMvW)AR=`iR{ zJ_;Q_yZ!l;m>y*UIC2?FwR^?NVz=~Vr$}VSM#pOV9cCs+?2T8F_0vi8Y%8kVeZp_c zwj+8))zj;(`Ph6Ui$6YBFnJ|qqFj~CP~`POZzbkNeIMe!wy;9cjia3%M|E7i{AzrZ z`c@Cp`SP@c945!q^V-BfZ}Y7!h2Scd;|7v+>WInQ2o z5xH8#RIk8j?gC`02<@nhMU@`J4Rtj(oAUVSr%2SIn(29Vq%)N=i=nITZvETl;x-n^ z?~ZfR|Fnte)Krh`-x$=-%Xvq+iqG-tDJl=Q!`!}%@Y47@49)awzggi`K2|+`J4T@R z9w&*-Sh@cLz2Ejzt_8S*{I54+AMHHO9K9B*#^L$KV(j`zh7ui@2$k z=~fvQkv;G%$dozXAK3*Fe9rBnvckf+4pwL{qlaz?GDUQ|K;FM(D(e6wvc&DWXSlQh zDx(_L5CBMzDbY_k+$Qv&S|2NdKSdo;2mV;QwiO}LPYtz|m{yJcD(;+O>d_~r-!rJ& zL8DTChxs03@aMHHx*gJcOtVii#!yz{gJTU`O(y;yQMBfo2C1ItX6)&XcPg$pP)Vw$ zEV1;oCI{sQ_Z(|t| zdq_5DG=@--C!_V3d+Adx{&J=W@HzKIbx!)(_2e9~Y#rHFu%&yFIz!eSN+bjm(mh-% z>Umu)4wc7|pBXgEzJqOCCjL2^!7jsB?R6oDNLLvw<8|F@@(6XY(r|tZ7YuNEBwIRn zkZ&a~n5tjj?^e1B6_@)y1&pi7T9_EMDRfpcA%DGS zJ72LA(Q^B)s+~qS8>2*#Ae57GC2mIftYxm8)kdFNf5x)`!@eE{&#?pR5S1lv1>Dtp z1NcE^i&a}l|F7N$fR&7u89RsB5X5QC*XKv&n?9XVNHL6IF|hqKa;%_35>wX4WUldj z9`B=g$3i$#@^3bp*h8wzG%G_`vu7o1owwY|YxjAaNoJ!6zR_k%N@^zIrIq>uGvo2- z@u&7DRjGJzXw0rJJFC&|z^JHE5wRCA(YAVHf68P(N_y)DAYoJ{h5RmiBmJhxQ?GU1 zA^n^p3yoyqHYwmw?S%YJJhNO7-$Q;6LB2Bq;ev;L}i zsilij_5fT~rvb*FEe}(P#t9Yti)`g(GVbcSKljw_EhhADaQ}EiHGktxiOgFXva>Gh zO&g`CZgVJ>*0DO}JL`I?n$&j7wG!btN%7BC+I2OG@fkvXbWsz}rHpIZ9_8m}D@uu9 z=uq??x{Mq&7QvjIG9Il8Y|ZK?Pi0kBw$ID}+9qLAAxR{(G`S-~aaE}57gG|=4@qcs zdf5uG-0TjY{2vGFLim{L@;%0k!D`5jp_1b+kqwP)YRg~Zk-o+7`}&9WUOE2F@UorO_?!AfSJ>`# z?YfR{%T2DapTxElvXEK(u*~tIBH*BGeiLmj=M@~Up0)vc*-hZDxd3Z36%zD ziZXd+<>b}O%zT!RKWG+q$Qa4>9Tf%m@6}tK;Ccri3I-Z|z4eZBC1bdzU)zr4RmT7w zBDsEK=%8_#B(S}S|M@b*v>izC@#D5p=z0u>gz-D`L%l6nr$VS9Zj4Tu z66Q>0R=t8Jm2s_Q{o$;vgBEH8M>aTZ*0%u=U!ZoTDioADP#&2mw<^i}E}Le__t8Gu z(3v>bBhhK9ikA*P_{p>$2VB9KWAFGAWF+Z=Tgh(p5PYY~6NzGT?`*jR!~5cj(#tUK zj{|AuWyM#$SYmuz-;vld{~SP<+`gpRZEqW6;(c} zJ>3*3J^iw_Y=qzCL`U;4Aoy))RysTVc#g4eO~^?2yrq8=OJ>9)on3!Blv%O(cSu+H zM9p4i>#M3kWf9{*DfbZ+nb{k_k+29wzMUc*mN~Z^S;u9RXQ5JRQZ3N~RvWVy*3z7+fI|#@AtijAx_wyThm3Q22>9ETXMCKV%v$ez=4!ZP2qx;fL76q#tgSH zgN$-7ezt1Z+3rKcW97n?i)hRvZS)IH}Vq=9M|bLaP27qYAEkEEc30~57J zjuZg>gx8{&2ob#9;K0^e?kELV6^7eqnO5C^7xMSqVTs-KhNj0}@qG&QimZooRP%D4!r9IpU^)0 z7Z9$a2!yC+JKHG(IO89~F_VXDR#6628q7Kt9(BlAbqGi)Uy*!j_{XSb^$!bx1`8d* z@C7*i;2YETXHRZ16bTR6;T(6xn2uj!n_&Ax$|ifB`}AkbPxEIvt00Ak0w9UZs^o

    jQy-)UFrvsJQx@wp{T^ylkzi1^6x)9P9I=rH{{QpG1u`db&~$>SD^)5_|l> zp%#S+;`_Fy?dd=CQcmy36bJPSlvNHu?#~45^<`l`QRx**IU-XR+c-Vtob4uJo z8jrw#tHtSRJDbmdQf0ZwvgAcXy%5=gpq_nE-oMBh^5W7tu?FT-J_$#EQGs@TPu`)) zzhQ%@(DU|zpjLZvQ|bng4`ZdZ$#$`KbXx7D3+|(qJ!;}hhevj5=dOmyIw&c`= zz}h~Icymf0OFHHye*FIt_SJDwb!*##w3Kv*bW2NvAf3|E-JsGr42TFyDBTE1!_Y_! zq0$Y~r666>4c{8ibIy6sdEVdm{>QNQo>{f;b;osGD`ZKTjRsyKr#38^q+NiizL6Z8 zF(4+_HN#b$Db8s${6y_->d)lHf!48|*?zfjE(jnK@KFs-Mf(BdWL;)u2SWi)pWXKJ z0NBjroI~gNL>nB9S>=e_2{|wCImWwE)qWD#t&XIY99cP+fk3~z_>%pg=zWRrk#rHV6B5z{<(41c*2$;hc z+J0@wqOpfZ8)N)WBXN<{{?MS~KB3|fU?d?}GxI!-X*q4pzK3!UML?V)f_&8uHvAF3 z7PwaYH%=GeiPs*Nk@vElJ5!|2#7QMH+$UQ_aH!5HgZutQK!mnq%H4Yq+D8{v%`nZw z7xnBV7IFix6Jtf2xEZoXAjlYv+-3@tSuBmo+iyk zwLckPyB~*yPJ&s*2ncE_!$hHfh>6prfWQ~%uKW+OA18t!Mj+7fhn4#;dg&i_qY7wi zHHsQ9f}AdukJ9Mx$Cp5SP8WL${(pEw;tu^Aj|)Gq?RD9)jO|{Uw~1BSo|h0H-8M^G zTrt-nNVkTJ91ksy-eiQ!_m|wDqyS7UCl78F{((Fl3_;i$-}znm#m`Es3 z2{6v;`93v1fB1*ONf)aH7<}X_jQoEQahbr!1`pxFdx**+ zVG3FnGJnw@HOVF-wcQX*EoM@bUX{20CHEtg^XR5DRTyU|&p2sAvYogMw9!#U?}Jq- z>X`m#9b$b1L{f^#SvjVEaA~PQgEtCa;QoV=`WQ5l)s-;eAB)rj!AQ;R6hSajP4Th7 zogY7|=@9%^tC3J%>SZ^vZ0T-oZw~&&fyM}aeLScIG+zFTaM_IsJ)Wfi#M;-ZC#6YB zPz1~RVJUIJ0%Eb#b7QEIYme(kDYm%`Jh)#{dVhr}(5C9u9pq~;3_D5T8Hl@>Z;{{I zyJ_-0ScM>=OEZj8nsaetJQ|+4WvV8HY9iQsf7q~pUL|7Cllv@_9bDDj7W+N@{@ z&np7_ZbERgZybIbi3SDqB)YtmbZ&Y)?yu=0dR@Epo$_xZgx`${9T{JNkH3+V%Rvf% zmIHU0E@oGL`E#`BF7iiw#F`>P(?M{33CX&F{DsV2HJLS9`;{IlDEvVZ zA`}H0-9_A|n_foB`Oj7*5D|{PXs-y7mI5LlJLusih(SlgcLn44Gj6|~0__3}_zYg$ z{V(LGbHGM?jpO-3iOL{-r-n98;|n?eXF#BReR-qh|F`u-W(gfE&n%tH&dnA38zEio z9O}OrL_>XjmcAsOq~op0zN_;?8&Df0h{7$OH}>|fFEy-|*MpgInV6XH z*$q;qh`8qykq|`pt`9I&Bcb|q;V6DXLaaTAB~;0Z{nASwjZ|Q}k9Fkfw;3&6!w>7h z@%4~B|HU{Ef2(h9ZCS19>Ff9D>YFeDu^L}zBldekQn$^<9wSfmG{=1x5Hvi%)`|l( ztS=2IHYmuz9#F9Ir2Nffwf(;TM$tO9j2XQOjN{$x5o^W0E$J@28ne)JMDCr*@<8n8 zE5bjVE}Oo@9zqRqdVS*;Va~I4YUQmr7tpE~JClc?(DmHG%`Iu(YgSEN;#h}`o&5tK zh#BaW8?z};d&Vc+jwsOXh{h13gb;v+#K-M_C5DJ_W}7Sso)!JcxW)u-TUE|1fTHco zL6Nm?9N{@S;6MrglY(53{xXu&`18X$x9l*W)iHz>MB}&qu?kQw*m!vA)z2sDTbS%J zee=c7euU~R0JW%E_}`m&TpKzH1Q~z0d@mItHl`17vl*Tf zhI)N!CV=w>Ddkc61e9jJHNu^Kf5#2l?3uwj0{NJlgXs#3(9&nM>_y@t7|?cLHU1%n z_O}g&o)B;&E1%gp)*`49i!)C6I%C(Ti#&j_##fY|&+Ikzs|ONi)pzox#$hZENX9)I z$4mlBLH~UF{+Ab=9*Pk2sj1D+m$Q1YX^@fn%*e>5Ba(RHK>BvteMrJ)(!{$K*B?m3 zEaOi0y!VhlTUxRl;XYtQsiFs)BxP?r^go*kLD1TN#kQpiI$73&TCKD5bc3d$AsZ7D zyn4W?O<>yN=6ua4JZMA3>_b7OAG_B~A5SZ%sAOS2LgI0GAqc%K zKGy7cLPd;MPqItkx)Zm{c%x(U(MJk5Afi+A z)PG&qe8y~2#`V=0+J~1Wt?z57tXuBps%m6P>2fTUWuHOr_ssJx1uc^&RUbOnEKZSoK=PmaK1Q(H6E!7(J>0;2O z%g8~?<^0|={(ctDcnCFywLvgRHaQxT7+9J8m_{)`F%V2rb z*Q2LM0HFdcD9xBl><>G6r5;Xm=q_$%B@$s=;2mU>q82yzJh(DDj@>q|ghC7rHL0TK zhs_|M%ceFsaF<(OSGVJtkL%oLEssv`vMjH_MWDq5G`{Gv1~}sg8yiuVHo^? zwZbKT5OmW3J2&@Y(X!!Jah5f=?90b==FsvJ5;j(Dyw5~c3UvI)5)vi^_ix?}y<={$ zmTkkDvCF0Fm8J_tPWbeMZ3Xxoj?}afa*R)7XByaut#P>v$?P)ttQHxp2dY@KB_zB8 z930#-tRR#GoKrHo{!2M1;T=N4!oq#1q{eio^WIs2!6Wb5dhIn*DVIA^D1NzJeXzm< znRZ0F_(rJF4J?E32uVI8N(jjho`USK9@_t9welkQzxMKfkS8}jHj$covBkV6ehO?- zbx0P4`FQXeFi}=aO`b3FtfEEaR|nEq07E^_2?p15ZQk#k3IfYY8F0{LdOq+Vls_Q3 z$X<79a&vPRch`&1r1x5HgSH}p8lbWM?=f34si5WnK)$He1Cm+X{`@cu6ohbtrYYaw zd~x7OXn;zy<3OG3{HBo%HH*&Oyi`!PTk(JAof~~z8-3&p@wod?LdZ)b>MVDl4>+MKesi|mqF)Uxk(0lYQp7f>7`1fy2&-p^ zj?W+rhL3m0#7?#yAgQifDG=YKQAz;|SQ4&-%=Y8wH7A<(ISklaBOK}4CJ z4a7Teo^xxxF6D}mI7zOlulJ`qIVndNsTz`+1pBHuj&s{@@<8PTxD<>;;RrP;NW%jl zfWW_A9I(ot^}ec#&#=lmLXiGF|5s2I(WMDGvv7(VoZD3cQeFkdejA@&4hT+u_A+_V zL7YQKR|crJpxD-LKE%+wT?SS^%lhV1ACL`S)YP8vFOgFMF8u+7R5|_)t8_ zN8Gatw1GOVjT_7j%h6)Smw>~|<79hkFTjo`aTL3rr+}!!1pp*ZO(*I#D6`v9!`jF2 z0)E^rcQ=%wV}*Xn=3_SdcVi&`{L}mq#yd*g#N$3r9aj>*IcC?aWsdNf2SFc1!;31=wqJO>Aj3743IQ}p@(^O^3jQRjy53GHQxXKn{$jiss<&}b4$M#Hj zQa#J_9% zY(D=p7HY%CkYgn1QAZK@ZTVU{=E41cS%81t7uP6=gNfY$=77ExRR^eskfQzDqrd^* z))otNqo@yzq*4DNg~AG2{>O#@_Z;%C2md=N0#PJq1^lDn*0`JfZukEkV2tL3luqVW z(0%y7UGm7P2PG8u3FF?OQBfqnllQ-yweksWv-;&2L`^SEz~0jk6kbRR9B)z#E-+Uv zqUwJFBTQpNEWhRb=HidO%M6&fvHWGZzd}Gy(5rtv2l?l7fA5uCqkk1_czQhJ`Z3#m zXe=Vq+lb1Y`HxXYOhPOrGG)D>4g5hA7xiCiHi*D^L1Z)Pzvc*-Df)*Fs!t}iv<&s) z|8!rx0_*=1lGt0@arHYf))8n9cqQr@NT<~Wzv>af0}g)A*RNRMvJ4l)fn26vW zvmSKx@MPkp1;T$vk-iguipXr-!}wJO8q^H0H6}yEC`q*1NThPukFy)f-0|%SydC@S z<5Nu4MDBrA2V|U%PJ=1_=fe8a2kAGcuOE<2kBn1W@)k!+d!b@g8&Q2i7)!Ntrib~( zHG3fs0^dS8cluk4n;*q)_Ygq?5EeDYJHhF1l6fFx0+Lcu!JP+qb;O@nxA5CitNw+1N0)lJme!YVSD1O@8t{_k9duI~#ldrx&@|ID7u z|9AF2EG?K@GZMU46qstWW*nVOEXzxdRp*}b-vsJW&w^hsOKPbg)+83bBEoN_uul&` z^dIXdA_MS8mg-NC{QuFUGW(|{)!!Cna4;8;pt9nz-Knjv7IU)E!bi3MvrBOzrg;4~ zdt1?FPnw z?ic@hhsdy?Z{%{?<)sZJFup1F8}$ z;LbvXpq_jb1rs5&YV22BF;&UqA`OJr$l)*c280IJKgaU59vwl1Vz!j5zi)iBG%F&~ zKB&`jKsnIQE-Odn-@&jf-gKHJtS6mT z>`@HKO%_}j8kfs{daA)wZOeSw5?jHHkBmg1C!?vOgOAS>l0$c+_5E#j_WItE*Zmjz zI&%q6q-98W-Z)2MRE0CqVX9yteZa3`spgS5lv?C3$}Ha8%*=!+omw}p`ATJFmWz50 z7HMnfPkIDyN$fQRoWDGb_HO=4<-c9I3ZRGl00Q)Ry4kndu*s{;_XiZc9|pO0LO6@= zKvXsL;UV9bwvWpwQhx!ua!N+e4-{qBDz&cK0cKRt02GeJ>FKUSr?>Fj{~r zL7>K?pAeM8?dqXMNoMu*^=9c);QKJ#KL^?a&vWeU>l|isNxf<5sE(p3gr* zYR=Bfz2SWk83^T5su{?d)U9u>%Rw{zVzXr6MHjW5N9pdwXwOqe6c|IYQcQnKIvVz;bN5qSr&HoVPOanYNbH z0HcDsu;3W3K&6jG5c8(7lgVMcgV}A4Qmu0ZEQ&d^v~F$l(dB%ywWA%CErDP|D=mcD zNI!K#IJ27))|pkB_Q5-l9XAkp-h^$e>fruZ<+*O|;l~nqJL6VCGM0afp2otE&bw5t zicm#CHvu)+f?X@ATdK~X-`M!Hc`hT)@K=NPHM9=OkT~fv}i8a^dj$0IsBq9C8*;1 zwoOnllF|l0Sb77WwV!8pRK!#s3rwc@6uv=?y_RTx!8bJ~iC)`CJWva!olwn-3mLdIqV)Qwxg;+#Eh$rGcZESvPg?OW$da zsF@DNu3a0q!RqjHZXE!Pt}uRv@@&zkSD@Laq_S7t?+3xyZ0V!vx)Z}wiiXo7W{ZyF z9bU{592K{^$D@=`*TxyKbV%@crEeIakdwFQ+`FlB7pUk>!U+z~GkS~nt*mP=SElCg zOS==WmqnSlMA;tlLmpD6C8DD(RBjL(okThW?a*sWt@*I9qq1Y0e|j-(lE0Avp#X)5 zXIoSgB`!M<)x#~k6fJitd(^2OsAyoB0Q{w!^woBWS7 zOyoA%%ohc~_e@(5tMVe!Yk%a~>TVB`TAc-Iud+jN`Dibvy817Uw+4WuFUuh?p(!98 zH^msIl;p7vJ_%-Bd@oS7v9XzGsgY5N7Jv*q3v50bRbhUwm|Wky!X-a8HuPxuUAjxH z6p!QzSQ^d?Cv}-=D*)FxiIBox?I1N%t=GRbol{ZUCn<5>8H)(wfRG=)`iTd5j!J44P?_@3BIF#X47&ou38ATn91DW9 zXxDbln7VsUxStWJmjlqh<31k&fVcr9bgJ0uLmzQhBHE(_eN$d=#XU<~S66pX$a$Nd z4lklE;kHT`r%WW)BW>__$iQuYM4Q)Zi0Zh??L)oxW$&d+SowCscH;NMm5IBaZ;O&S z@meC^HHK}nmSfCbLgabnR#ftTMaFvgN+AV3>NTgV!J?))nIdW6lRSQYji6X z4g2r=`lmvNs4FmLY*2MG!cDl~^w|%*9buJfeS4MsPd2w@LuC1j6@R-0Y`BrHK#?!0 zodPy+FSYGT$Q#x$p=zZ^1O`oPqDoF~QD1lj1<;>t?aYVnie&{|I8-~1vm=nv{NB69 zxtFhQQg>;^L?H*XX}c*okCTtRTfc`(a|09KuwH?KEF0Vg+yGa`c%`GGw40)CZH>oS z&8LiaCcb|^EBS1bz=8W~3)V_AmA}2{XA*S2!o8ZCFRhW8zR~j>BovYpICf&Z(%zd` zZip?OtFstih#7$+9cEkQp6$3zlp5X)Y}jDNBbXZy!s*d4FAI>$l)uzkJ<(Pd>1buE zC_QPt7>60Bpcrf1Nf_Qp@W+YAG3A*qD#+YK8Ft=*3;F3&_45S$yyq%D-@;PNuJ7ML zcMT58_(g!W;f|SxqVC4I#9a=wy0o%WQRFL=Sg+^zf+o{J&I7s=_H4xNlfGm=t=I*X zbv49pNa#%#?h{jps4({ckrbh>SFuqBaZm`&$D)Bg)to`~@Otg#ODj=Qu?em^wU8yd zesx$dYD+i-wMy`hLt7|ucy0velj^hg;--jCmr6~Pv z@pDFusJX*$9)@*ct{l=Xz9$-9h3!s0br_5tVR!$?z3+v|qAjcZa#$I_6h zpB=OP>w$a_jn|@YjK|MONu=fly+*~wh=W28=Z=J(eSQjgU#scFsFQln;Nx9HhjEF# zOqDR(`D(Ofv00xuNn5j6>6`y@yZpv7vHm`pzL~^2r(-b|%f@;OLNAi{WzKI^bvm>2 zUj6>_wX8kjmE&^%7J!W9gf$j3vOv%;eR}?Z{mOL9)~#3WEJQmWlaMJWj{cN+aMJ6y{SMtAFuuKp-0kM#G+C5&=nqm1AE)~11+*k^|D zr%x{vPwk3-*0v=R9`o;boCeN$SN1J%*2AjSJd#dPnrHKbTAh~&Pw&-iIDIO4@#%Q8 z=0U*_J2Q8MrDF_B)yY`ps+Pa|LUQ^0nZ_@S5c!==%V^)Qu+6TmnA1^);i^&7tTM7? z!~V$QT)eYaO3}&oFeDSb-_3oUh5_BP+xcG#ynGe1ctyu>0`BclCY?{bgIg@j+dm8B z99d48v#^SBH9hKE5cM*qHrnIOMqhY^jWt?x;MNe*Zpi9lA`5*dULX0KTtc@a-PuUE$g|0WFq24I$0Vt zh-ur$Klz#Mx2Y7?Kx~X)i${TyAUs7wcQgJGbait3>*=td*S5^sR4i58oM7EZH|F6*)S(i5vYB9IsCT91zgv!~F?e8f0F{Qj~viW2g)xl3#mCr$9 zX<>AmKrC@Nx4_NxGUzaE(WBLD@p9p`F~zM>p$ROw4E_O-4eYOPIWF6oSF`pXfY5C_ zkyu#EV$lz40NZx zSsAp{$(wC1SU}~mPLA{qQT=E#_YzGiw_z*QXe?E2%P2Z&5R>&Tn z62v5;kwAh;b+2~PbVDr_H>V(n;IWO=<@AtO;T?8fy|pCOwHX$OIA=T&lEg22qf=w^{_zuegQWu1HS3*0h3PaS zjz{ONKgRRD3hGfHGsm?#8+#@bePeFp%B3}GS2_(UpRM6rqB}a8LvuQ(N-{_~FXXOC z>M`Tww_nz@zFFa88g^VBx#%D-VVz>z-BUJ`iql&!PkB+{pw~a>GEsLTGRtUF!rYwG zydYMOCNcj(h(4JmsQB)&zEuGp2U?+>1m=fl?&_gQ!QZboA=jpa3w_DK<&|Y_CpXdK zA173e*1u_RZBd9>PI}LfRkfpQIr3Xhyf6Mx?{Lwy6me1Jb@8#?sZ>eElA3SQ8Jl(?-z{f7s38x|^O%CX zzx?U*d!*d`kS%kj^v#8bX%1*vw#6GQ>jhHRMMW^!kY2(|45D|^7Br+p&W}7$kT85u z1SDn%v6)tZ44qk(RCa1($*Jnj9ZP1dkSz?~1 z#AzDo)tA7m>y{?w;T@s(4kPfmn6-h4M|s0OE?A7I9cNx^@LZ(`xTqe zrf|zXPQ6oWrUY_Wgp6CwipDVG(MX^W$)asQT5e0<6_`wI!4n?Ji9SS%81;q z_fdna5`FYJHc0CbJ6q-*r#gDX4kDS#3S&`}bOWn6%d;{|kKgn$Rn<abrYR9`+(nv0pI;5-4N%44;34l&hM0cql`k`>3@V{$Er5*shO#f%Lb&GK zM%63v(oiNVI#f?)i*^US(y}|%15atzMZt1_14S8^(I){z1^^Mm9q^a!XPR^f* znXV^VKAzGSwTe$xeOa_`Y+58DdEgxA`!z=5H-K4x(J+E-o*F@KE4sr?$S3EJzJ+4P zygmN-L|S{%rg!59sg!R1^c#c{VD0+jf`%~h8++!A{2B$g2i_6LDvU;Cy=-nfe*nSD z)?vedZr@hk8k!g$wdc@QGAeyp-lb4ldYIqLfuv}8*n(r;FjJ`d~0Q*72gmNE6B=} zU47KzN_}$FC3&4e)#TXpZ2o7nYu?-MUbSzTLWCtu&3LykH-)2yc*;qOX>TPBxT%qg z@@vUM)xO(CZdhpX6`RN_d_?x`(|xz$`+3#Hv6`3Ydcvg|rygds#PbqTNpcs}i;3!1 zTU*J7EzMYIY+Kvq`ckr?^NI}S^EZ{6|sNM0RVsU1>femTQt;|h) z#Y}LX7d_8)r+nNt z@ieWxGAPEz#l2D@FkuNoEI8O#?bS)o-JoS~<~_Ng`6w<@i$74J!eA5SNNmL2bj0}! zYa9mqSgDYRhBK^3M4^EgoCTKR#W}qo>3|0fz+gNqlA{Lk{c}tKA)!WxUo%_rY}&?8 zQ=J)Nr8k-8_cvL4qAf#viB79w!K0}yyD80r)hQ^w&O%<>Q%8{elan7fUqP`eX<%erJ`g+If!SHAA>QG{Xr+8D`*W^nZ19OV1dtq!}FJTNme zKkY0b>;G!x@i8G}3Y~*VHhavEC2SxI`|E6CA>En+h5}jqs|FwS&l`e0L`^07inEuy^J`?8XW-PDvORFK1|`3-5$o+pK2M zz<#M{tE0zK69;1chY=oQ-AGlgIAw>$IthV|YLtYL*4(Sox_@>@ke zV~Xq>L+~&!z}y@Y@JVddDOu$7XMKDGpI`V!sd0x(Q);8dkd+M?NLQpR*M?ZXHf)!0 zjueI@t*;IE9{)Pex7ymu_cS7)%&70hhUrgzV6dC|CMYEHp@lq@C@k9cCBedvZJ%{EbCU)Vr>~@Xnwe73H0G1II|->9<6eNnNx=FL~??cd$*U~X$YXlfQx7`V0X6cpN&U9WhNA}o1w_+~Fl zEmipT9THQ=+Zj9?+0bAJ^X%DviiM;=Z)2lqu77)7osc)|ktUMP>6_k>Y=5UH)bhF~Nk!~}SWaF(}Dt*1rX>Ax3aN(5CnTnnU&}v-i zy!Q8nUZBnwT7BdU30AARU$-3*6>{qeNyyHXVn+JzR`hzZOu>Hmo(>J;{7%Z)hpUAg z5EDp|B&vi5FV2oe&G=y}WKsYXPQ0%fk$kdOHC;EE7A;I0XK5?gxCLM zIunK364b1k)923`)RQ05+s%Kh(0XAhb-kfc@Rg9&{~EzE_s+x#+fP^`%MoAO7C+?- zCurf_SUgndxUUwSVI?Yj6cAGrS3ZH0A82!g<718wk^EVsd5v;m_sIj}Y!DUa*^+PG zllbt0e1*Du#C2uR<8k!^9Lm@yk8xJlJn3Je##w(nErnLxo!esB z&*>fIV$>=?eOG5;KaJ$NmvzfOhnYXIyp^zn3qox{=+)ukyb`U(;oQH%GG_}&28r_y zpbp8$aC#BkirF`dp-KhkB5C&Q=JNXXcC(Uy&a_~+4t9PV!-E-s1`f#`ylm7qq0TVi zAO2`EQsbGjtA%cf6|u_HIWFLXjVu;&2YEd@dr*>o7#+#RR{yGdr7wvR^L1o>&S#Ek zh-weWocr2Q_U(g14?L&8A6psS76PL8?dMZtsTSR?x3kmwOPS7A-MZxk0g=3HV=X6( zklWI6c+s>W2a!XlnLr}`Owyf;e*U36g{RoI8-w>Mp06b6Q($JAI!_|}?^W<+0Z4u{ zt0qTVtK!735?m2J6jE~w%R-7;yy{0?p=+z);HwI>emNE^W$B3l$vTb7o~^g1 zqb#v67LK?!K^f`LtPduk6PSbiv8FkD0~Kk0t{vAvaGx_w-s(CuyT9 z)#2V)Ung@?T=vWl!V@DsNyj0`<+TfmpO_eGpEBJfTz^9Tdg{lqnUrnF1_#us+_gY` zr=im~BDQM$m|z>%V|A8T-`GN6%=Kt}q_dAb%f&2SHroK@VlCupsn#$L3SF$skG8x1 z$Xfu=n1&GAAfJUJyC8HzOvFhdk|T$5R2K-SZ6aNK$JF2p7^Y&kkV`}^=4ca?T$4?? zzkg9;$%x-oUP?Pf;Whc}BD2zq_DJ(k>V5pFGh{nf7FS9C7KlS+kh~fTYeB`#tDpBu z;}|)*aV94WRP&n3L0PLF6z$`U=~|YzWL>g@N=6UIs~l5ojQs* z@1GXggFbCw+|CARnb|Y{Wn+uBR|&b&__}Xr`EXdaFzWa`pZ7eX-VgQ#*fDtpvf= zo9qQ=pif?0wc z<;`VfPi|XhAq>HrG<%4{JC@x$$?cMXLTx-F2b#$RKoO|i^8docX*}J9i4e4d< zKIbpG#5=!db{68`u#s}IMoIwza@=_(f#-c@LM3`-%CJjP#$Ue3Ma`-j0E~XS8p5e! zi~vv?n-Sm%9*g#ZT&vk5ypcke^Z4`IlyZ|Mh)@>uhG?(cx{bhhGN$kHDM4fbC)HVM zZ$Be|=RMw=iaL-E?y(wgp7Kou2J#S88EU4QsD!SY*xD^*lnmd>uDaUGjtN=y%xIQf z=@Zt*sg55psEwDNCifPFrxmB{@#%`ezcOy-r}X!C2_eZoIcmWg*SKLQSdC`>kp+3S z8`ZH_wwk~eTD$0>&BN1oWAI43NiMh0A&EHGF07F{`8hePn<~q4!s3&P^g>f=$X0#7 zQ#!Gf*p8<#<2z=@&hxb6hB=!dV&AUwGQUyD8{PVC8oIE->A_AxS0ncCrc3epJ(p%` zJg0s!ei`-6H_sc@<4lU3zrcv zuo*Ry`I9f(O&gRg3G3{DBdvr0iZhrpvsns=d}$?7^JVRCHXpDo_ZAF1FD0zbc}x0L zvulz1MJ}N($zZ1>zSqHnsOd*0L0h{R%)cWI_uJB3w_$4#GeqfJ(vT(F4$qTbfRvTf_|J&3JR-*DsFZziPKV3081+Zq z_^xU87zRxmAKa5ol(zhalC(WQY0l2BrT419ZL~9WCngM)tyXt#i1l8$?wfu++VR+` zDwf!qHK!@a(*a7pDp*147K`Pc_X0p}N-{mKDQfP)9IoT**GujmN zZ3TYgb`yTF1*fWe=acU>L9R!Ga^!Dzs_6y}dT#)^#RsUppoTq0U7Bv=p)qZS$atH? zAqqF!Ks@!@MLiIt6z?G0w`-2xN5RsNHqDNNbmb(oT@*07$)@IVkd+JLcIC5zS9G$Yomn%v#f$?wnnRg((^z(wI0=t^Mxpu>4i( z`BFjst#TTQ!*@yA6TPuLix}-R?zxCh8(5t2jL^u74I#`vj*Uw1Dd!2S*&iCc3H#Ht z?#-6Jjy%!=NYc9smqaf=OsxBB-hZ*{{knPWf}?JgpJp?j8JhVU3 zavZj+2w)NdDBS?@3&gf-5_XEb0cV%Efpc=2S>}@fvN9gXMJbpYH9pO}M}UAx&;Vps zh{xF-L6b%ghfE2wTwZq9RRo-2bJH(>!PiQM1>{g4E}lU;h5G^GLMWS!&KEE54)My$ zTjC_?fLAC?SxT>*H{q2dYO;=7PAyn^K~)ivU-pSs_9q&S$Jzxi& z@=IRW3~|J77rGzzLDRfz)xM)X34VOM90lB@2l(|r@#CTMZe=A|*x!&`7C-k#?xbDo zs*b7Xq_5duJ3Jpb6aSa3p3J`xILe#`}qXc=no?! zCsG4VUx{M5>sd2ev(#dq0`K(GJ{Gi*-2Mr3s)B&1r0T$e@T5ik!xS(EW?$W*c=D#IoiP}F-F(#6 z44;e+N~r#*ef`r}T%<;kPc{LzU>IE%_^?%!G@~--53Irf;8q#?s}&5{6sN^F_%+?M zAKM(Dz^_kc2S7kL0DvlcOG(9LkY)pkzKNlYf-KjgUK`5;PytQ@c^Y9|lLx zmE0Mx&xdv!=`Q;Kub#@?7YY#iW7}Yv4_XP$l1TzG1||G}7qg{9pku2&QiYD2m*1hF z-TOrS$(F(GyR>3TSfMZ*hn`pcvELg5T;!!~hBHEj{Yk9=%eUMto* zu*lWNq%3Wj|5*5}BkA?0dNR!S7gtO_E{E)>ZRtXf>y{_JFQBD;oOe5Dj5A*`9DgNp zLa(#uM$@4*cQ&;mRM{_<&I)tE;c;XR+pCYpdGb`Sat%plR2^b9!I7k->0;pLo~>LP z*rv%VdI&1B1RyCay=K;xzp}BPpSmEBmU&hy!Gq_q_TtfWVsngVW_a0oExOv}#RB@7 zU!k_3YmA3r2ETIf`T*9*>oSm~*tH4`w687Ta&w>Ds2dI~NM*Z3MI>TZXPwq1HiA5( z7S5(L)OOx6e4P_)p2vbUE?laa=4dVQLo$9gHa8TV{yaA4wYT4h7-*1)ySgERAP!Z( zQnG?yZ?vc%7ieVd+7kJ#EL?(WuBx{7(4O@`Ht=^CFzJGK6)(0ak*^uE-z|fqHtnU1 zj`j5Yua_e)*2ayTo+ln@JcU`NSbpzM6LxEAOzJAN%pqam-t@wG*D3otc<_DRK>tEz z=UQNX^0NtTeq!F zi5xy@UR!&p%JcwpAg~ybo?)b)-^Gc<#`(aPa?1PKhIGNPxB-1#oBYdVA*jh@H>&D)2vE;{Ym(V=g(foAZ7@|QiiQd+iQ9i zX-l8$?bDP2jcZ8>c_vcFpp~o*&)dhNSbGGA=$5^|xixHCeLn5=tQNq!XG}6A%yO3` z5l>dJWFKlol!?|INydK{S&@QW;cu={NwqmSm)%3JI`3JW-t~i3`qAa-$8UVJpQq_U z6KXgN`l2S~Rf6q7a0uOkmt&qVbX zRm8WUUZs0GQG;zkh0I-pyp_fM*sz}&oyA?L#SU(skWBiYF3)&$q_hIHg^n=SuPAHx zHYSYRPv=5wzi7|zg1h#whfX#Vxr_XGDIj^CH zTZHLPP(CZyI4nW2Ftw)o3e_*qj|d`e58@o^SZ#a{#FG`bRr2<6kIe zb9}{j$&Qv|o2p%n9bUanODwL3ox_7xN^unt;vbezXkwGXg@b$_>#B=P`9WKpMTLzt zG|j4T^}E!=RNf^|Fg%}WoGt~ccPs1QqlPqjkixEe>TyofWam5oFCMdGsX^Kh)z23q zEK_G`vr=68W2{WCgt%@M%th_|AbK^WnSQ|;qpk(y3f44w5*<&d1aDP^iCY}bCCGV| ziP<0XDmd9uC%kuNgwv8Jr%E5^`Kl*>7dXN7-oGm|zV}jFw$JfJTvga_ql@jbVsc_VPUG_ns$>Sj;P&{z-6mvpk zs`z~CvQMAU!V>a_Q<&~g$J>)eu>ldY^Cv$=IX`B`i)R~6|JV^2Fj!F}EFt4HkI8cm zFMITIo;vEp#)^HK@+Q;gqwaLE9T!;4hTfWwDIE*?1tWSFe1e$-j&^7qdzmFRoegYi zFP?EQf)a0BGhA#F9M^~0k)-`jlr-An8%o-{ubm?U^iv8u)g@)^9y30lOk9k1j+T}5 z&a#I0By%oQ(jA`vlfTVX#J6Qaq;Ya9X7;oz*`{K?2zSh4NZ+Pn;=GvbpXaQ)|2mb zA zjzATH|=i=VpJ1;n4YX_T(#@_KuerT}FH3c2Jk77l42PzJBmsQM2M+C(HEwl|t0IXR4 za^7o+ucQh6H%$MZ{DZdD$ACbHSUlL&$g~&3;jahH$uR#$ApoG{U$ppn*tm(ZN&C1+ z#N?VbxB zgJey0(k=}W8C6_c-biiW1e-_L{c$-#P66!;4m0Ecqu$f%mF^@NawH-~6y$w_{+%Zf ztNqLs%N?OQ7qcD|NtX^w)z%e-U^kWXzK|Br+5v6wSdTC?ygZwsf{L7?{Kzg+8cE(hCZ34I?}KZW1p&60xEfMopc!tHKsk z&}Byvr=&5nbYh_uvbdlO&>BTa!>BR1T(qo1posGb61wP36(_SLR?r`aTtZWHn8_8O9 z9Xq4jj)j5XPI(3lY%S!DKu%scyi|*Sv=-}YC`eHMK1&x-PU6KwBE`a10blg>z-lc! ziJk_D^xi{p02{}R8A{20mGqGYiG;7Mf-(4xD5sPcF2#S+O%NS|K4l)BH|fo_+r}p| z2457~mj;(fx=)Wps9=jCBv6BUabK{;1N%a9njS?t0nBYt(93s}|I7E#%b3gQc4Wuo zRv|0>KgQlNtjcX|0|lH?N=SD|gVG_5gmi-lNQr5)zYcq)QQy25AtG z?k?#wUe;QB@9*2+xz72sE^sm5Io>gzF~$@3bKh3|pUYc5Fh$-QkK%KJMoGlX4a3yY zWB$eZ?8s59sHv5&_-rjW#gD{M{or=M9)0uttG$zZ!UWLYlEqzOp;E;H(Q@oGUkCx; zXVnoEcZU*{5d0fD20a)gJzYWE9XeD}EEKdwKwMMdD-rh@3ss2v0D$N9JP>)LnGRNY z`E(ckSg^M@P+|SEJ8UkYad+@g2`PxenkU|lq#q_1&Lb;(?UGo7P+-!q51-v0L$gP6 z_w=FZ9uNwj;5mD2r^mP1Vql~y*N29nM+uMshH&)$9X9kFA{IJ~0Q>Yzo9}atAE{K) z>)A#I5$R)RP#IrbQ6aiSKa6~MrT41?rWz$|Dtcm3nH zM8yMHD=GF{9570=eNC4%=;@@l$l$K|pOYvaxe{m>9^oX6Hmj3H~M0N@jsjJ2Ci_9&0IKw3RMyi8AA$H>j`2%PG zM2uu;ut&p>4?^!+&K$5K?bn+cy(c9?YPe3{N+p1Az8{VEp#|FSgcPRjgzs8|^ZiNc zuBg=`??|QXz~a$NJdp>O*V_yG_8qW5G7GOw$*x}o+h1xHOdLrHoA~vT-3Cp=fJA2n z|AdYV{d%inX?TK91X?LYT+b|VS`MtXwk~8sGFc|=6SV$=*&U_O`V+~nz(ySHsSo@F zE8YM9S>(aHyoa>%ZWhn_k2`SbHS%B|vf>a_sANoU(E@^>M%WW+6AnfQr?_{$7-ZNMa=_*aB-C)x<4P$|iQ=4YqrxzE z0J7q6(3Je&|AfuT0X}AN_|lpCdMe^*aZ#z2(^6p~qIIX^&(89zJdl-5gx>zm-1K0B zsz|)RCHy}&)*SxW(2((`aF>ILi2#(u8+b`{_;iuC3rf+T2`9eW&xJ<#NcVUo;*gWX z!rJiaaF|$bHARhHBOE#bceNMep%ajhdW;?hjv@9mp20yx>2bNt?^10{qW37;ze8ZL z4v_JaV^OOEvusD8heAe2j?N0$mHO|-apIv;VYP$To7zo9Wr#C_{xVn=Oo4*wD#?Gm z?$|Eu2U^p-TKWvKxPv^1s>&2qPSuW}eE+2YG>H zYd6#oGL)-rG&A28#<4QCdK=B}&e$biO|f_h_4U^n$M6Jr-W6PtC*}Ih*#u>9liepe zsDM815)jKQ0U0{KZonei19?8Ew0i6xc}-4##;Jg2`Ee+x-vo8QBmjEklhmsd+QE?# z!QYKww_Xx7->7K%o`ZLNXd`@KYkBP6BhPpUl^J~vv75*ympfa zNdOW+A2^ejJk@mhGaHm$Rf-bu?2ak6Z|RSpT-Uo^5u6eHhepiXDgbL)1`)L^U2G$o zNP2*L$=3A#$>8*~1Ej281BfkpJwUw=yALp{mxIJ@?IzTl=(CIm3swb&Fcud;fJ&-( zn+I)bYHF!hblWxK^oP6zfw0K?v@!s=pJr9ml5mHZBV?8Km-~c(tf}!MLZ-*HtH`8J zK`UV4l1r1(Fo|A6*7)$F1Z0gwa?SY6J3wb4Q8h`h<_--|#)x9j-B8oFwawSB_be3# zDjRp#Pi_H8Vnp3Apay9(p?U(*0^L|IkV+ea?V#KjG&>D~u0SG$OlULE>yDur__I4d z*mVh%fki-5&|)xS{E|7S133>fI(K&Il*$tCYbPwuTx@wHfo{!pOVB-Z;wncCWua-( zZptXi|I@7*!yZu7D;RBKThX%3IV@xPtL2^4J$(!Ib>^|bkNa+9ej7}FIa6(xtxKl{ zKT~HBPE@{DR?Cs%8(UkPLY07kL;&j2x-tJr$kBLOF;MJ|mi~l%7zOc4dlRo^H(5F0 z8L7{1_Cb3AY&V319d)Q25(xtuwrhF$-GK6_zxcb7F>Sq!DIH`v(qV$v1B;B7;!XHb zlMD`j8`y+UF|QHzeHr^Fp%@6XDA)(Lc~!Zdl^^y*(xwUo@AOXyzrq#Z{I}Uf1KzeR zZBQ~)7EJMaHZd4XKb?J4;@-Ls;M*rEru=&&cGlKBJZBB-UG?!PpbNA|-(-@>Ur<2c za5!tS(Roqn7!J69D80hqYuc&KSRaS_WbnkFU(Q^eAuK_k4(HdOKM`5VVf89Jm_$jy zfhffql!P1)Glk!8%yD_R8p=BP1;607WGqK2M>W$rHq7w5#?1X7-*;l{pZfa6?E!p? z$*=oeD$~U-&gj=sokMlb1nx4c#OnFss-tS7C;2QY9-H#2pByC41CDQ=LaaY7!3me0 z!^6Xu!v$T_H$r6*>5eQxI_w{hi2P;)NxOkYfH~BKi#(eKEp?%DVZF(sqYA7If968M zcw5c!Sp@?w{W3ztDNccN07hCy$L;+*xn|)TU+3JD2k&Md{{AYb-0i}(usqBy_RcBh zFRKWcRZS~PcP#BwS4ilwaIOz+56SmuX7~O+Vy#)%Cjb$OAjmEZQbgSH^0%3!J4Zah zIp?5VWW8r>_&zh3r51kST@D&u;z8A`eKH1~DqtJC|MUXXtE|9SX<+^Qk~wQOd19WSOLcb;+v(@y(C22>vUm<(iFELVSKt!@S%pFn&@poRShBv$^(do6d>{+?7?JPfl^ zL#6fr=l}4>#tC1nDyCLKKvKOU@H!>9L|;+{TB@jKGr2SiHopu{;z<-7CM5|o|PUKjzA9zr9g zqAdVW3UCMK-4k#QgviCh=|TK<-AaxEi@QgIim69N0!y3_!MJ!ZsU6sn0`1f!z`fS< zgVJ?(u5~NxH925JB9!YBUj789O(MA0k+1VD4_PM()PT2ijg0I8tn`Ho(6J{&HXT!P z;8a4fhkJ7d+x6<}lIiI!(tY2Uk&Swd(^hq8Vq&Dpy#w8Q1Ga~2!;feZwUH_>({n!1 zT9i{d*7Hm<>lIZ>f+qED&>UqMbMT@kQNR+2u06WT3Ji-G7}nVLaoV_24B(I|(~(5q zHt^+WGhV34)uUGx;~!35-3rW9GPz*VSXd^wpvFCft_iYRsF+0I;8>Y`@{6TZ+aj*+Ps(8E zSCj^2Ib?tN)+|35HH@O10!yECGtQNEW8b{@-f?|q(h9W>$$=SjAC-nQ=6}bcPz-R} zUuv8Hc;1r{u{T~rw-^~%Vqpx(mm|k6V*>9}NqZjy{rZJqbpFn`*KbRM60k|vZ(m!L zaoC)Q)Oy|27Z4Wn7$3N1_);F{R&lcwsKgYkW@zga6X<>iJ)6m|8`a%01GGS_HXOj` za2=}|aT8&x6cj7rF_460#f6mO^NC+XC;kd-eP7ErEV8Pg8@S|0V8>OUM#RxhxOOjmcDQJ`fcqyz=S`_LPm zjT{Z7ZAVi}r|Cf7vc}S04Veld1sMTMM9b*4M7WN}@jrjSsOG5t-n3>%KY z76qd5;d?aeTO6gzDJ=KB?9Z3jzL$G!0R#Uurwj<^SE1Ddf zvor%YEiz}bMa>9x7t^apv0QRyl&CO-`7 z#Z``#Dtj*9+kEX>F*f7VY0MHb5Wb(ojt1!!>X#P0Jh1Rj^ipa3m_)~h`V_axnEqY{ zB$)TEgV&2C*Lf6lRC8d3xBfq@@H$YpPzK|-lZ*?XK_#XF_eSK}8cpaF>}Vir4ZENm z@I^dl@kdEa#xaB*Mnol*3KJBnI@ZtbCpHk(3f%f|3C+z13!mp8mGoDP3PYP;&;Moo zj3>!{qA}QdBB-DUTQNHbhM?~kLT$Y7$Cr*Cw+p_FF517MqdHsTNZ-elxc;)?IN(s4 z1M@Z)C@0ezrwYm6f;J6^!p%S&=laC&7u7xJ3qTYOSjzfm54~I}uGa~h>@j%vpyBo& z5DC%&lk|kxGyY#5042K=1DX4W64_nj{EkL{@%%a4aM955Z^j3ERYX>j?Ct-tW2D~W zzr(qF;L*&9aYZ-I0K=B0(@%r1T*sBfjLK+ky5f6EAV%$Y{xYlMkDy*bJMhRy0>+6pU>oyc-OXUpwN05rmiUsYMT^=5V>%;`+Kkno8 z``+@GQiNjt_Yv`i8E1;8d&uJ85aD|R9h>sNny+^=O#RqQRi)j=4=MQZ+ay$N|&GjwYe5CkpzwGy|$oduH_% zJEVGQO6edJXV?!O&73a@kxwETUJ-LV_6!L)_MOC__&CrB#QaEIKo4Ug!1oLe4GqPU z^I672Q3J#?z@4j;7aATT16Wc1b(o>Rf)!El`k(#Ga=gA#S$)tb1?{*Y=!zmU-&Iy+ zWo51XtNvfVK7CEk&Fy&VyW6@1p?vc7ihIYV5qf8#=oVUZDG)3Sa#zsgD}bw<0bHah zi%Zb_ONsTJ;1SP1clF=jT7&s~x^OYCNARD2$A0wt9YZ*`y2ME)xsb!tuc!A8!$m|y zx&iF)ISmjS@?hONJ^#}#>EJ$TFT?_enEv=fNVx>75CCcD)h;?cJKY{iuDur2^XET2 z1K`r$g|3#nPiwD}n!t1Y4J60-g{D0-UYCHuS~ibWzs%qW8GoXwIih!#Bba4;)C5%V ztatbA9et8cz|WG}dU|mZ1)wfIgO$FQ_)+0Bi_5R$9BAe#Ia;BjA^yiLDised=y=Ta zza3??JMn`__Dw0-KokdJzKeqsmq8Jx0W9 z>`qM?F@JFjT(y~cQ9D_cxYhf(E;WkTSfK5@Epxo==J$H^%{Bm-q|V=LX?(VeJsGi? z;a#CP)WcmW>v%L(FOsRHcxPoxDPd$dM5R4;RK=>r#)pPK{`u7N=mOCE$lSw6%)mF# zjdy^eeLo2>hg=DQMl$B%`me6zE(k$Q^CnV(d6nTm<4Bu5_|e&&i{G7-R)ZW@CFXVq zk6jwuH;)~@7cjVtn3=EMS+VnNsVLI7>l@WYmcI;puRD|o=wp~SYYg}{rF?KWx(lm8 z$ER%y@8-G*e{L+!(U*bm(Zfd*BE5`{RCvGe;g&BuY$Xu($s!z|Inr#Z#utE_DZyV3I=`IJtfkP!UJ`^SKmIjuCt=~M`;LcoOTi){O!1}Cu z5;4ox`KP)q7>wmG#{E|WQvtqOx*>bHWFWsL+b35nQq%?NLBRjXK ziREbD3P+(p&B-V~M%a8`N+%-rP6cEYJnE7c75ZOnJ|+k`rX+|WJht2|DrRbjFL6s| z!<_X&9G_dTKyS(U$c_VK3;Vx0Ws+WR@|`;nOI8m6ju~G2`GB{#XC)}?*#khmIlEN9 z-^<4!zjXQa3hd7Z+f7W$vGL2Pi0M!Yflj~!C9##|8L6GEcDPySp{bqzk^U)iK^l;3 zaU%eZe5MB+b>5IygHh(Y~a;;1wt0Z?Bj zL9Gx=K;q2#EUrDvlNt~;FKMTq>S!X;Y;N0*6!3s_Qz%3>Z2D=I_#!Qqi2&`#PM``)JoSY&B5!)fX{N)svz4)_tAygy6_sJ8gTx;Dg(V` z4XcZFE;I=9?8E9IFY$k|!_@4TVV1Tc3&%nJ{uq0D ze2u8KUi{-3!VYnC3^9;E)TeD{VOf=@IG0fJkrg%|r9z)}kR}ob?h&`8A#L>@-9USM zR!m@0uS_!uuT2U+;O$!k`ZYW{$+1xwMOHc$ey$K&zfsoJ4H^u5T+VmfG1gwljW;#z z1aigO7S*)95UH?YhvHNtu^!rs75PWnDNLqsRiEXO;_bb<#mNn+VD3KV5N`bX z>YdAcxT2QX4IYM{J8CVC9F)H#gyZz-_jNYX5*-Mx>Hj>gGM{=@M^MibqK~Kg{!=3E z_HoI&&6KI!coy>6xkNybR*x-TYoBLU@{vSac-25-qW8L?Pi3Y{F$R}R-FAU)7 zZ~u^&ORJT+*B@f~4bwc|zQV>**F_L6TO4+G%Q0HWWwym_cjlz8b|%`)R5~d3c|11^ zD;iOKWS&1WKNc#;w*D~))OoQ}`wu!BKEIAg2-XxG?^;JT*Zi^qs^Y zBNsg;ZLKxVV@Bb_ZyJ8_|0!vA{hJAfSyPK{0l&k!wWK|RcFb$GkY_DK+4-aaoHr0eNFki zyPPBMVHzC_d*7bzhsFx*+UHVAkn-`V(|qirniSQsu03AAC_XkkYysHcn`&q^mCiG= z>mf>L;hkp0jlrN#N?F~2%t9i=x}R74<;W@EW-{fcmZy0SOgC=-@j1v4&+!I8T=mLT zcwD{!YR_Nw%Q+-BCx?66d$g__G}vs)euJ2#0M9d}gGleP#IZ1gnbu?r^2qpV*Yg)d zwcC43OTBeC=)2te>fzyw)Vuf8l~t!=(-5o!Py7@f#CF`^H_>K3fGPs3Yk*d&#B1Gf z#>RcZnx%ai%C>07>!`OEhwe|1SP+QgiEeP0dmq}A0~~T)pVs*5b#A#Yv%GxJ>+4#1 zwAETTFV@($SD(+%!Q6%SA&csCUlu>6SsC23X3wT}5P!Xv+&@#ZD=d2758+iSqZ$;4 zVO*d3DqT#lZnEIVYI*ScTLm7L5yzEp#=-?6MD7gcr$6}oUbAk(FKj1RB^$YVrXp>> zxbXUCN_ujViSsw>+TcUjMhTkZ;`fU|xqC63sL4K{jhy^}Kj@oL0g$JfNUUJDYU=?m zd)7CxDI$+yj_`|ze9nEx7>J35^11IiGp-h2$iK|At1hnS7kL5L(S-FMAwY#v)O@)* zR^4rQyFREj6MIs=H75ZusS*2KmF&?0N|#H-L9%?Q>_>9HW)>Ca@3KB&o1hig2x+k> zv-*12_0JIJ_6px4tJ(m^&*H9^)&{l6;dTP?lmlwPqZP}u3xyfB}wZ^fwo;RQ0^wxp4!SUo$sdTK2b;Ke@T1op4V2o^rrrKY9+7+Fi9@FkV`Vd_n&3 zIofnYh~9_W*wl(q0`?{=$8w)0DBqS~m9mOl`a!7k>=SH+HTnMZJJ}5KHJOyy$UKcG3c!X}@ zGH>xV1YXBV3${K9F*(1h7aeNvO@M&tRe*5(HN6VVsL9%<4ql;zqU+~}K1X1YD8V(j z%*ypt)$ds4+dcrf+cLgl7Yjg02w@Grwg7aRO^ z7oFINaOI&s0PyZ~<@-p*H_3EwPkrViCIxLnQj>mamr>#R4YxS~!oj(0L4}2acT`nMFHmVe{9TjtARQhu`9^V) zQ0dzYX;;=mDZcdQ*d7ZIV+BUf)iL5~86Ic*cmUA7<<8Z2*Kv)3Vqf>4-FHh~4>Pnh zv53Zb>T=ApU9Rs&vb!+;z{&0VYy$&aaPO9(AqOwk&ILqR4%<3!YvFmDP~&W@d?=aO zC8*L_zDu28jnKO05>YzuWEShR)zqU$;+4L|8XcdT`@0OXCvvb}AOtgKqyUWbMFs*0(0UYFFI}F1?Brn1F*(;U^7?U# zkh!T)O|2rj7^B_I0vK+D!b|}D%w7IDt1F3{s^oYoFROay1-?=esH#Wkh2gkvkdoe@ zbS`*PZqjalv{Aw~>05Aj;nJB*!aGN|V0^BAd^)nMP%QXnYN(+m~IFHS)1Vq0GM{j>~k ze~X7|R9-*hHVz;4F1J58GcGUN6HxsehqENJ@7w;UVHV^7Ts`4iR02?WA=*JZmUnn#;U4e`QW5~tAOPbd?TS}VIaqPomVvuuRZ@@f-S|Ol2M)j zUEECP>=45zE-vF(M&rGw1V}D>8(G#X-;3$5W!M11sZMpn74%^YNs2qV(yV>_n-WRZ z69_?{E<^7I?HE?;5|z1H0tN&dJQooh^Q?|tG?(<2Ye}1S=h-Umt>7cmXT)ygK}7qi z4@cW>)AMG?Sq2aq`n8tIy5_vwY2<&j9UyB-`!&h0_{@oxeYyACaIY?-$s@`V_qDJ@e&BOVZ+^R3RiIsJc)v3Y5 z30?etQ+$%P#oa!At@kLoitS6MnSQSY)ja#AL^i17=|!N7a@mKtoPYMwCb<5)#yz2c zlC+!&P4a2JPP=D}Sb}>GI!6~2C#Hs@il(QM5pp}Fa~jPB$4F#@^3ZLoH+}3BfoCPu zxUv;Ep|3wut4Zw=X!lT}&A*dpbkwx8&6hS9u4}h7!O`kzpudAgz3@acoq=iega2xG zaBA?0KHnxCL*HPfP-^L}94+pF2H*ZpbM51A)2Ng_2ZDDEtp(dBLmd;I{*gv1$FAsPW9$*{lY^bEhlNZr^p}EE6_L6*w zfgQc9+Gn!LYYHq@o@M6p(+6rf>&y!^%WgeBtN2dSaT1!^40s`P(*ti>PqxBI9JQOK zMz>cV;_dXD7C*4C>#f5|v95ZGBhm37pOKYY z%9`)6*JNWOOLt@?i+nG)`H^H7Jm}S^ulSRnLw-$cZ)CaIK8RaKOG>oP%`Gn-={@%4do_iw|?{Mjpt!_s&n8L5$7JWeJO5vI*|KD!7N&VBHV(s zh%aUY6hBiFCnjh7CRxrB#@>A~2deg?p?=L@1jIbIYN|3N^s?eGOkz`fDt+K<)-ciOSp?eK%GixnH z#xl0s-9gW2ediGd>$(n`HS(9MpD-4rem5MHJXkaH7>XBrsLki+e1UsliMHzTTwqLrKNd>>d?=ilbiKV)$`9*n4QZJlD(Ni#1dt5mC|8eB|Y<57fXDP zhWGg>a~>4H-v+-AKuny0aP@6x|J#iRwKCJLBgCv-PhQhV7im@7$Hp4JbY9>*g8v$g zIGo;ivY+ENXICL)VZ`a;)1IU+He-kGks$l(droo#=akPfa#=N-Sd8h168Fm3&ZkwI ztwSkwgx>UsymxZoPX^O@(zo}O9%v=26jx$8E<0J>SvT5LMb4fiZ3-U&;K6h{XuXLC zIkD~gJNE+#xm)diCQazeCup9fm_6FszmtRmN0@A;$h}Xy@cR)tfOS*!lfUFf5PI~c zuK+!l%CK&IZ@`oMFy+ZVKR4ItZC$*W909p6BAH%YUEl}ixM>%A+m_`uKZs$3rk6;2 z{wur~Q9up?GL^lxO|$i%c@s9TY-V#AZ68&uapdKNS&E;SKs0h}mZ$Q7gMm6@(nP*1 zB`q&O_#~dU$pm!tawe>R^1NlmR%_4bwovQUXuQC!bwF+TRKiFL18q{3?#dLQHrc=- za*bOKaAx`)L0#i)(MPyy*PchKf&FT$kH08%*O)g6biR zmQsUUu>EwodE(8u@ne2+0C5(cuoI?VNx8)8n6@!3r`gUlz#{A(zFL8nWo1-IsdCoo z(=fcTEiub4Y#FX7FyS)VM(<|XI@o&{GR5J8;T}>P@#%2vfh(5mJ6V14;cql{H~yS6 zLG-H>9Prc2pz+*JrEtbR;g24I1#7bw*E%7xx<=xlT5`WkYR3 zS;Irp1j-CPJ*dzhbZrFjcpW<&@vx6jI%C4yB2i>M0Wo7depKI=EGl$yLuEIlo@H9o z&t?*|_ELSx&9M-%bWHp}fVxPLfu1g6{Ao~npM4?V62y%BokZSy9V|&8`GxE&O7$ty zd)jr#P}dj!RynXFjp!xemWJ%dASBuH>wAEAK8ucwBP=x4@Q4eN>@ zFez)Bww{RVJNNq&tbdOlO>n0SK8-Y1pjV5FZ>jmp z{e45dl2YYR#RjM2a03hAqrX!=ZS!dB1Ww8@1VSI;`Hu`LJ)tu7y2mLt3=yt_La-Ws^wDYt}?^&{<`%!|n z_b=OFZG6$2q)upQrxrkU-cZy|-Qz|YHOYcV>7+y5xVRBnE`DmZk*ZmF1kgg=WZ>ej zf`nQBu%$iZU_sCw14_}yzw-04NAmMMZ1FTUg@u7wl8?bMs`F2E_+Iq$5(l;~I zd+U*nNZdU1TM-|rUeoo6EvP)4mcFR}diBKb?CFF)u{1KBus^Lvh)E@T@pjV;Uwy=2 z@nLLC?dZX{C{o_iuM`~PcZ;!wRhCgi~VGm0unbJ&hS4iw4GL|UygTPs;vmkLg=~j$qXxj{>)><@X*4A z3^E_hv3mCHa}uMQe#3c{pA$MV3Mnb>xua#){yc+dSw3WVx&whU$Co4n-k`yE5j5uX zjy<48x&anwmKh9EV^6a-U#bJ_YP~A8l#UVu?+G+s%O_SIR}a)<9R;NYGBRz>1GK)h zL}>}TOXf~oa_3HL&8W~>e?sWf)qPDKwo8(ed7AGe%JN0XwDm`y;vufQ!r~j5%%3oE zd#?=3$(v5UWcUsfO&3v?wNjLIqHeG1BW3 zInF)`cgfDXGx)2{eG>f;!+u+sX0mdmJatp8`5@0F+cR_hqp5F7jsn&@e4l_9?zhdXFep=w~^&IgAzxLUYJnR{JfP1$e66Bd` zz6+4-K-;_>ZKBJm8R0*Y)&e28Zs^paWx~x1W~gi<8gJk}psV>YPz*oJUtpg$$SYdg zJa|_RD7_aJqSa#|IZD?jp`5=V#M|WT08qLtuz!R5d7HK0hVNi52>tot=T(4lD?bE0 zh8af^Yinz=dejgE@}7PfwpVCl)CAF>^63Fj{@$%9#bx*@D0#*E(!Ma&UJi0z zdN!Pb1*P3qBH6x-W{UfZo;ex4J7XOYG1SvNuH?L%N- zg!Y!H&0VGyLhb-%4?Ltw);jIn*Sv=vzPX#jwpu=)EZKw1Te%QEE9qHl1{gW(< z1*s0qH(Jxh6FZ^X<86m2KG?&dI~!Kz(FlqsILL!CkIo-MJ~IiVLmh+>f=RhlBZ?vV zzgS1AJA^m9!mB)+==!6--bU5yliP?1_1|y30<)V`+OxmZQ{E0#Sa%ARqf&veLV-z> zF;r>{HQ8IA!8jc^8Psb!)jjLRgPoHo(B4|~2#bk;WP&te(N22`;Xeja(h#-3dm$JD zEio~1hZ$dQm~Hdgy2({-xoxUdcf8r_OsJBufx(RPW>x0(vs{kCb+P1$e4feoW8%aX zH7=k+&;>G|98kg37u>-|EeYk`S$wHWcjT$ef@WAVT6` za*wa~0Ym;X2s4iu`;9bM2~uh2A6A+416}qL)8+caPn%WFHmg!MPrLwJaqKwk$aVNf z>gSw)`U-=3DU&9YAHPneIM{sehw4~rs%v$6LwLGk9LCfj^OEqsp_u&;{mPW;Zfqfe zqk&`43y}ss4*9KFwnwDz-V1s9^xZu3x`A41&#{U4Mkz9+-Gzno>+1pQA$k>mgiyj_ ziq=+#PfCN3@4&Fc|7h z$i4qnh)_@777%m#bmsbaGWX%-i-zMXt?Wn@j;5f=OT1yzi?%q&odb9ge+V5;;>%zc zuRY-}?YSrNtW%rT$@fCT(g_~`ikr>gO;A|-@YL^cIOUESsJkdTud!V`6Bn5BT23x= znw1`4%RDGYMnhtZKOm1hyIur1Y_gK&HV8Z|RU90gy&gySAUvTkb-O~10PO@n4daDG zOfG1*!e3!Pqlo;O5(Ka>N@YShd{&VJ8zA|)2g;R9L0_}Wr~vtf`&SwxSo^B#8c@uw z>d-TC`?yJ-*d_cZoMxq!DZ*Wrl84AKR8l@kDbVIU9Z4pRk5mDq! z^*X*!HcQUZuyOdryg*WBODv7&Z`1+DB{vCO7W|cX>j(qtR|mgb`Z@p!Sd?m%T$H@z zyY(RJa+JjJLazKkYGJ720k-cI$DB4Tx^wZ@t>im?x?7plJXY7+H zXTp0%=ADL1!o>la4HzESiS)^*CPTbU1cmfW_P@vOIx+SIQSevRT7 zp^Ca;!>H$`t50$ITBg>uZ{+3f`w13!PQwf{QpyB=43*9L^_7x#G>v?u^d*>x8M$I&XJ2TVi3tdJd>hFI}>c!;aodsP^O4-M+#VS) zk19A}^Ab5HzqZGRroWq&d)JA!CJ8DICI)WFxYzB@lQy(>yWUmnQ{A*TiQ>(Y>3YkA z6v`2br*Ir+4&yL6jLAS~%T_nN+xfK7BHH^d$(P{!-1L=;w^kKAWr!{l9zN6;B9gZP z&CgeS*&wSKIbuEeSMC1h|PhuvUsmllQf6q@SY7 zOa5`Sg3~s@Pxcm%yf+7`0cpo1O5WftzEnDq+a`KcqemlCQd)bkOMe7eovA2gEtmb$Qnfw1lVRgp~6Hk69YAkYb5c%r1?Aev5C*C0R zS|DT7X0z98Gi$W|TQ~Udi!6Qaj^y;DX+pp1Xii_EHOXEW zv4Nfc@T%F`!hqK!w;Vs=Y>i=*T=GpE+}QaG8z=Z6B3~$J`orc|mt^VB-oK^^1Y3S3 zjc`W3!y>BDF4XU-^*&hs6~tKP_R$VLFO*_lYf$N97I(VXr_dsgI9?c_$q^8xdhT6| zQ)w}T*opCcOOs?x{kB^|{g=$4P@{XucP%u9VRK8x!}t-JepOo%g~7E^K#mqRVT8Ul zA(Hie837YzV(G6~x<6h!R_B@C+mgGD>PKtKWp?bNa8*okdUC7#s&pEzXgGaIOK9hK z)ui5Zn~eWrAv5Uc?)0C|K6|dSg@@XYdX6xZ7I)Z|JVX3a1(I4n{8F6MtaCYYGB+a1 z-I|iDDIa5dbgC{k1ZTq+5dVDWN9#B}*GU*j?}JF)_L`|viydBy47gZmkS|NMI%o0t z#S)j|mL587p)UN#N9e?6cfj_F<*Is9DWPmTZlNKLw{L2E3Vp>@{gkO9e8KtW&TNal zhATwkMBi7g)}F}_)fPePD2eb^OQcTL(~*Y@@B6KtfUYXar^9KvGt?BI?hKzH`LSt& zaP*t}PjX@eoZ?U*ha?zCZ_S6KLE zrq}B9ZDW-M>6&-)lGJwgr>$I#SOdrYgOc0vbXMs#ekoPes|mRgDr{oD-Ud5-+Kofz z6MU@vPV(>VU*61C*=`szwcHaS975!gNifu`+r2OqtX-5*cdYTe$W8|z9{XoqIJ{+F19l5utme0{n@ zj3vc#t!bX#`XL?8bBELAZL#ff=0Rm$+Jq12&8O<;RVvvJ9=}36y&R;o8bWlVt~2_n z*UaekzVxFh8Pv4?s(ySDUiq5%;`#3i!slJz>QrMZFIARqy?z_PA9tsVx$@pLDQ#s?izblUYY*HF)B?}Pi_>K)%LtMBn_Rl1kCv~(F z>LSDWFdEUN4!f*2tGT`Q78d#uhbqTu3{fJ=wZC&?vu*{ioo}8pU4iU%d5UsHd6{sK zl*i3_69Knln>_&g|6A-3_&}Er9}6$-TViH<(DWQoVk&r!xz@e%Ijz|Ue-pHIFCedo z)Ez6n)g-SwZwqx$hF@T{G&rV4Mti?v3xi3_*CUtA;IiQ^ACNpep(&&g)J5NNPfwuz zTsD|=8;{CyFL`J_?KS!)hMMR=LFQ&MVO54wz6#eMobX6hv`}UhW6Sx|N#*DDG+EDn zci!hCLaKO!53LuS()e#4T#Y@fOJ)8jOSK@VSz9F$%YsK~R1f+bj=&B83mzcpaF(t+bgKKIu%fWVA9H+9a zGTuwfZAA{T%4$zS%TkT%vFuYIb*F)p6-vVMgGYfsO$3YkdGI}@;S~{Eh%fdJ-$9^xw1p4 z)ifuJP>ip^(~jnRoEeZ0=0{7N>DW!8@AGSBg^M6Ww52FlhDRF@PT6DJq`Ma^ez~Le zYL7b~4Pd^AeEB<6JL{`Tewxb=vN6JS*n|45u^HO$22}yVM$_q#=pLU^p?$_C=ik3QFuQk}qMDqcgILxXnI0CQmR*db9$f#8338$2KM?`NHldKC2Xvs0@00z)QPGbv9mf)0)x(T$PNJAc z@jQj2)=qZqX)?7@8fxxNWG(>(e@C}ZRgc)3ecuC|$2Py&Co@~#7&d z&Cj@Wk@wZzURh7^3P(30(TsL?sgOWm|KnN{&wazL4rFN2+HWyv&R&4fG7x?<=Pyb4 z00QFvnosT9(ixAO_)-&*A+OVH0Y`H;=T{cYzD4NEu zamd&GZ6yEbp5jZTvBC_1aBo;Uq1?-DezX8Bd{azcwtA(CJO3q|9{kwU)Xk#I{r_1i=nE;kGMTHq5x+gLTnIT%zg}j|JSd?^w*0)Z|6nYKW0Vu+YvUjy|*9!C=heu zVe`ox9iFb{&A%jRBTc|;oE#zr{}So_kpWbc#hz&cSz-YatOY$F*$Z#T#ewu{G51)3 z{B*>aPF$20Dxu_pG+k7PH0zB3iMvuL6%qCQznK3%a<6Cg`agddO0sfWQ(uX)eJN!c z>CjiB_xCsSN69gKZ4b?pb`qL^?q>(%fwK+bS*h(?pJA}mcT&|6zgVube*b>XUk|&h zamNP7A6-SLyreIn!!$==N1r@hJ&B8pIWRU0<;tLQ&tee?iF%_;f%)E(fRvSQgDy^D zFg=xA_l4dmV^a;y7`>@aLD@QORdsQ2Sg(@_aYNpft17a9(G20W+NGo@ilx^z$TaI&YOc?1c` zK{Ylv*Ut8l)T;phVpV7FvSFP!zJ)>G?!?`)Mx~_I;6}yqJEca!skr&%qA1(+oldAf z9$GIch6Isn6r}wOnla44=t)LWC;pNUZk7i_lbF-uzg`id$6!UwE^?`_6@(%A=`{vU zxoc&2!TpzK&(|fx*7|TRFjvsl1 z_-q>em5+$g)sSw8q4dA4kO5d>fdgJK>@})M5)o!atcf=H-WM8d z`Ls{4iL1caL8F&kkhRjJ+~nscx4Wfrs_VwvxaJ}b|B~u_hxHAL%1kfn4@WKB>?U&Bcj*hA$#afU93t$|d3xkX?4$lmgGP<7+X7r!A;s0Sy0nWrP z-pTI6ZF{1g{JnF)19BUkD=7cV@Pei3sJaoJOk|iutqp^iFr|ldPqD)?!=Pa^mtyM# z%2ZmT62{yPI>%8#e_gn!_jaGw_5NeTM4*%&+PUqj(!2e?7NHC*g7Q&T#_V5<@DYr^ z=*@$OYY}oL4zRDEn}|?^L%(7gfi1E5-ye+IwLm4Z!BSjkAm5d}T<<I%_Nz+TsE3LS&|<(P;& zz$zNT)g1l5;jhTjRc$Y*vCk33eYHX*)DB0xarWhq>()CM8Y+VJ89PrdJ5?w7;^C+X z*Q9+_GL`4=0OwA&Y^XW;`AYw1=N25Hg@W+YQ*U9Q?z{-Ht#JTBD`d5(g^xCl*?)td zl4%dvGT~EdOqgoFcJ$@##p=<_`(uX;WOjudLq{7T!x=rGeD)N6K2JNDwy7ykZSgwQ zvEeXQ>~Kiic|eY|18^8FfFA8l&GD@ez=E;eakkyGE%99t&5IBe1B7hYrQfYjT_gNs2D zO_&heeZ8z~|K}+b*O-J#MeDO?zB%#583@lQxw(B&NfE1Ik#oX9YapRRnXr|{1}8ot zmy$A_+2g2GNkpUl0ifdk5*jv2mxBb*UUk=~%#=)!Pm}?bCl;s&!D8Y{{sW6yNPx~Ai9j6+zLazfv%vd`%!#o>Sid&WIrR-q!9Bg`2|vJ?omu0<3Msr z9hKx^XSSdl{l6j=x_)I;>awvC_xFs?qKsp|FJdmpr+F_BVU6DIP`4O*B^3^%$SC9} zv@(3kNYW>0Tf8qYWM64fe=^6%ivTh(>Ry08qAY~m3)j5_G-+A&y=gHc=YOGrQ6u~I zHu+tEo?jio|D(Wo7ivud5>TB$rHg*PJz~4WSB)+b4{Y-j4J7VA$G%O#W?Oh9qM<51 zXHsDv-w8HXN)ko#Dm5IV59GKp^_1wy*e~Dyl!iS;;~8~F2qxVxl@%O!{*YaIA7w(P zNtNWj!2+8LU^+-QiMfepgyEDdEFOeQJz)(OFGlSLW~y$Kyb1Mw?ScACC!p15Wl46` zf+kkV7>4P;6qum&_r6Mtzj^7oj_BM|9~qM*3$977z(FPU+)uV}9%&T{2msmg?}i8p zxjc+m6(4I>GPFG+YBi!eUam-?F-(MQaV*yc=)6ry#eP<1*j?J#2t4lAc zo}$q&s3D9tj%_zqFEClKpas~dK>qWyl%O~(ABnxU@zI}&(3?nF*Un;c4UW34T3he8 z_aZ{6MMsugNu<|ScsF(Pa?8#3dcREM>*&RdrsejY<+dy8X*>S{6ONsi>CI%7TBK)`Aez7|0KhIpCVKhK^L(` zY06FN_^hz>{X86l^hbaO56az6&Y?Gz>Gp%(ezh$mzs^`vTpXJsWS5_=E%OKnKXlHu zhw71YU}~gEVfe3FT5k5)aV`N( zi$_JJrR&>dg7|m7(lU5{u@uGfosUuJ2I9JFvY$k8T8Z%jc+=lQ@zOl-*OB6^ob`Igw-e(5&=ev^(?SWB|E%#qSR`&1uL8+#cjcP-tdI z8&``h1Cbw;A_=3ce?4iqBXj%{`63b!BkdVRqyG`W$SdH#&lB z&3PV?4HP(@%NkUe@kY?_#tK+{{(>mv3q;=WoeMrEDLMM>v8nTmCrRsu*!0xFS1&>( zg5}UB@(o1uZBd>LL~e3_9iP?Ez@YQ8EdM$_O`sTtdfxYU408{PVUC6eZ~PSzez@uowt($%v%CLX1{gEmb^C1wt2^1Q#z@{b`NG{mnJQP~Mxi?Ns9gn`2~ zkDG#T^x4QmXi~6T!Xq%KduE7}FA#_k9+4Ej`&6-v{#Xesm=DCso~lX&alzsjuaRJ1 zOg~SMMqxv9;1|s{XwW*lw~KJNA0j_QynpgZX=p~A6vX+o#M`tW8#t{H@OmD{B2G=5 z36g_JJ9FQMSCYONVN?GfzZUuX>VM6TKl`^Vv2e$C*~E;7MOu?B9zBpt^PvEam zn-_%Iu9n>@e}&r6hhCDQ{u65dYyE*iBpk9pKU+^eYtBvB5Tq3sS)+Y6TOf z2UD}q4`OTTyURP^$D?BtEUbTic%=OGnB~3&nMP5X0eHHMBPT734jwdf=admd|Cnz& zjX04Y0;Ok)>_~Fe!@fo;f7isDjhS+{p~UzdGD+|OHXG9f5ndXhjpsyDdJos^l4Ijn zK0cuRg71H7mYwXHUK@AkJuOiKv?MWx$wy#jQIjCZ!H=j9{IYegpsDr$|E=Y}uKjg# z{71fKj-Q$>VBRyJ@2xJJv2vpW6qTS;;XLP+|Ex{?+Da;^2?=l}Z2OE`ar9HFYV{~^ z4rmCMnr^Owd%Ae7TNNQVJ=vxa8k>b^YuDlUa_<6h!bKKL2#8*GvGx*@er=jGgwbHE zNQ$(-{#xX7C0N)x=003ke&v|NAXBT3K14=BAu>@b$khJFW#BlFC�Fp+sK(of7Xb zAKY&1L8JTlKZZW~TJT}M(JQs?*5pJZ>7(9xB}5d-lcW6;_ao>`F{;2jG^|1MlO*dN zZ#d+`?2FGInCdyn94jHNfT$qxLrJI1--p=mf62Gwg#?ei$vr(^<8m~bY+-_jEE=is z<)jUEqnuj{aq|^vp#5uq+5(`Wy+?0Cp+`v)OaS?c1p)p!Y4GqeSJt6Km{3i+La79(I`kipD&Ih; zWc9OoR_%JsndDAjoc?nVy}g&<4toHV7~ry8gB6v$&6I?--LwdWtii7x<^4NHnaLlnhih(TOU#o(cNB*9!nHtQUJz#&%6+ z4-HagdF#UdJPK6>c#GcyToB#$O&*PoVcug)d>9Z;8(AZQj{GdjFo>UN)}`Hjtz z>0_?)&6@m#3RU<{qcUZQImggvQp)ofp)wP5RNffrE{WFNlRj8_KV{84bcQ&C(a5lCf+D1;tqv|^kNZ0ZXM(TQh zz%?)LHVCom%W(R%a?r;#j_P9ir7v2Tt~5w9ItiH5ILCB;wi&M9re2ZzNpbm&UdLIR zSeY@W$xN%^$NNiKF_BEkg0j({waseB$2FvrF{Sj+T|g0*3p?;fBC>d#cc#tS!| zqix%NDt)p$LV2EGVM*@Hr?Q9vzL%)r2j4PCA{!+wzkYdo;bvJ#)N>a%dkwHdZt2^<>09gH!Y8gO8e= z;dT9*0>0c@>!bD5QB=>^Cvrcssvp$$jxT)3=PfJPXa*cW{lGZ}1j|~BeXluy$elD~+AoPyq86eaq0FQy& zvd3GBL!TTy;lKkWMbIXk;j=#wiu3TDm9)Me7T9*UT2kr#^Fh;GA79h@Rp5IP4V;l~ z11?oPSp|G0FKHeq2fST^q}MJBJa}>pblEbG0Z=CnVDyYZl6w`fBCAZjmjT)%0Vohu zeW=`h!snK|qqWL(Sv`p_L5Gvo{P;WV!aH0zd36h=RLhq%NP~qFsH;S$rlwkNZ2-3B zh4g{#6U?l(S^2^m)2&pf%wWv(wcn$KQCkyIfOA=b`t&knbi$p!MHAoxE~^etY2@~> zI)604^U(<5q0#5@^DNiI9YS7;~Bpz$5^5Jr!=ACdTr#NqB3LawgvSuYW>V)37ZgAZXmjGKx!#r$g4P03+ zo?<>#fiV>DzqJjBRC&AG`I*l~0PtW{p5WB~^Y-Y$2#>$8pXcP>lsRL{TN zaCY|FeEdQz6%)^c-#dsY9ujia)x+95$zvA1dZ$T!;Z2vr_#=7NE5!5ZEL5GQ=>Qf7Z@KT@C!!04T$;&GPW%rR zl6s!fnk7q*rAMA)^p@vqT)t}j_JQ)Y;7JQ4+V^EZ?!z-ubw)c}>^2Z`AM$D({Fz?W zINH)bpV(Unz@9#2XkFhm*au!r&bUD*hkNQx1$^m>`>{v>Y=5F9bF(po0HgFw-4Ih`aqrp zF!@qp>pLn;0jy^`a%WbC59I)ORAx!ntGXj>kxAP7kswxauvS5&R-h@V8B?m3Zz5XyY3dt2p>nUF{w?4LZye=iU{|alTu8!yYnhHn$F^MIH}b- zv}HrOCy}?RB88xKPh=AGhNcSwCdTG=xC+u9gboVX^~aZ#H*%wDw(`CU^6C4*SMR57 zP!)fjIq#Ke`Rp43>%)*(-jWQlpSKEVZEikzEf+o4Inco0_-g_l)Ltns6`(nHc>9UjGH77HXcBD(!^|y^t ztitxOM9XMxMynhTd^&^WqqhNH{Fv{vDaxIh5{6Ol#nvhO~kMM|b?5y4Tu{ur^t}aQtgUg zLmBtJ|JK^WD7a(v4$!}Sd(yTCgg`3I)mrz)Odt^0#DK)%!g9?x+-$0Bdrv6%)C{?M zFaS4$1agm>jzC2;9uk~DtG|be3G5olgJo`AFXgqtq(ke!TKgRlA#(An$LAoArc+KA z(vJ>N?}0Or5m2JE#&v{@7M-W-?=ZrTA4{OBN9IAPp$dLUTYTVx9uAfC9fDVJ; zb{@>za{EnPnN6+9s_&4!Ad0l8gF`#xI+&Ez0Bd?@m;4g1;pUHnr9P&muNuHIyH3V( z)^Ahh;Ar;jVa1qPp09SQB z<_*!Zhuy=zCzr>{-TfQ&J`Fl_%@r_`lh1VR)|$97uEkD#&kuOv-8)Y*Vf8bAugksh z!n&B?<>Tt3{1$)+m}J@Acu6NtN|v%J`_Yq#czH86UxY-OcT?=4EB!Gt^e|&x?tsu? z5Rt=kyLiVHE#?k)^5j_A@=Di6rmw$Ob7|L>G0P(1OUWpR(FADp_4`>?Q@DseE(PT3 zSF zeWEH|FI^T6v})V{D}O?I(oY|rDW%tbysRQuX>QR92eJ~`ZBlzZY7l-A)UL^GR)1$; z!P%69mLu*<{Wa~{=_Pl<{7r!rOBealq|wQ{R#5Qi1X>czKoRTADb#K9wO$g@NyTKCzX|YAp^_a@mW@;oO$-t zSaQ}z=HyZ;3Zvckz#y1ufhZ{TIs6_YS$u8KE54yzy^i|OOka&Q&Oo5BgxYjw7~KYD zk_zNrDIQ8HR#rJL;Xu!oKs$Y|z)alY9o9j`*&+Xt-#x$NRq32V(fSxl!OaMRM>(yMX^x_cJ=2<}Ve_-KHk8ELsohZq%ho4-93_nzSx z`9l)#{#3Sm2+h@%=|V}?`*F3JemVIBa(ik0NyWk1nv+)gOIseZB{JU*v#=e~FCxJ; zBa@0_GI!rW3SNc_o_c;nnP9W3O_yXf_gjf|yB2+$NQ0oaIkSk1;vxC2)Y33cYX1v3 zS;kkzVG(u5=+TDj(U!@+?@RYLa(CT|)AcfzkZ!xOQhu3!e6|Jl_`=c2sqarzav3WX zR59(#qNtVWmfV(CJYkwY?<1=ZH3tH((!?y*pa6qURjRIl$8fOMvZO`22DALmG%5~> zcDzi7fE9oosNQRW?=kjCTgL%IQ4XA6kc~9NahTs8_VJde>e8L@0ky*>Be{-TcG8!Q z!vhZ{O+dj@7HR`1rbzzT_6zYK{k$o({oqbGPx9zM-lMgiy&kYscDU(?g94%|Hg(bJKGjx|Lew9*_bdgt44O@}wBJz|vYVaS! zgpqw1XE}l2kg?LmU1=yS;ZMke#~0vJ_2KXrsrt32MCUC%T7_lAUG}UNOP6B1{~_*c zhtQtC`#F=5Zn$do0RDu7L_UYZDuusxpr_hR=Ekx;bpbCc#Zg6w2hu_q!#HS`^V8n_ zUTZm-e&jY^#^}~h5>I7}(GPY9vpS>N>-9ZTiq*Es`cWAZci*C$Pme}|HWupwH!nL@ zn3H*U)dw0-#}-)t3#i&xMnk#o9=2+Y7BTFmEv5zg@Gf=-V@UMRW7qEOCy5HCr210p@|uCNOmm!(P_MvK8k0d#rAv!7to)*SJd-^g z*;YcW0pqNVtFE(EQN@i<&&yJr3frx%u4b*01s>x%!1S?TVP=%*E!* zsTl$*SJE~RHq9rqDxZ(J&cW^+b~dNnu@c*e>set zDdQ)*M`vTdYgxpJ!#n+@=?#bB0SifDC6yU_xBQE%4nFq*JQ1(CZM4aq4es-AgZt=N zxzmW0Nqk|Bi93$FuUS~4S3-4CA%QB4QRn1?g5Q|+DjS4s_PQ4@gl8tS;%}@qj31jO z9C+*vIBD*NOG0Mxp`l+zriMhs3i(>&xmK90()UVZq!(XI(eW0S#*})K+R5Y zV~?$rGV_jlbS#%x)6ng;BbSwIuMuD4zzz4zDtCK3y*Qz4rGnU@nDY#53@<8yQhsUJ zyMjb;7H)pwdSK@L4S$IL@t#+`)=`dLXEa4;lDy{`P~H|n0_p0%(v|(|2pREo;fK?3 zaXX__n}u~>2_<7|lktAxElN##w#xn{HSbDkLy0ac z=lFQ}t9eV0SK|r$`>e@=^6|9OdrhUq%1;ZNO?RWaMjy}f-m6m+ZSjmapY|ADnesy@ zAS)Nit!tPxzC%>?ZX9q&;_0gHodl_#CHfK_x!{n-kma{*DcFyOm5qGxuk=qbwft)O zG!VOR=e6CA)Nsm(o{FYNqWORmmcCraE^Eu_;o3HEso>xrp>{alAY(Eo!$o(syEvvni`s&MRPtpJU> zc;NZE+0T_BL}1e8Tim_bw_e=NNjc|n$Bk^Tb1_PTV%OmE;lZYd#1qpcPE^}S3}q=D zb4+AU9^7GXtu{od`p)Vs{O0#adYie29My*12D*s+O5j_4I0e6$4lFmG29cSJlW$E%xGr=jnAWNS{#=;eX7z|x zmlMRqoA*?qZ`_aKi_K9D_1UPDY|NMHfc*@DpOlEqZU5|G@}k2+6^hw-t*)t>iHv-6 zO+`8%?yjWM{P1KDc%W3^BVnA6!`05}m&AY6$AX`~!K{Cd;8(8xb;Yh*HJ|86?Gt>E!Dg)q!$B?*xLv6qmClx+$iWfyc8Hte5skG_dk5%l zb!JRHoy-^hVmVCJ0EW`RaTaBq7n_we3U)R zy6rN2KBLhi8Pt>XH~oY>gbNDvG8P{%uJ`5_AlyxB2PQ{8TsFvup#Xi3!K~4FqLOUj zz~?=K@R+>4A@*S1Y@pP{6}jHnrc(?rMG|n54CU*Zkmu!}~nTm+Hys?2-+)Hs}(Z^YILvg~y zKn5rWrB)&$PiJ)m{KewnPXO0RR55HVw!-2ap3VnFFA-FvDk~dS@duPDsB7Ie`N(NG zllpe|Y1o4m>_H38CNu9>IwPxMnut~zii(bN8i(-MmA&zqBV#`|vSdk>+T?`kbfmLy z(V ztSPbdT#ID~j3p(OCjNt22(8xoMwj!?$XC4s!VfmM2#B`^a`2fz!U3ci9CF4OPzR|1 z>4Vj)SgFFx^%m_oO@%S#IOP#@9`U~z;6=X};OQli;H7E>p6+B$fp$dO*;Kz~io^12 z0!P<|g+xZfHAZxPi(OM3$Tc&qJx{Dume@%Mrp|PYs%PIieS$3^%EvNjut>+n3z|+! zcSH?d9;exh3o;go$@MyJ*}}X?uRCCgiY!wJM|`kWW4U-^G_00?sB*L)yST*ODco6j zC&D+@_xScGpn72&>CTmrbO=uW zip`l>s78}3W!RHP<6r~J-s5$(c6Rt5BIQ9ykUkBL7M8y2ttF}fDy-2Jn6P}Y!z!`} zE%BO3Q&SUNrKhR6cs2`gRJIXAMb^TuwI6;~6}^@ww5IGl?h?||diA?c0ms6zxw@6Z zbaK0zqmZ4nv%yf;Yc`vVXJoE`om5zIlLa%cWiB?j;TCq%s5b3xU;NUPv*G$w84@lD)VF0K)QCOSX0){J!m%@c^& zYX(bMTW#h@N8chfAQ#|yREH7NR;u+$y?8@G-)lf-K-E+27V`*>;@iVL2UOIk-`UxD z(n|@61tqvYX*S!Q>g>(EOtM}jWr(WncYpm{nXPe8&*604JR2?}3G;4t=c~Da>RMIh zYAKgl`ohJ9olYBoEF8nk#c9i?yhQ=Oq`Hnh_uk$VA)!T8nL=|j$GNI1ev)*D=fZ?N zn}_-e=#zf~r-Kkj{tn4<^m>E)(xm9>4R2Cr$P}s46LoLOM8R((9i)YxA%T6lJNl+@ zAZOU+Ckeb!;#y>3Mh&M5j|OBgUnG=kymAQ+_!{OZhu&pYJ$eUW4L2m-#6I0~Q}pwj z0m%2Rr$mr8YiC{CiHj6}Az6z#FeLHEG|eVJRw{VLl_pdydU0uT(Yw={VqdMpf}EJW zSBA$D?1fB!$mRzd7tj_>ruoCSyS4$4aDt1l8q)y+Q$BF`6i9RyWMpz+J(}bLpN2tm3jHn3Jp*>y;jMV2^FW4iUF^L?AKaaY$|qij-0CyPnR4s zX-{U%=Y$H$B=*cbFC*c+jv_Q6TS8P&pzNT>W(&r)xyWZ{apDIs!0{g2NnXrP@bN80VdWh%09uFv_z(bnSgHY~7}o$_t{&{_wrJLsgsB zWVxCp?fT_l3%F~v%F_JH@w#(SyR(L+{qz>Y_|K+Wc=v2C+FlB%hCh6|loiogujjFI z&a=NinaJXE-*M*KzL^8Ji}~aColENY(rJ+`Tajpr$W$!HUOe3&>_(1ETrjJk;2^(?_@h zJ`h*%U=VMhmCplN4__r65cpaorl6@Us3$G1xUeUIgy$}K{7@P2n=Ij=mkaFjXPHBF z(PJb9c<0D|iG&w-nDCy@IrI{|-k&<|;tyx182v!u_f7FhTc7Js=4b|P`$XEQ= z4xd&DJu%x^-n|LriEFRluLn*|s%0alWK)Fmx&TkeCj#dp`Q7qXyCsjCDtWf_fn4tT z)6P*xcQK z%(8gW0lKvP3$3~TombM^oz@ImCeOka*Ote3 zrW&#hH4nM}_%^va<00v4=W%j#;2Ml?Ku7J3$V|1_y|LDZz5(T;!6Q+*HhUy92e&oL zzO_sFtk%j%=owUuXCQ_&H&sgmBbCN@Z@}A!FfHTB%*A9?90_4TcK9%Nbpav6#r$!& zOGsQrPi-DNftO_S=X43ZVz~jN zfcaGu*_cGq9_yZvCaA+K?Q@tiI`wsgS3l|}Un;HBG&8;V=~EQZWc8v;kx)q++-XzC zYZ&9S8qhd^SwPu~Ep_xiy^DOgyX-PNAVl{BIiPUpqpVV6AEdjaK0cQhIFUm|*ZW#xC}ORF3VQKyC==o&!A^(khLx}RBj|3V3EHUDE#?ldiPU32XQQ+J@sIVr zYqzLu?MwPg@)T!br@APcq820fl+@R+U0gL!a#a2LJgX6R>|m=ZgP=2Jvhzs%I=HQt z6Wk&W8<>H%M26qbJ$P4J`hKWeb4|5hc*?i$bj!W$_#V_4_}e44wwI5}Ux>Mimek2@ zu1_v?KR)M;6GcbiVs_ws)V-N8y&}bO@woI-qx$5F&S%P$Q{PcCsfk?@7K`|0ban81 z%smpOS_3DaHR8I1zh#nb!^5<0Jl54&q~j%KaOGK?b_M%eMvnsuQW0-SHH?b@N0~ zi7Akc;oj^NBmLxv(vs4im6?9q#%s47egscZPoWIkcFIau;%(#SpGL3HA72rX;@w~A zh7Yggqw`BHHK@?Ek0G9fX(w&HLkT6)GB#Fyd@Y*4r=i(*rI3+3;TlG`eP9)-x1RJS z=(J7Ld%$&KCW&`VYK4J`Pz}Y`VS7H%*6--dVO?zs|qd@@v0xdDADo#cuux$KSfgNqBW_rrtjwN zB(5pqB7WmVJLkLNP&f8oJ-nsz-XoDrdQ1LjVJlU{$vLI*Ml70uLHx}Q{=6hloIByJ zKG@f^7wZ3o={%@6b7u zCNt*7kZz{>iuy#{7$N}t&17Fm6KmBlTEkZd%`5nhi&UdNN_wHS?1OZxa$3SVtBm36 zWiMNFX$Bb+LovtXMPiCG^l~7NdVJk$`0ngEK+n}Dn-~hHSD8BTk{;asoW?yha~5IT zcapv<*c*TTgdM(q4L9F^0#8op+?3*^&@Kt>RD*0C?`!@Sw?pf$=-nFaG|lrI|B;(B zy4XVdrU0kES~Ac9PnTx8+eW65yGhn9#iRPU>}^6SXgg=wEBX)iN+T7a0h#cJPnar_ zYE};s?<&qa@2-&O?`{eXBCvZ_U#NH)!^3O|4agnJI`D$O$GF&1cF^c)yR3a!aT`@u z4Yn|Pub8i9KgfDfW73v!FMsY0X<9ztAX52G8s8mk(ud^9Zt(#h>BLfeB;-V~J5TeM zw)V+zV>V0}DwifUzG~G0DWLVOeeaLgk)v<=Wp__#HsreTrc|t{?ZGZABDKzc0RCDy4n_vE5hv=JFi(f)VZHhaUC zzG7X$1$}v35lR2)O{&7em}s9}b_ZsCzsPXe+W`j5NZ04fA2xGkq(g0CXe9ObCcO&u zyW?i@W%rlTtK7m1*8sCOriEGf z+&}(UO&uXYH{cg)_4D|g=|A9$dJuf^s`aLnMc^}>!;v;BXP&mJN|~Z(@mvkl5K423 zr~BleD9vCzobPy?CyqfE&GsbvR}FW#XaRt73#Op(rHJYm)O4E#5KNi6k-@*%SolC< z>>h$G78u)qUiOPV3cqkW{>0h+-7_m~s@8S`IwVI2xEXYnlybRm!}oT4)f z;70v9_0IiSZvT3+_%iwH8J$NhpGb7dCBR? z-@As8?J#bdy-_ZYc#gl{hpD)=X(W+WpGO> z7@(G$UP|;*M_$C~W8qLvXaG+j2oG+M`N!_CWynlLxNicGp`iyAf{HxiPvjj4pQz`g zR409l!M=EvQ!7}q z()(ZrMy%~2M8MF~5-Ng`m-v{=35`4nbR`br)1L1j9rQCXMQ0W?ybpA#lCa}|88V<_ zFxBXs!&v~!tXkSkWdr9Q(}|dM@Jhr&%c4;lq5aHI(Hm?0_iR&cbiilxzgHJnOm{hv zli~96IE&xx1Vz4l|G)xY`?cANq{hR!>3e=9!k?7-=1yIar5P)7k`9+;3l!a;Ax7I_^I^L7lr>W3;(?` zfxl+Aj+1s*P{Y)l(=CCglilQXqZ3ob!kq^|way49IOqxtc}Eb>>hpP@q^x_`R8n=t z2YVH8UH}X&Z?Of+YDo}Ov$&*W0a$n^LU2*YnA`{GPb>h=<*Gr0^({mJm^psj9WV&Y zZWSIxKzX9kaNYF3sG2MPo2r@aa^0j>143g?T%SEV#h~^!vUY)*S6CWs%(zyT7X$Fu zj&`3OlLe;{?miR@OA`*h31%;q#x^fPaaIHcIuscIQzUgTAGZLISfiQ^01}3B7nK9{ z-gkg$dZl~7>Kk9LExigkAP2Podc$v&WsbK#Oi-yZ;`sssOZE5@s-(BN*i;gULH#U} z0OCHiToHhBx=df8prZ+y&n#b6oydiQI<4pwZR-|3QJ4_9G-UJqOIMQP_0_JVdJ}9W z3GcHLj13Ult{nmqhyCZzOm3Yl=z8?>{a;MaqI0n2=JNf(ACY4D8EP99MsMFg)K0#^ z;`5R;Dc~+JEK%ls6Jo})R0RgcMkHfC8HFcIv>&l{Ww z!)+X$OEhXWIq-|~IT*&QFI2SdHec}=RLICCAZNTzFZM9rIS_6r?(R#*AXz}!UL z*xTF7X!Nag2khX91Esx5$b&K4ChmV80TPtS4Qomgr-l68lY5gnub#;NBx&9`<(`<< zIY#9^C9KEi4iUgSOq%*;fdHhU?i&wwFPR_*EkD?N-u){5sXtYuWW)lVFI7dRH^A;t zp{bdr!})ss^3KN}j~lk$lf>f; zkD~tkib2q+0tWb-h0LfszxbOBrX?LP>^wa!31vp#WtP{d;*xSo(x6z1&H{PNsq#{l z4H8nzpyHp40DQ|xrK^>&HG`tEvOXhiR~*FJctui9lNM*z^A@|Kb8 zYEpR`HiHyI{?r^vm=MC3hh^xHMmweg5AlG0bO)#^a)vQ@CWA%Un-~CQbPw2cGEfV^ zptb;~JuUGT#MYxyQ0Bkk{N=`F5yyC!rC&wmmS06>K+S|GAD4HmCHFxqZwfzEcwYj* z34DML7*U{Y$akX|5IihKriO#yh)U*yI6yBsTWx@<-^<8U;BqHZE+fXn!}F^u9l@Ud z=si)k!XUsy%@D`|9mtr}$N;=;NbI56P4DtAE<5}R4zy9-2LL??sNDr*^LYVi+^5p= z4=>a6ELNRQmC;Ld0|2D}={gIV;^ocTYD;gXA9pZpck@(y8C|YuSfq2Xw~zi5LCZ1F zB*x9nZM{9;wV~+zKd1X&Z~ib`UK$>ACe^2$E_vPzHxty!-kPSIKv+{CQ>RXI-G~{2 zP_`8PMc)MMO?s%vqqMU|;l+VP7uE5r58+e8u?$SJQa#g;g3OLUg{GeTZRCGfZT~(v zmU(_fo^^C$*q`s`lEC?!2hs0ANpCVok0}*w9S=jZ0mQJckzp}^ElCICT&(0R$P{AcuJDqdWF^;`-q7uujLWBEUikrF8mE5OC<7~|^ z%CXVTcd<_crahZ-Fnz^6wWz$j5i_<#hP#t)0wbbeeiMX;haj?5{#UG_(8Y<8-gScWU~JSL-9;6GJmEv53C79sx+2esPF delta 71654 zcmb5V1yof1`Zf$BC?TPgG?F3$3eu^FG)Tt~DmiowHNXZ@5kYFCYoxnFVnF2~q=p=% zRl0NNfp2rpdH?VCuCvw`i?wFWu=oDeo%_1(dwc9g!`O{VL3iNu;(S-u%sP+ns$A^b zSfcnB5K=A0;+-3-WSk#DP#;Jj^;sO85To3ZkT>`4eH?xiqId1-3nYPBh$2bsZ28e4 zkk!~X+qbnbvo$*a{vC~37qo86WKE<^%bz9wu`srP8z!Z6%-qkZudjC+NSBJU-rL`A z4nFC%3P`Nj-UM1&WY>Su=*SN3-x24GEhfPHGFX&!;~?_uqTILNdLZazF{x~SWB0Z-N{zeL` z!(c@MZmh*m=B5r~Zp`dg%U4t2mY zzdV;@2G2a@+e_G551ccE05Sxb)S9WCt!+9_{r;!f(a|xRY84*TnQaIlc5~b9ySgxT7wISA|7TvQv!*2sw8><#}rGw=`hZB6mV>7{+XeKeB zxHrmU%C?hCOIk-;ks}k7zn_RR#t1yPi2(b;q$C22i7;LLU2p-r#14eg-H=&xAqdF< ziMdAyo#$rVxET#IOycO7Jz6}*prj*%QTS2(M73>isgeIkflk97zmQP$hBaZRH_1vY z84aKI=lAcgJ9%=s;Nl!atlvE61$KtL`hKB|G;}j|RGYu+q(0^!DwU)ReM{`u$*b8B zYRUnMlH3(ysHzU0ik>-36aj>Ugz&{I5edgb2jYC~jQi5irQowodpN5$(d(&t7tf^> zd$XBnjLde+GwnZjh=w`cx6wz>-|PFW6%{OWbH!-hn`^tkRe1J`jhoB^0rd3ryxx>I z?#5DrdtQCP9ivGK%Hi|l#as<=#wPBcH~yvjewF6;#v{PVC5dYnr@GhyXP03$tJ`YQ z(SKX~_e)8wAER>Dxrs!0Dg$t9s=5)a_S-G{+9h=>X>P6jzB6#+g+O}h9cYMs>D-Tn zEZ3{K=y?22}|MO+FvFGK87z zh2;WJj0*6dph|ffh7dTE&zhE#6A}f`-2?df<4kMpOuJT&vj!5`(HVZLc}dTG)-`vp z_JS&l**y3iMl~6UZgo78%;2#^QXVNj<U=w1kw(!VhkUz%<}Zra?^Twr1QleD&U^Pw7ZhcjMI0i=xjN*!f_90rEgTBZb9qN z?3qb1*v@`oPINqIpATJinI&95tFQw>qmvc$q4CuJYgb0npcBhiijx)L0T-Y@{5gyO z(#WYu@Yr~Zdpwi?!cS`M_vfAGtpBpcKkk&)RV8g$OpN3)3D8jL)%e(QBWLE&ry(~~F&&24Eh zugxPN4b{BzL$#8Nxc~Tm-~Sa1I71X)5IH_0NWe^3e03HT<|yum5$p$CJ(7X0YJeSZ7#j=W50=TNft%L`S2 zQg%mC3OA~6yE6D?4^)93!G>v0(2Q$}am{mh?E(3-4NarCYU&RpJ5irrR8mue;vj6) zH5dU^ua}vSkce7U6V&L}+AQw~AYqD%*HEWg0oK;m({i1|2s^SeGa>WX1m z1zEvOvPaWi9+4~}HXo+FN3;MUg?X|nF65O$7ipQ|>3%a#Sv?uR)L!XWq&^#jgh-2H z`vxy@+3+YB1gUqFUdG5B#?oj@es>frU=K_WeX+S!J~JnlY@?IDpu*Zs6Z(v}7J8Oq zp9w>=mlPL2o2&6Dh|bRD9D1aX6}YD+wGgXxyFY%OddCp-1rk>qXVD3c*x)PQBEh5bFW0iJv9${BDa}&Ha~=A-Zrl=?r;*gtxST2A@9!U! z^^N=%$;}YZhzJ!6RFR>QH0&!7>D{G`iJGzqiMksWYHDhO!V#y-g#|mLFRNoEaJ7bwS_n2LY2IefuL`G+ju+hI^S6>DK&u>@{t%lccXwi*3|xI z1%P)N`dL4qa2a^AZgrX3%+Q;#_!(SmdQ*mZ(vS@y4cG4&MK706axB?DOWsZKx32Y!{1Dui;@O<(6DsIC@@)v(6c&IfBfd)0c{ct=<6aoY+`1@}H z+Y5a-_q&aPgINCR;#L#8t(F`EqtAY|t}@8`T9-s25-gAbUS83siL))r~oZAzN&?XSE|7L^e15(aGAEV{aN5!niH(R2Uj2fv>yo}(=2FY)UV8man0=DqL$!Yh zP482S%!_;5!e$CVrzc5t-&e}i3_8UMHF?H?AXe|vj{76JWbZ~?#0b%aY?irjo-5Pl z{csc~f4dLsitRrN2b%7NDDDtHH=ym!hH7gcKXq-w`+EL-x6B*ZVs(R_j!sQqANlra z%E@!zp={-9Fnuve8u|6>9uE&sum#fz=b$j*h36UrJ0Dl9RA_yDbRXSwHIf5;4%%$E ziUO$=D`*&s1Xmv~35cboHIwqW^_-ka;NFEO&At2iydI8rd*MFf#%lKMOpNl5Z_am$ zx0^(gMeVga-(;1)F@3_(OH2}CcS4aJK=KQ6O2I}^??PdD>)%8EYg~fR4H}ZgB-P+W zS}Or;oUsG3qQLv;&*Z`_gX7iRETQJNP!ZbY(`x0{kc5&9H$ZjkjIAU@65gOE^HjbH zc>f4EG81`PfunBCS(j7)*X@C{D=RAp zk9@aFWn`LLd@Jb)iCz=!S=WQ>(QTy_d=>gCVR0&kBI512tCL;N!=l%!tA`>twrnrX zPlm^oIedO^5%BQc2NhJ8KRT((j*y1!l$zvS)3wh*EQzPCZTmWCyYvV}EZH%>?iton zlJ4prI&y;0X$B60)U%^x0pz1lI`Wr;WYh$#wGEl%D#$OywB#H|;1Mrf$SslGLRQ_N zHM>)Ll@M>1E&upG&iQIL9EFgKjJ{U3QG^OwR2W>y{?g|UVCKUCCV~I^LyEi7b-q5z zN=wT^sk1j%k?76+SoYD6`fgWl#F&}rd32$>0oQ`bp<7jL?yt7KpNyd_G~f!YB|i5) z`JbcxkBme6$Y|+ly&3M1sohIocKY+>g|Gc=_K!{5Q!}O%u2^P>cPQr5S+^llHv;kkCZnAgLqtZ5{6(H}w1a0NdQ;ou zfA8`?-(sN@p> zXfjFXIwMkAI6@Syz#-3Wzg9pJKBi$sDc)OfBY4a`a}r*E<0(xkefZsng~aFhGM3R= zb0pzUns{*{vYWM_iwkYN_BjR_DoRh<3c5wyr^Ki$vC)yeM1f+YJYYoyCVe*~q$l0c z3Lz+TtF?)`y4UiWpdx`cYMWq@>;DiHxCuoRak1l?sO9D4X?|3aS||z8RUY_an+5m$ zzhqHL-HpNY`pND6c3`^qFq^r%5SfG`zY zNs}x@@gaEg^?7++WLz%|vXYpHMD+={HZq5FWGI>jxS6Y{V!y0NuQNeHqz?Wc<;swt z{9kK!rZH*$a(~{ixXh`?FNO94n5pcAnR%>z`oF8}zlCWN_U(A6z22CM`?iQeaQ@~A z6~Dk?4ej1@jskEYOQV|$um9uCu4WPVw}x_zK>oP<4(eMY|5UbRxF}6sLlG1Er?hJJ zC~;ImE-s9&C%*fb^vZ^io=j2bN+aJjcZ>ade<1~+w(8J}b0T*U^rZh!7cKwi;{R!C z;JUwskUkb`uxr`KU#kCqmtUci!%sgsGqWjCOI~75YJ<>olu>#B@>U0p?gYoffM*?7 zy86`r!iax#Cb6tKdvwmTn>Cc>HYgS)38cn5=4SL+0yNhmsvv_DQk?9;)3yYCIT+=}G^`!jO-|X*2-h zrqeXSWv+)Gz>uYBw%;a5ZgLU^L$43@FX{cam%~HyI^NBs zynlL|?2Orw^xb}Gwfct02q3laL2-a|5b}80m7f6_BFb>19uy$-jq$%Mj;vb>)MY7O zjYf`t{U)gd4RZ2!%fF>N02*t5BeaJY+&Vqd<1Mr56I4x&z`L4LUv(%#f^R4`Bqby*`=aWS z$n6*)gfzb!cwS08NLXVFKK>)(vtg1V>G8U6b$_Y9=CD8c5tm&S`b$oZ61LRTx}wRF zhCyl${xrIRCFo#iiM=^?8booVg7eJTkgrB)l>TGl_~J!u8Ps;Z^AlFb`#H4UxvJi6 zvX(y}sjk)i&(jnc-*VaWo%YPLGwHegT$UNXv(1LyH^k(g2Wwv+TtyDVl)}+hHZnN{(_x+rGOW&q@6rbxis?9&%=(9*Z+;W2cJisop$-&Qvu32ux zvEz-`<^a=4U`<>dy7_$HQaJX_$11%SQkM4%0KJd%>B}jM%-8q^&2Vv}z|O~7SEyhU zi_ras*$QPYpZL2n?2fZkMVBciM;98y)j0fT&7Ok5ALp%_UKvx>@PtwHWwsO4O@J={ z;$6GX$|m(8v8G~HRo%9{fMD93*ml|2493JO2FC~m#5-iIRq`*QHXX_nTq4;u)wnU| zvp2KOCX&6H`xxda*jw%7nsN9kGhEd|5=GLFFUVZKzMDnK!O15GF5x9v@r7u`-g>IM z;IXIc_({ja_70`eqZ0YK`}yp%ZT1_8Q{N|!m|u5gSd(|!wkejj-+sbeig@a)cVgl^ z!T}kedH@3Y3Tp1q(aUOzVM9G*qTzdkV`Ds4XOrU##%-qwE7kcaqV_$vR9NmdOnYiD zX)sNj!jnU6rV=HdH3gVqec<&P#)vR)@lw&yp*lTDyUe$Hk^1k~MErdX+W0mNx07X? zw)5<;V-t2gu*Zv~35ThMMpA#Wzj*Yj8OsA68YWp@Y&7c0I}|GZ)Iu^&8Z!Ct9~QS^pz3*dz`z?I~fhg z6pb%^>dkwFYePzR$$$0BB@KQ~tjIhrlyG{K*+8|rUuQGT0y>5J#LFR+`%yE;e+8>B zzqGcKm&at5dJF7r7FvW=JEGe4nhLe|H%g!<&g#ocNuQ!fLwxGRQ%`5_YpwUqJ&GPR zs?k1B0ovB)tnTGd9#@lo&~|6}{<<6>8ATsuJNx6de%Y6v{HS_`D*;C}w(a8+);?jJ zsts=Z$u-tSp3g%|FxGZMyzO+#ype~oF}kTUG3xXDgWNCtXj|;)Xm>o1hRx<=orwJa zONy9N;@ft7n)GvodMgN0RC}-L@!!86KT%`fJ6fclk@Q^z!g%Y>XTlfwyjEbl^=x4q zxNN=xZHr4vY_5e6;XD`K%n|pnVViHroIKcotq8M4O%ZsD7#m$Df`ZqA?-9bkC z^t*^{wC{Aw1}g}uWM+lV>|e$IeTXJXu_Ud>I~SZQRcH8KE1-36HSAITg7xo zF3z@v8O+^7Y3X^fW*7_gzfPwF<7aMsWtaUDa}R*ik*p0ZtmaM}9eq5#!#HZ8ei5BH zKBmVb=wua}O?o4R!$Ga0t8;3i-ao=Sl|CPxf5;27Xm~;Pd=q9C767P%YLCW-Wta*K z4w(adPHfFjCzP^>rWb1@IFTK8izzzhF|z58oUaO~cV150vMPkX1aaJ9_ue;PSo%2{W9 z)stpI0G+P>T3J<7HpiLQ&Z5Tb*H7((oy~x;iftUdL&;#BjwKIGOtU3fj|wi}flL_B z3p{qddJ_P)>V^fOv&gy_tGhJO=uTPd@S}hY#m(vSwe8#yq9ICwF9_&|u}inBX~Ko0Rt%zmHK6wmSdIu-HyDa~bWVqX&F?<#6-; zTg7~$>lOqD5qz1IkVvKjZ(fQ@5Nf}yfjD)sAxh~89`Hv{6g#}!+3OWficz@u<=@nD zFsy;aT(Kuv7Y7B4AZ8m|W!=>jJw!+0ez7FcR#opLA23a4Z2m{u6^z72mP|Gy;e$L$ z=6@%*&4&mUaN}0&u!ut7Y@^BHh4!^~Xr7D~d)(TXi1*5%DuKbZ3?J4_FSxiLe(0bP z6C|{U+LNA>SFwU0o`CR3+8&4z4jHPBYAN}1ep;&qc{%Wk@E#g(cVT}N$~K~ld%t_`OG+b=PtHD%vxyS*E-^^viw zhqJb3d?`yXxo50`!}ahK*q4bT24XorxG?;&>qm;<_vBhfxv%om1NKJ-y*|SDi=PZs zuF|vLzqdKznmC*gBcN{t;i<#{Km=*<>ho_uE8RM4P#1$n*TTYU%c37nnI*h~rj z*v(VHx@0auvxjfn$?TGHTGVGM{)Y9NtS{%BBbz}${@9zk5_CbGkujpiIe1Z<5m>mS z7)VPCGoCSO_!;>9prKHJJcT}KL^-Q%K(STcy5V9un9*x+JUjZxSL}2fQZc-yDID4v zb1`2R>4GMSDxAL9HonK5KVc)RlK;xMQ0DUE-mhfk`;AVq`zi{jakzI!kGfc_#f;2a z{HkOD(AgNBD=}+<-UqH`i=UdiFdC5AmKC6U2pi@e3hV_jb=k|^7{G+$EByl#4#Z}i zvEGU2Lr4+X&0>PW>(}ZGb7x#kp|orPD_^QXGR%~)4&+Sq>s_XJK}G$V4#%4AeY-~W z)vQJK@RzkG$euheJ(}^g0yzyznT?l#*fREau{hoMe64B2bpyCn%R!l2kOnx~T_;Gk zPSBb@R*K1Fj6QjNydtT-Ui@Dq+hmi$kqV#OrM<#gN}Xw1UY6hAi0m>B95Zvj(zP$r zkq8eDPm8y8Ma}hZgO)RV7eJu0X7BPJKKWN;I>rG`WiZwjyaFpLtLL-_(DgH0IkHsn zbTh7e{hBuPk0bd&L1sXYUD5)YhcS}LXBZ-gL&$(^kt_Pck`@JIi21=;ob;|ewjS?AECE}SYG3|>fNBctohKU zj9Qr%pQ)QF#JrpMT}xo94uR%^h19uk$YYPQni|mod9Zg)*~=fwDPf|a6F-K$5cR7b zu%502eu2gKnc*-O5WCv$Ry3AnK zX063+luV!9_OM;?AaDXTu1YX!L~-bpv3kF>J@A)W_$|hxA_1*$!EYKbGxTiD1jkzU zYtd6{#WLPU`thz`jOyEEMKd?Ps><)uJ8A_E$KzoX^I6R+Hpe}ig0vO`*g!$7_%vz zn=al2RL>JwB1QNNs6p=V?Kc!@h-1TQq)=1$De)*lqWI>;VZ)5YX|3rINKW!?w_UcL z<0^osXh*$~r-t_9AG%J^f9zrVu!TvwSSdWMVNbc3f11 zdL6}hwEwVVu^`Ap`b$W(Ju^UdB-_3bv}@Vx`?xF}SCC|2-$^sF+q98e_UZXHP>*&* zuO+`Kx0&tDIh2_q3tscVvkI^_lsNdq?EWnH91`e^^9)~`bVLT|zulF&12y9$W zzs(&`6=+4u)ZTCTWHXZQN{!#}qre($dDhOCwlZk&4TyzmZUdN2poUrAe1R;O{!`{% zt|j36EUW(~SA2?F6FJosexW;<-qRTM1d3{GrR^8BnAWL!lNB&VtkjwO(}q&a>VX{n z6Nhuj6MN@O7{7vZ0}ZFYpWuoYG^|XH-==PQ2H?RYrE8| zgimI6Zl_dk2TsJ_9=X+iU2NCB$EX1jlV)iDd zaG8j$q5dFVymJLb&3h_#SEOzt6R-pQw(K(zY^<<*UMJId{fQXt;kL}&Ni^;WCM7Jo z%YWR$S6Lk9>nHz}SO z-27blyI^Z{R}c56P_06xYbUDJ$Z7fT(m~OziwF+xB&j|YPDzK-T2=c_Gg!V{F0{3D z`v(t~1w*!IZFuDS=6b2H&*iKhM$*vrX@usWdULBvLCam4CW#ayR7U>{=GIJHdEoTB zN_y_ld|a>jm>`R|?%qcMz(0KcBDCz!SW=I&tacOqI=K#lL_NB>Uk55b+{ zo!UD~-CScG#S&~DXa)g0w!3*WKEAfb;;>9j*$aE@g7{>%fqi&PrfTWSGX~0Xn*O=F zxxdLxFuw|`95p%*AA1f7++;{pk6WeO)-?J(-FTOxSK8xO;`mm?LN4al(2c;3eMV-f z?gtL;^)P&zTw45bHBi6Io%n_um4)8K4D7zX!9+BcOVLovXV0m|Cp&yFfeFU0)iDM| zYroG$8;ZeN!rm(K5?$9K0E>Xw>tyTTk=Jj|^JEaLUa`|9xsRala zwgV|F*A|}pcwX1tbU7^1PNsE>=c34;EPGm~T@pLcfYUL0u%)(xmGgZT?I>uf;Wr0{#>(vD+t6@T}!Nt^`|`FQ)h< zt~mYuzFaX&q#$hl^67}6X?VyWeV%36Dbx6a=ml%RI&IAVFMT>_uWL5)SdC$_&BY}`;9+t{`svvV})u0AGRUP zNKyh2)sE62d?zQoSlLclQ!gPQ0d2?nv(eV+F3S%+_FC#Il#?qt)ZCHwW(NxiNEYm~ zV?%{RK}YWQzTrpJ4`tZYr;TvDW#UI}b}t0=lk$*a*X!)v+u`Z5_%hSY?ctg|Qwb?(OZB z(w9y;8Xz;7-Ne7n_Wkw88Ht}&Yj8-T3hVY$|H#>;rQG395$c%yIGS=`j;RfD_gI#< zuN1R5k=(bJl7vWGzIGaY&}PXlR7#&H?fiyO{=0UW*!pHia2pobNo6dkDu8+o0U!AD zBNqz-`i&KGRT|ViGheiwiY*t5efcm`Hdhyxph)W1uf939DK6~z;HisPu{DF^(TsQQ z`xtt)`P9qj)%RKCC%moYIjh;w)N-Z?JV$Kr8lVM}JECCFeXtv~f z&}*fMyL|K|nYyJ~@(l&T-xlMBqilzNsVXHt2`||{? z9Ae`iXC@h2I`?XamJehq0zmG2%b6dKlW8-&^6QE`#o^zrX%t|KAn z8e6NOZdgz>`i{jvAZYk-KKMd7pyso7s*_eMZA}dTd3_1 z;t=oqq`#Uk&mS0Rj@y6y&`QKroN(}i>Y>o%QNy1|9r@K_lsNBN*`L_`m(2x5l#iBK z3c&nP$cbur{MNYg;~J<0n2Z_4qi36$=65;h&LuY6I(PHqQ^At*5J7 z{LQdVTMaXS&KxE<3(IY_Qz;zCnqTX+47I_8P2AB8O-@X{yA5U%BBHJW+Efu+uDC5> zp3;bf?C8zHPeH;ya9l0Dma)8UM>P|5=)hrOa_W0c)QEucjnAc2hn6>Z8_Ra*JN|B9 zOUo2`dAEQ>8zH&yhGZc;b0)gVKFo9&E6~1laNz8|X~XjZ@@U(1pBSm#b1Neo+OU3+ z?3JUV?v93S-liGcfMpa2B0=mp0^w8QGbJI2ysZTS+6jUcvneZoSduy*mfMXh_ORSj zT|yogt)y}T5JW03eLt|uRx?mn6_c>?NAF+*H&C_2!bzYV9zNy}3poRU@|+ z#IViCj0+wsQ5#1Xg`2ovSbz!TFe&=d0!4V?qQMr0T6^ZzespSXlkx5!VsYnH#`^-h zHF`3`Tf|kX^S;Umb7{a^?-l;Xqd1$G)gm-M;;hu=sJ@oZWJ|yeAp&D<(&1Bz)|LjU z;wjVY3uetePIm2lR0N2xE0e#lPlBH@$TCWvC0Ssqnlq2!t)eR_jwn$}MkbP4eNNnv zn}JBis56MMVoNtpA9`UYN`$m4W?TLDq?244xdv@{89jMosFr5VWolYUdO)P@H za3X(+k=b(<0i`w2-134{N29OMB7=cb?+Vj8*GN2PF^FRGi&;OQH;9q|*bl}{tVN~y zd>v;ir@h63m3s;{=A33X`dmS0mu~ECS@HBz&WE?cD+7eyv5)LB1f$8rH#uyPhnff8 z>tBWTeZ!)SnKK;cf_oP5p;6P@vQE>3EBMkII;h^8UcWVWGpj7+5{Ge_Wu{A%zo@3l zj@rm;3BTvRYa4cJ>n=}VuwC|_?$m-eOS}j4f4e%zxxRPT^|da}9ww98KxVny zRuB%(zDXGfi;SdUp6#S8T-$+1aA#;)@+Aj@HP5ALg;~Ibztb-nIjd_~5_5qyeQ6Rp zmH--YhtCY)&F*=Gr*pro8=RZ2J_y{bFK=^@?mbQwb2=@}iv}82H?-0F12QTDrsx2; zjnwQL?TZ>$3*YtY6iZV3AAPk&rn;h8nWfHiV&p$Q=z@M2&lZ{8UHw&VH@RSeOzlsh zX@iFcMZqgiFon7~8_Je9QWRCQp`0COOT#5%FFXO81*$ZDR@~CjXcW66&&ZF$pkBT8 z0^`F%p_A2~0$^NZA|(Mb#QwERy_58x^*f#e3;yN%Aw-F3;$_}&<6d@=N$=iinNU${ zDwuf9a0TMDtDKbPpm49~hU&?}NflouPb}n-VE^A1E3}z;{=D2&%ZVIpqDAC&L0ObK z-@vC*gdlC|&sXnLh4wb>JSj6!O5xUp)41gBFC97gDS!*&=VVtK*MKW#%uVho!Q*V& zK4Q#~c?hy@-B=#JUS^l@p=*TB)xXrHwIB) z5j`&W)vS~KM9!&n@!8&d@&o!0!E-#jQ3-{@>iUg6ODk>UJt}Fh=mRdBQ9U=+{+?$S zy zuWarSlT{FuXUH(Un6S@Eu)?nyV`DNd|Me+E{OU1$17^ z-@CluymMisr!N*N51X{|pXiss900>R6P-^hiJr0vV&Ix46|^&ZLh0|1O0I2o-_RNxGUdK z*I>P%Te}8DINZgyzSUV|v4{&*Wy*zo`L%81ZPb&zu=^>%`=!VdOAH0bW$H=pz68Ai(?k-5RzcDPp% zM!anMq*}jb+;wABeNt>9qluG{_s9qPhLid%c$$W5xuL0c;;FW88(kf1CLae3 z96O6C7!su=@@j!n|C>|uHsg)`%*J1N%{;-V{}A$JQ*bZfKFc!W`mX%eHbV|}j@W4i zVEDZG3qchP#&?%s|A31DtZC@_Kv8ptBd6$VRYRg31%W{+$}^0ngNh$n4%jUchNIhJb$0{NWFev9lA zt5tye`EFP5lN+oyn?nT47~sMyCg-2i#8t35Hozv<&V?*%MUp#{vHH;<<|bRP z^zuMiQG076GO)9!5E#GAO3Uvs0$HV^qz_JEez zvobL|6E-~*F2s4Cj!BF1l-jU@E$Ps55`59BI z;?i;e3V&h}7oqdD>)|I~*(u=GO2xmB-Xv#M>!(PcusJV`+~QR;(I;jJvoe}F|7F$_ zP}29nUZ-YwA6Q1cSlugBlZaCYXA3Ay2W#W8ryz-JkN8%=sX-mKa z)zcYZ1SASTw(u?!iDjdwZHvD7lw0E>y^}1}OriKy!TSTV+HR&dihObbpz@RjtcP@` zdY59e@Ei76kxRyO$V^Rfa|OBt{T=^yVoH?8@eXH5+m0IX;=vSDhX$e*7n80kI)W7Z@qhkiOi zfvhSf1DDwa%Ja|BbAnAP$v7XH7ml;X(W7Mnrp2>?&(@6F=)Iy`2P5KM$B)-1hy9A} zhgf2Id@ya~*k+-ehih)6;{!y-Oyf;qad|XN2w`47>2L~9!e)=Lpre^({x254#8}~@ zfG#PlWRs>&V)r%0Dj#H`d!QAPor%w{mBo8R#Y0-yoZ^P($TO7kdqdhO)-RmZrxMaB zjcgV-9?w;RNOm39#>52AfGl$<$G&mv9$oLc4seQGiIEE~+V^7}H?Tl58cg`h_cq?8 zJ|qzWi?cPp`n_F;G@H$o1=NxUR2St0e;RA5DVeDq5G|A4zu#vA%i99zxdNhOQedgs zYw;jHx>!k18_FsAB(etnEHb6a%-#U*!#eJ=MH`7CdR_)_!s_3fuGG^tNPWi zAJ((bLt}3YGzNb!7Y4Q6f!=|7K1aD1GH$hdD~kes&&rEB`kq?I1_anxFiaLJ_>?hF z=$zvhrH879zYfgc(GP*pIYL&5ZGU@GgxC9{cv3s($uMD$g=zFkdePM6>wxPH%p#FEx zeLNyMVTHa|nd|g&+Q$a-I+{EXS7Md0UeRW08GK^fxRmN-j#dSeG;Fs!e6o;V8s==f zI-G|c3}w1jh3vfhcpzKwNu9)sMCeG59Q$)ZEpNlXD6G09-NiFpui%&2$Jri(`H>ur`h zcS2)Vx9C4oKEwvu$d~v&FaMI`q3_&qAmS^~QXWkYtH2N9Xl7hm(WgPMx%#Kljta2@ zofN~mc+1?I7K7g$!uh9Ce&kvAg(p8<<^@Yk$~8?%qZB775n$CG(sUFU9Qmt)AR28w zYXR2p;5k0C9@kgeoaz=-? zOr<4jC1qptbcczBFe56@;j_#y)5{C;at}qZosrqf4!CZf%sObW%O0N%1$Ky z&@J;i-TP3SN7Y#Wt}i){n9kZ_%7ipCMBX}5r8z7EE|BFbXN~Jct=N~2n`8u}1N->N zuUlnbIWhqT?j4?y4`>BpuA?)(3I!fr!7TFoe9xkFY&UklGQ@Gom`%x<`JM&32)fKv zeYl{_@t~v#bPQCQzcbm72l01?3>9U>&{~ZVB20Suq_TH}aiy+d1*|2ob*r|7JL@9u z$gah$V?*%gECK%|s);9) zsSVGC3QEefC$X2%y}m>35gaoqGdeLh zqaT_JpI?tR7XN+@e3^8FWZM{~U|KJ8+kQGBJ z6?X*p=BeeG>xPNu~MjZ&}S~njQCK?it57*`+p|N(B6N|`59$;X;A2@<&^RpD~Nj34}3?$a)e!u z_UF-^SC}NSfa%U<;1%SsK=n_hn$g2r$DdYxgZH%LOp~eHm6p?g)ePvpB>t#^q0LBX zni;|c^*RBkQFs%wKGFJFZ!x!s)LdJTUAHbb&Psz`;GyF_6pjI)x}pxRA0fntjLvvy zn)uW`qkWrC#^y|wBv-cMmY_ASL)QapTV8oZ$2=|_goEkc0e*RdaoDKF2&=G9Gqfpf z9wsl7@sp{2>Rj;F6;Ly&>GNx3D#v@+QcoHP_BnWkZ5p2=W-8neNYI&Ss{m_89g#W2 zZT0WVf1MzMG9^r?ewmHM?rOL>yJ~5qNT1Ku!bYA!!yF^E@k5V27gPoAIa2W@YrdM~ zGGczRn$83EqJH>?yY_C<*4cn>tF_p)I#|FAX^kb2R$L3AC zg`iSr6Ua)AnxG~j zWEJpQI2Z0d1rDREO6Tc}(7T29c%UvXUt44}Yt*Q%A080ATKTqXqcnJ3&G^XP_<89~ zve0V&vowz7nybRzZrqwl_|QD-q)hlW&cB2wAD!9bq!M-}muCFBHCEj{eWF?op)tRJ z*DPrRR6cGwOxwLN^?ioTVR(G3=G5en@Uk3=`Q>llbDbu)0V#h~3DpN!!qJW|Y62ZE zGpkZxAr25&qM+Mq{_ku>PHC**W5$}sr>V9&CJL7Rv@sLP_wDft6&B9qsF2o6$GMZF zm4DoP9B04&AW}vn20sw^IO7=cn?Of$qR9vCF|cz7fAh2S#4Q z#H4>@I9E3hWx3X|Wk_JLH>vll$9yTOuoLo^XY=k9bl50M+WU90VTlb`0%w#tc^nlH zgV49mvVzE88l4)t>?sq3n$ry``v3VF#AE)KEk^V(Kq+3~bu?pxm*z$}QT^LJE#I1< zt#bHm!8;ycVN35}AOmIwL^(vg#IHX^IS;dm9L7d!IV~ImEWp}3+E3GJovZVycs zXWqTJl_~`8md7?b~nPdCU z`xKxbvzrp6_ah&K@7pZ2TE%~2FOw~ z)(A{c43%vl{o8n<^j&;HsP?bNC)R8tMz%lxY>ofpgguIz48rfUQ!!)|K zf05I#vtNxQU0k)mgKMOjmSUi zWV+Q!P95~PX_Zik)~Xs~?!1t$pNh?(jMbg&;i|a2+~^E^C>*`XyZvs=wNuSGMl`H` zeAl@bun2X3RTOoxAd!3eDLnTA&(Qz1LaaP;Y-^S4Pzw4{rGxV9kMzUu2`dB+)$9Z9 zC+|cKE4NK7{*)xi-w_SAyw~lTFy6l2)WP1@P822p?lt_(8(le^uRb)t#}mi$5QF z@Z{0iFX}+B6e?P7!?Vb2Ph#Zn#R*szy_zZS*yd&3DXER?$$^IOo&{^wnGYy|{T^?4 z!In-DpMSL#gB`=(cK#Xu@~o?1lPW!4+!>`eF~Da=V@_dka?RA$7?I;$Z;+PBZ3o2X zW~=Y7Wom&hBkPW$wzXqv9tuz%1IgSozIWvUAXvjX{y5hE^QS4b_gXO-lC~V{pNp#D ziSUdQi{{m}sLDw?30A>)5yQ*xX4;4nMoR103+3D=@QeiMipuF!Y60grhZE_OTJGvy zZ~rgG-Z~)artkL$q(M4F1VQPPmM#@UQo5D~Y3XJ`1TPw-8$p-uZUku%kgi4PMmiUM zv)<4Boaef)^E>DK$MW6y&dldKGoN_B#?G;rgvPcm+xa@%PMQ)fG3g_?2xXVtnzX*j za9FTm60zeQp8hm*8&SiAT4d-LNuHto+GI=YC&jo;{gx;{G{W-ypBns<*jxT&dPY;%??NOsffSpVW|Jg~O$U`% z_Yq^A?#TxFF7iAmdsBW2;$tsTNgqvy?wXS;cC>7TOP-sP@N0I?4%>lwUpz|-kiQlUVAXpItX-&-P_?#fAIHaVQ1B_jT$UmB?QaMqB@ zTX;GNaoi_93Jh7K05a~NksK-8!-??%qM{U`yl4MXI`zhYo5hz_|0BJYis1_UURvsE z@mvCXr5RjEZ|nL)hWWdK>zuXR!KEZ?#%Ik?eQ~KFBN8Y!*kc|xg>uII5+c(H#f_kn zFu(dle^dyhczf3auka8Y@0*ExS&V-fAY=`+dg(>d3om zG09416?Esb6f>br7-|;B)>JUGEzs0jS`p*RC;&& zQ=sJ_D*?(3^1qqvYh~!BQ`sq`SA(VzQDBcv*=8vvF6j;^0G+jF^Ex zmOA!q-gl7n9|?_<;Cf(5yuzZ}NWD8bjvAeNl@+OAM3)A zN9wfMY|?{}ft4$Pq---!S@koxE{;GJ|1qzTXdTR8V`8F4ZB z#!kQI#H^^ue<>fq=cHR)#sT=`KU!ERW4t@r7%`PwS|PA(@tO`z8mL@t;*6mbx_8k? zUuYZy?Xru0m1-}Wt-Q6GN(u*9{>nfP_^g^R|6QVp9GsCRDVBMfc>K_dtYOrV7*{Nv zqa$B}>Qn!|o#%9_XxFvqTkdMQ|JBeJ;P!d-9)$m^p}S~e!zzFMD7b?LLi`tXyhNB9B>(6tBKhb5 zOldzQ|Mh*_)xHOTsKnWL*8%0~EXbS0G)ImGQ=D-Zd0aadIu7Qu0F;gY`$ZtvP zbAMz-{{$Q3+@9*aP{yl&s+o^*$jDdRtSE3_K5h0B_M2&P#tm+ot0X5U`Qh}n{GAjx zS~bcp#s@<%A5mZu*?~^{8pqRdAJZE^US{#))Ha4oW^O~a>UlyPW7h? zyO@Kt9SOl%*@A(URUQxm<%G3ev{dC0cowio zp-XrX)N1f5zVWtf6O|VSi87N>VTU=MSPU%)K;0@Z09F65_4TQR>nbZ>mo^_NH0)Fk zJevEc>v=Mvy^2I|yes^EWcfXv6&bO9;M`09eTd`n&mtgrHS+3v@DEeKjFys@LD&h* zBxwSBN4abMq0#6{q_qrjf8}~K+Mj_wZV^Zn9yjgQrg1=;bu#@gtsAZn2Q0wm-Fd-n z>Lel3wk;F-`zzKfz=u#w;eD1a?yaqW<_uo&XB9rAB!@uial zFl|!#07A6c6GNn`Y9G`O*Mbz#f=xrgK;43SS}Q|*VPWB;1-O8x4T#2>+hcfl-WAgM z!y8KEHp*1ZeDEH8HFm(McFYkH5+E4>B%f}&C44orR+(Nft6c%?!cCAlAj)81jjgsA zNbEIP?um}(n8S->Qu+v}Y_^)r29M}taDYLksKW4(l2_SqbhzJ{;pk*)>QZD^ftj0) zje|oX0J`W6_#v!-ULujkV}4X*CzulQm#_h|+z+Z>czAfvn|kAjdf$Vk3Q1@}3Zr9IKNnSrvuMIHgAB5r*AZW; zYV2d!-(E-)V^by*=9rARl~?hxK>u8)EBmY{GN?Ghfc}nQM9nB&!_Agj|D#B&B$fxi zTivX$oE=Q})*DJ6&ncs1R?aZqinTHTy=y}~?+f8>N`frVyGk&OX;lPcXufZB#mFzut= zR>@?9{)1~o%jo%UzTHCejXWvlhl7X}#(jP~*A#~!bjh&@1@~55*IH0pRMu|37cM1N zxj)BYUG`D2lgdYN_*dtL9`x54`He>Eva&%gE?EQQsaooBdM2Q)Qq zZf<^SUs16ERt!q^{gP{iQp#Zw$2jqnlne>F7G56VqpB=GX1wxU zsgTa)Eo6peJKoyo!6SA`$+Vr#*kw?UV!LXCrMU|Y4Sz)SP4|tRT|B5IQQ2#|^{3)7 zeLwHDo{?h~8ZzW!{Hr7QaMuw!XM=hPIC}$tm&QFx0b@YFrufOt+!oyTzmb{u@}l~L zxFZW6CdPn)dT%rBi=Ca$Io7?)x^FxYus)c=_qfT;`Satbr?fJb!7ep6K_hfU;lL9~q7QsJmZ*oa`HYef_e9(YU(-62IH;Op<2-a$V`qyrC!F$8Rq7 zJlbqa;2{5YY6&3*FeQ}yvS#zB|G)=;m*!akG$<3P_VDJJs&^vF`IzzoK-%7Ul)Mos zGMNWk@K{{%KDiv8Z*VCYn!T(pqsWJ>SKY+{;5@L}vUeXKS6i8O3|HUpJZ9ZG+6Q0~ z@q!Ei$$Ihb^XJbi!NlcBOWNWDBBhhCvqh++x3>8Lfp=nJVwiH`GnsZr@cjDY#r@%b z$OWST0mdiIqN}~8;VO5L*w{Z5rPP*%qwM@hAoR?uS?y!ail3aE%x_sJg5F*&mSEU3 zrf#e#WKvZS2cjcy3A@V#c}o_x*skvGCrxa~nvyS|fB>0su$6ga6;cQHEr(JF{u}n&+pE<0YqB3azM4+3@;M8_1JLZgf2tvCQ;uZ7CA#s3g4bVZ`VbV zKz#S#Q!gpBJ)20Ut%iPiN_TE!p^-Nw?2V9TMyNO$ZXZ$&Sd?~kv7W_!v_71KRKMWr zKmZxAa?$^)%lryV%DqZL3k*P^mSnpKBD6L`JG2BJR>AyEk#W57o>V!8+i&yLWPP^L z?d4R-vVf=ePDM}ScMA;ga2_K1lYNP$P~zeTiJ)KUp4jNT?B=ejuBYR)fVj_C@Y>Jv zk4|R4O5&>V=U)a`EInzpyYHFOxaxnRjfO^5ha*7$8F>Lc!y{eA3=vJF(y`yEhlX;d z8wtn*P!#~k17H95HI6P1i4-BXXhEjO62W&so?nf|&D)~`qRvvw?C2NKyI z8Ii1n2mFD6%@%(1nQxP0|2{q={gW89&8tRZy0Ukb|H8bWpb~=j)3Vt80czmPBY`Q@ zXR$?p7%+do2>tJi|JfFlKva@9*4u-geVipU|4$A}`hhn;Daug2Fb1YJon*ac-r7Ih zkbw4fya-yYS@lrW;T-N)b4TK3EbPaOKD^KAD1mvdEg7z+{>N4uiG@N3lno0ev@IW= zJK1uv*yoH=l%8$oeUwNZ3=XCNa?iMk#uoQ=fD*?a3A}lg==-N9aE#l*w-TIlX8+j{ z)BQ{WTa$l(kmtWY_;;rhNl%fcnAVy~KiHZQsI~aJ`=Soq7bj^*_jYU+<^P=(Bc=N? zP%|CvBNm^Xpd_`g30{qGO{vl)*v3{$k<>~9SEev1D~N(%G^=5e>0 z{XaIHy##?L`qaFT*h|>|10Uv|P!7lP2@5xCFOrob{(pq!NIk7?p;!|j(>*A#oEzvL z&Q(fN=`c?V0#41!Po-5P*M$-jP*_;@(f{~2xV^fdslqPC4Rtm3 z3m^fef3xNG0POcNH!q|fUtH-A22885p!wsBV8FR`51%8QEgV!p(-Qev7l8X;5FaGR z2>}+TfBA6t-hwvX@yE+Y`fyOBIirjtKl&fD33%yCvaxl@YF7E24h#8sAbaee4Rb+D zvrXmGo<9A!bzlGg?zv{?R6ha3$JRT596l-O{}by-O3lRM2`?GoYV9~0^O9{j3}|V{ z%QI8&iMwHkpm0hBIc4QD0;8nFq62Xc41VRLU%IHIaoKX{IL62bNw;54B5;i;v5}}v zvXQ44f1qmb=)s%sorUWCv2`O{&^hW-%VhsBOR08>jg6g3O1{%sI6S4Ik_9_38;tMW zOPbF&e;W?u=sd|rODjy{h4{<9(*FlhOqm+J`!095#$&QDj?1cY@mu;Y!8a=^rPwRk z;Km*_7Z{NIeSpC?;=c#piV5^__NQIqNwKT}Ds1uoXYJgmjqO))`y=-;euCS^iX#}k zj|ui3DEUTl6(dxbcK{pEwLNYG{0SBz?0`$;TcLm6dQGAF_tgS9+}M>!PrSQmab!O8 zzr1bGz^#m2z;yi&ql+O&qcJq3DwCT+iBCBQ2h6k|3^7?S-(Jjre(k+@v;>Ri3sW?_ z+?UH%(Ai9=*UP}HFEK)B!i-?nK_U?G(Y{jZlp6;|(1=c6HUO}vA>}a;Y(G0)02{ed zzwZfAQ3QbP8tD>OlfH4m!@SlT!kp;NLBR8fYD7Qhi)Z< zT6!n*{D58WwoF=D+T$xw z-)MNXlMswY*$?(p$Ys7(=sAQ`Skn0|0Vk5v5%qKlTNvz3R)J5IG-U)eM1ng&#oM2n z8sNMj5gT;|Kmbc(pDBxPfcx(uh5oB2+v&(b=z*@~z(5eKf#vr`Zgl^x)xKi;N`61W z&pSncZ$^kjX8I0JVEKDZG9q%yXWznqK9}}Y9S_eA*tlJCCw5wNYVfuoRwYuEHdlXU zz9`FJf{Jy1AKody%1hqM^w>#y2LOvgg&dzPVYbzMd{}q6KwRnJQ~N@?%ba zDHX*s zq5u>y6QW{Ep`iWi&%js^t8o3CLEL=@6?=v8?YNp;3aHB+wO<6KoRwSl0Ggx@V`Vnk zF7vDRhYA!Mz5R+ob#O$p@A-jLlvz3b-qY$~zde^1b1Pmxz?n}2%v-qc&pQvX$4xN) z>nHx#uc;P*N0a?zkjDW#LI$nMYFN3dbW)ZHA0xjYhOt140icdyESeYiD3aAViI~82woc8N65^q@h4Y`&YntSVM zrDaamFufjoW0Pu?dOl5~)HxWW#Gsd&71}r{G9_WrIMrtSk%oH57MdH{y}er_(&sM- zOX=2?OV*M7H7sYk%{F#1OZt3y;(S?2l*}+I>~WTqBb&8|mbe{Bn)ZD|!h$^tQb+jL z@#*av|H~Vg<$Uo+nrD@Mxz_q#We9_2hmX74N`-ly->8|IY-=6*4b&x<=_uzMvqi^w z?0k>&U_DWbRXA9;Nzeis$@4`)h-7oL7+Ppak<$8B&-}@(SyFC|nc_SoPhi3sZt21e_NMo~_wpiB#j56+@Yw!6 zzwTqh+48Ypr?;IEH;u-kz)Av%m9qKU?BtH=d7mN;CN~b+4u&@~bd%Ld{blrF%2H4i z6<_xR^stZlNHbTA5OsmUwe6;Ub9bI7f7JGQ|7VCkMDU>RsHD8w#rR4}`?c0@#pZ+0 zq!I}GtJ9;8FC*tBzA=BV5n8|K^i3*!czaK$lmjiJOx!{bQs1Rx!OZtIbkk5PeaG9Z z!Y6WJU4s-_wREYf*kLp9BdsLnST8$G!=nR7Cv=ECy?myAEcTQMkt{`Xt8f-Vskoln zgLsaSWXrr4h@uyxYGJSFR+Bn3d_Bn^$VrGVRml;kM)d%|8a zu*YxDecL&mrxUo=7;^GaYW!v^$a21!P|xX&ceXMV%5KlSNJ=DmRWy9s&p2X&(N{7I zCpwQ((xW$g=bePO+4bIi$>(s(r&R1eA49-f;TAd5c-~mN&6coJOK4t~{H@z!@9WoO z!$q6aQ%@_n-=@iQqE)FtckPkB>GG@E=p@CSBEp8#$R=9#-)F&XoCBS*bQ5!;Vxp$p zMEPh?H0*56w*%S6n>mFhw2!=3g8HKJ)Hr zSt~Ly)3Z}OrtH%9!MBNXa|wXo3=eY!Y$uaJ?5*5*arDhNi^K*%EZKo8>)+K44Gq5v z+9{H5)Cl1j7Ts7o-WS26+Y}Jr&Z>Z?(!ea2;22x0iwur=dzxo6xCPey=+1{Lr)(+ubjm}z*J-t%W)eNU9 zqs6Xn{Y@EHomElcaxngv3qRo*8ye1N0O=Lm(1 z{_Kd1>ojH{vDDAnIR`$Heg^MLrCW`a0_IOD za32yB>ihSvHRmh)$*~SW2;i=r`l(|JSKGBS<&m0tcAm=csd+u9QyeVRJe6x2>gB39 z263MzQZ7UWr+)3=uF<2C)4HSYP|+(Pt3*asT90`rW%B{D*3zE0us2VbP+Is@3f}Snr0LW5EDs@zIHoIZy^aND7%Lrvc7D+V|idsx# zZUC|3El!Xtq=PANj5@qNe7lR%=s`YHpxO?j0@Ba~K}Zu~>{FoS3N1Zm>pnJwc4tUee z9L{GAonz!-6;U=!US1njegmz%gb4W=ty{AVE45i0xv8&X$V44avmTXj&##rOC6eQI zO)t{o#xB)GCEQkKRUmdePWEn06YGeR#|eDLs(62hbv5qQ2rEWt)`jT?FqE)1weLr8*^r?dNZNnI+T|L>p_zXt$8^j+UI zFTVA#&Y1IU^ts~d)}R35XR@MH529_!Kv^-gmlgl)1;RzWxexqj zFf{V88h@@of!)vt;S#sodbPLKvhPMG(EH=y9Kak(BGFHnDX@YwZrX8Ew> zZSw`#p#8c(P^8hrFtA>-DR_Cq5cg7uX+Nvx=9g{6VP?D(OW714B8_={1M_d^LwEeP zqm2D?Z)`6Ao1-j`9!+)i<&XH*Vk9}M6_>NNWes^b$fEklZw8F?vzbhvCY!z!NSy!N z6^;*^t!tKy<19n(eqftvpVzd%EfLP@Q#Ic*3mIeG$MGzC>EPe@EN}1Oo3TZOLbrsI zEqrg!aYgaoumWop^IhhOQz)^I3p{m`t-&ogG;(eTx5A9q0!)Qt^mycQ*=7j{|60=s zD`NJ^XvyEpOcIub2m*Ow?oUyggx+Zj8s{sej(GD=tOQT1M(pz-mr7$l1WF`GUimOF z??j%g^*~3pyQ0`Dx*&BYUl}|Kp83e5gzIp_gwARPfT$}yJvTlFs_>X@e>iI5EA>VQ z4WYo0ma9f`G_7cc=W?f{I~*VYUGJ7%=TX(EEIMYHMmP8z4FnE{%=X=BbYChh4PR@4 zTjgOJEyLP<{Mh(qc_uHHXaoh9x4SgHC+X>4g_QBLk0aP};#)xk81WAkqkM{7@v z^yA>`s%H$bM?gl?c!jtGk>$A|-!bu7Pz}49ZY^jxZ43n!+K%{cLc(*lm?qFFY!|`L zL}+&y0#HPxj5R*Aza@@Q^#oa`r)mmg1jqb)NKPqsx;KdjIF};@oRQN9Vyn<6B&+dX z^2vFs@v_Zty-C2V-$I9i4FW9sEz$}?%N zPYO`3dVv8M`3 z0`n%w0B(wLI#WWykv4;31!BO_OBX7VxqH3~OLO2wrlEXz*;G`$l4yYr@rEtMBD9f2 zoKbN&LsUq;bd%1Tjg4O86cm!XHbHT-NzB~*`dmx=u9qKr3|wz!NmcF&=(8dM9vO<7 zE<5feT%;@bM$FV_^$l@bioB{V6L9h&)nIvH1v{(_*08$#8mLWcpPWi8sR=`jE9x3I zPR;LYI_;B)<&v(M_k;NuCQHE<7C`IlTH+)4WbKlM;O$Z#uA7qJtT`2pEwuRON>iYA zT6YGV?9J3-7_99>p2Z}4#Sg&)^cN*<;6?XPI8=3#@^84sK?~ddcZP z`p#=9uYba*v$*pP+8Yt+*HEjk*1s3R)Y#PUi@eb;qjn9hTz@{Lwxs=dAV2A`lf1BK zZ8oVlI(kM^be?ABF3vPk+D5JfojO?;O48}Or73#aTao+$al`Ku5|VsiI7rsadBxajF-(eM^`wi&o=M;^cY*J{c-l@Xr?g>&pw{f zxy%cVgDn=Ve&m@7juTxtCt3`aHnX)md|4NYiRT5*7MMxd%+aq?VA%ug{8=vzG!<3b z&I=k3A;&7qpu8F?ie~smuO3}5Wl7Pdu-WKZqX9N#%MpgtD6Uwc)^=J`)?jFBcUX8g zJZm=Jvw%dx;_#m4y!x=q5Aw?!dBx_XFm^QHSiXvx)bQSanTNjO>v!JBK9D1(zj*Q8 zLxfO!S86I!%&%ez`?JU-2HT6BYDF)Vb53^lrjvHn#PdL?PQUW=Q~hdspD@*G|D2&E zn(KpPJ3{XV3BY^N(J6g^rd)iWjsBY*#j@ z6J;wr@A&k_YTh!c=LJ^a-rWH-tip)NF!Jp>k?oT;p)NhG=Al5n`o;PT(}#jMr6%FY z*LCVG;tQ70?Y7-U_2Y29iVJS^<%{(`A^ zD1wIdI#MR=oO6*13h;v>uw2RO|B{Re34;g14d ze_=pBKV`#tArdIHi8Jp!z*(w}5p)AKV(A`O3IU&?_l*Dz6>V$*L=@boUZ{*Q=&kXE zIw?#pr+n-4g|RMND5`%KKYFk4xWHYgz%oQ$q~RvO37523w3U3Yub{Dlv6YCe!(!Ih z47C)nH-v2-;u6LTknw3|er`})s#+&6L|mHft`r(YRf15{x?9m(Y+lB&Low-riZ+{_ zs;Tb%L;Y{v2z-48`1^}>r;G0D zLR+h2R+~&q!RS(hcSZ_Hs!MMUxt_Funz3A#2wCTumFh2XJ6BNL!Ek4Ma68sSr82vs zkZPH=bANiBbzXA&aT1WOj9dw6zX>w1=$ri6nPnPP?(Upb*SzI1W2=q^TNK%KwP>?m zjP)#OV=13M>?q-amuR!>Pn17bWH9*V9ZZ7lQ?&iW8E4Z(jrW>Hd+^t`Xs!DXXl7tx zhHWN(F?h9o(wd?$}|1QnrtoUYBE z_9W>gdV;l9S~_!xjq`UkhhPA+wR-R>WR`iCUKKyklP4h`#DDO)11Bf7gdz=m7EC_S z7`#sPeH#pnfRXSyfxWkJQ$^lxBgW6!d;yL6C7AgzfG9yxR#J%e0wE zqi*lW=jxmxbJsm3^+Lglkix^%_^EtFdq&!j5+d#i&Tky^+67aLui0A*D9bb}j82xB zvz{7mA+%+omtT_^S!P%6n7s0Nvw>OXSm;}CSg+)1m`U4`@-(M54-OSm59@$X3EJ8^ zsx+P8D}|SX-G`XOBzusQiqD9dZtnL4zo%M~TyLcmBF1g=s^#oBngVqe;#-u?E+qo1 z%uYJp?>o(pD^=l$6yGZ$jGNME9^nv$P3b(=gg%uG1F`d*d!z_hK7bJ0iXh(K_1mmw+t zg&Wzj*U^%#;&Sohh-V&#F7{>fly4-uLn3;u47am{Xo_gR+m%+RhmtGgoOHEW_Uq(x zp@|E>KTDq~zPtQO$Sab~PTn{JN%98K1>P9%7^yEi_- z?#JNKQCG$Y=8R*-*D*6&Hg3wHLB~`C5y1A|Hzpij!y|V8uU!n;*4o&fCN(e7ROK?08sa48Q=#6fUi2b>ddJBxD5f zDNyg`kKC)g;u9R^!O7${LgV&;R_uaS3bznRYy;>%;k;`?HzUmjcxa4Y4oq~?cZ|9u zFGLca^b+GC$#Kt1^=(FT?@8OZjul>L$kL#@52>RVYBls3!u-C88)ky_$-D`7kzEjRR-=G zMIPr2=(Jrs!hAZz-PzXrn6!ihiwH0ho=c&9HX0w_K=7C&%V(0E zcD_55DR$8Ph3!N=?$zVY?^4jyCP--VeLI6WgT%_L8W0B>sq2e!ENf-=ei9Ytl?6aUKoNUeq2z()f!{F7y}JtIF*04rd?ap}+jJ zKmAs`gjul6wXdsE5~`4T#&eaOJ-kV;?bFW|9sq?h&8_s!m`J$tD=8_xIC}>o8G8_? zaFppHyzz4;So;*bn7~u z-wUL)bF;JQME6TMV);43Lbg}*&mKmL>EY>CEOU6+7mu29QcVHs8j8&Vr)^my3pWR2xCt(P@4>7O|zWuYvG7snvzH%679Suyx(KC zNUh$Tn#&O9?AzzmEA+oMGih+fK5yY%$S>)6TWYHXNi`X;p)QTFYVjIgRyvx$E!RK7 zlRSXY44xx~`PIA*gNx--s!EuRD-x$Ge^f&`S_-DO#;L!euehaK2$w2pGc#=ApL%OX zjP^NRHb(~OpQEnTU*zdUFUwYjtdm!BAaYMs;aerCYvORTCE*SwbGC9-a`X;*(&Aa2 z)BWiRLHG4~goNn@`FRwL=$NJZZCTDk{FCaLE21p98s(_w9cJ2_hGOe8O{XOVPa^1o zc-L*){VfOeh0PDCWfHg5q$G??-eoes#yYM}I^J&5$~%c3rF3}oh45lP=x5{KzH|qb zu=O4{QpQhl(D!>oT!we>(1?0pwM$y~J-Ic%Rxc8AqMYnBveuX(#Zyfw9WK-N#O z5n1+NpO$aeh#vcrPqIu#lHzjP>S-O6ogIN>?}|xZ8t!{1T_6G8BUK(xd$g^u4)K8Y zb^QosC^eOqjGJ%wh~rz&p3T05Z>G~^5P!uf@wWV|O;%2xUG44()4`=#gZ|aG$=4@w zK63FYl~;V7vebUJTg1r7Jnx#V0n)2KJCy>fcl6&9g2I^NM|Cti zaAGvH_?$e(s;907&+mNaRdP`-@38)DqHa2NSgbf`^Ffm$@b^jK=z5doYF8Jh(rT0T`*gpIbsOv$%u%vePeK6TkzhbKsblNW2zBbhr*ggq zjwv^>Y1>+}AyUh267gf!-%y$+x0;#b4Hcz|PxKv-GZ9R$8AsoZ53rLjO)fMf7IWBk z4nqe&VT!IKC7VR_H2OmI47L3YKU@5P<1E5GVCufRkhTfQkEsgxC}Xfn`8r$Xf5JZH z+J-pgySXL~+KqN6y2lduvn)8U7AApCr%HjAT3h-wuAjHnW}?vl4EiZK?Phb_XJ<6Y z$+;_x98s%Hk||*OLmiiK?c!>y&WV^Drjq55>36EDeUzXL;kkfm=gUYq_CIpx`^;uq z>**vS{ULcMOR~)v>R$>1nXEb<40amn(@0Kn%5U_3$Snx3elxUsY~q{nG{hj&#XZv} zQx<128Gq*@cj+%^H}jr_x8P?}?KAlFM4@ zNO7!wpn{9btkNu~Agwydx;iCHT)o@d0(J4cH+qL`qm>nUv&+86DKeQcMY;btT#CW9 zrEzq%UeG0Boh12ER3!8|EvvC3EUWBTvt}%5n`x-x(bZG!((Un-$iDf@rQN-+0g@GJ zg#)dnuT{J0s|j&~!zgfSB}S|0iJtZ@*Ao_CdR7&P_T86M9`P1HwP6nR$R zdVN%0Nr*y)rcO6^gk>kCg?`AK+b26sj#Lw|jP{t>ZtY_bY_$`a`uwnbOSE=3cy32_ zD$;UC;vxP{QUZTq!E#;KyE=k~)hJIjNgnY5$L_tWlv~8FFM;Y$tiGq`w0Uf8yMWp} z*EibyV1wy1{&gAv4%X21Y9{zS@zLhM;fEWi+219g^dp3*7M9xAy+8p*jS~`9ms8Jv zq1b1~U*kndJ#YYErA{;BCuaJ$`8Obb-gG7-*{azSJ~v~ul<8~6qEVXX8^*&r)=Kd7 zXF&5rOTmm6d%Sg!5q~1lY!=GpAm6b0i`N8}Dc@g}zXhqV`#6~pyD21Xf6XQ9$^gfk z7P@Nq-of=jWV6uvC)WK7m=lH6{VGUyd6^K@F}@f{Qkr=)eOLcR=NY4wn~ z#09GQipim7NHB+<8J4inbmiCypP7HZv*w( zqo8(PiI8ggtq>&H4E*AfDFhd+%ln1TBq4 zAVmSr1XaeGbaZq?hWZ>UYke@Fv;K>IXAPblBs}C<&r_NiMf%C+4q<8h0!(dw%hZd# zc47zLNcjf7=sn&19LH<;()TEn&JwqdGv8na1z_`+$5b||q5aCTWg4T4_^w`A z#%ssrX%q5SbgvYD#IDo}YC<1Sym)*^kttVV*2quQk*X-ovdz&Cc+?Ryt0kvCwJJoK zx>rSn8$+%w72mw)PrNnN{-nlkas8CnSfsCc5BMBvUg*RY3O zRLrAM9Vdt;Yu?u?+~gLM?}Tq`gJpy(A?MLB7U z2i{&}-SQlwC%lXqs}3}92OzZ6sHo#=$+nSm5Ub@@M}!+)=OCeGmJ4lk^uwg`s}|k> zIe#gXWUjlk1Vz?hUw-t=FPLiB0gG?L5z$ahNiy%PD6B6@XJNn*|% z#+e`o`llZ0-JurpSHC;gf@pQso3w5_R^(^bKOo@a%>+kv%SeIl$%^k{R$jmPI)q9Rz=- z2*+;#8N5Cv@{(Ypm>*2Q;V9}q>nOdTr+9(>*l!e00VCY36}S;i*1t)JpvSonQ;_F? zvaGEjaaQ1@S6-J1P3?wEWbEGM=3?E-TeC8a5R@gXbyGXKQngWX=={7Cx!lM#OgsW7 zdnbeX-)`S^(Z#;Mya0i`ZU~f!rf%kH;9XkfWlw#0;QUv(t@mETx(IYw194{_6o)KS znrg5a*KMsG5$!sPjlsU;Wfx@X^sNvV5ux6&?m5UtNs?F!h#SFkDFI(?;I$_>C|B4~ zkMYIpT34P^-tK)nvzo9)`I@4#*4b+i?{)5sJ6nGoEw)rCOTh;*)V?t{=&GiQbbF-b zdSH5Q>)gj>eq)6V4{Bxh5MR_Q_Qe)6dg>|DJDleDnu#=A%KQy|R!zxPF5L&8&`l%b zs$RefmVWJQzf98@aIENNy_v{K?Ch}NU#Zz~`#3;7dzyrwiYgVmcm!p~h8t@>6ssp~2%u}mADu|c~KgBHKaafPcU)@|f< zCrJN_PDyo(=&L=^otcWW|N7zBWWT>y*_lkf`b@tR%V0KhTU5}Z%S+d>vWKB0ula_M zVb|5GC3*(2%Vg7%)R`!tPj_rAWRP{zPoKSb)%d|1aWXZO;Q>!k&u#`Zg?ZFI-4|!^ zcWRpU8V3Ys?4flFgJ6vn9$6dzAk_y7n-cLEX{57{(N{1EJvqWsK!V^o2OXj0j`EmsM#Swyp>tQ4yab*2US2V!n@UD zia_>))QT(&NKz;*B&`=VQ&^D5VwsYJddylwnVIjsDo382V(4gk&P(4v%s)~5MSKGgLoT*@u3yX(4R9saQ<9c(WpTQZ7d&0Bccsgs-pHy#>)A9V z18J&W^Pn#&jqU23w9F>OvyWExGcq1Nd^TOhEa47=x~L6Q#1M^)Q8Wv-C_7vDM+?E8 zqx&TaN*s$b`+J|*+LL^Cy1L1$OZ6i3NuO=NR^cZYqdAY=ViGG+2cNmuYJ%-*T4|CKXQ|hFrCgPrFt7r&CoO z>7!XVD{kO;r4tLiV_7NnM)KPua3Fau>2ERV5mHIU_x)*yAZa+v`g=1pQ%vea<(0~O z^C>+3P1m3pILmDaR7k6wsgz~?!uP6zZ7f1>!EChetqtRLdOV z(q&;SJBXnFJ)!Ct6o_0XW`cO`=Pp@h`X2Vd3RuNy`lxe7kVi&c%ucmh(t<KoRm; ze48?DWsOp)&`tA_T)}<3>f%fLP}6L8>O1vGg^6e(E;u_hTxvW-Ru$8|t>LM2>(lH7 z>LlF?zpA#dg0zo*ojKkhwKUw^C^Wl0Y~_^Jvf!xT`Z=P@WZRRGOzwMk@rjtwYKs9M z*Hif3-Qkq-v8?_b2WWq+i*~xMsZ^?!TK{6uh1lGSTeqCM&HS6*VJSep9VJFm489fPfSo(y( z60Cgf1S0u4SNT}9p6v!eWe|A4E*X?`CB?VW!6ZjBlstw|;~G0VUXIQyQ$imxc+Wzu zjOH0Yz-)ht5hzXK33zu_$FHwncI9b^sp5pB%BU)W7boMdb{Y3tF1Ly-LGYxs1vW;> z3!UKX=5H~){dvjf^$rfs@UomJ)La8nC6|L&6Izhs%+>96AXE|ZCgvg-Skcf9FJ|MLg?joTD;#>FqOZ`xW$O- zl$lPy|9MX+vs!3JbmqPMa`F1QUYs2cZiyN-JHJ?Pl?Ea{iLE6mP6|PQmHcguJjIY< zZt;meg`xG9$I}sI==fb%RmHGb)^(FnjYv23#gj~tGpN2-5|AQ|Un-3%3%X?M4Y(Fy^78wjl!K;cAw8l86Nc*(l zND_0b+FsSj>NE2})vwKqUkU29oST#eGhAGI=4Ef|CC$g`+vaA+^0Bets-BA2%9`mlZfD zaO0}Ft2VO}$xrq(G$|sNlH~`V9wv9s$Ti+HP0cNCeL^?6v?uN*%d#HoSgURQFxB}y z_se6NG#8euF~rCBDyxYx%=wmtZ1np;&$qi*G=2uEPVBBZ1Se+Rv768a6v=4zDao={z0Ru7(2w2; z17x|D;3#L=EeLH@Sg2l9eZK7ddy#VZsD_W7(o)1LCfEHe9pA1N%m~F+I@S}j+q{s4 zr1dJ2_6h`*=&bdG-l_^ zJU#^}+;Dj~s=)5SUi(m`Hhj^i8um3J<4l^p6k=cGKpL?Ve|k!OR?D%Q!b=&K+vNsS=nTMv^eUUV9(Uxulb8>`Ih5qx zz6q{}lEWF-2)pBd8Do7Eq^IUZV>M&>F{;a@@ToO`sA|JjM4HELkY^93Wrx~KViR4UVyLB zT3*O1Z8R=6o_pND0#_>p@k4KaNgFS-7!`y<}4p-VNE}M{Y7a%NN#8^#z*?1HQsdZdNiI)Lg#aT z8<+eZukP_dqm>J$Hg_z^sBa?YUo;Uqh7QTurtR|ojB{aHe3R+&?j&YT*azzu$rY6y zd>bL?zo7Puvsn8{Ab6bSY)>CkMW6eA$YE_*rkwT;x{jK-lXGz8H|>?aHus^e*}q%* z`>XFp-Mr|oh)Br|yT@hc@0gsRdr+)mtb?E8ODi?$6#U`mQ8kdX%e|>i*0#;Yhjl;k zsd<^ne)Jew&p9(dth&ETPTUyLg(?wTP@#Y7mhYLM zo0?fB_YN|}>D2Wqdud;PvxnciZ^htXDi|y0`$Hn=@bEW*Chemd(FZmyQbxSuX(i10 z+`1o%%)Z;j218ExQ9~n^2}K5~TRfXGB%n8MC*3|5@0*eEHggy57tgn;`9;<7OJ@Gfj)xtHh$xO_?pG{Xxx0Hzgz65Dyp%K85YyH#!74D_ z#T-0U6LKphw^ONf!%FI;PWfQ8zMR(Dzmr#0IR@T&+e)J_Psk9S!t6+*Lep6G>Q&frXwdmwC zn{NEs$E=Gc>`?}eITNjRij34jTr}L(Tt+9&tH7rB3E1AqE)yIn7T0pq8e`n`>y=fS z{VXTWSyZC6_$K%p%^_~A?1EaqKhsCSiw_^&>fhqKZ(Gs^KXr{PUkafkMVA;)2_Rm0UY=Zv2$Mt{VdcU!z{yrRj~Y1$K=^bU)9kv3xBm3h?baUn7JKc*Heron*wa1P!v2XWvZXAH zg^!9IVUC8xu^(7vJlxT-wW(G5N@a&tP0f#rBMMCofhfl%#lgm@TSAP#Y*}H;W^ZEC z(M4ze`n?tV!K&ITq@beglTLN%%=j#&pa@RyZk4+#sd60YPAu@E6PrgFr zNJD-+SF>OJMOE{Ou7heR+@`tYBP|pDlqVK1*Om+r2e4p(iq6-np9ztV9OI+6NPqes zmay!uGSxOuQtW+HQr-w@lvOsQp5VzlY~Kyphxes(O<4XZFz9D~B66ofJi4^zG0)DG`3ItW z4N&)-;JtQFDCrjt_=3mp+N%#DJ0+rcPM!%oaNpP|H^upFT`;ZQlSEI8O2&8(2R$eD zFY1t+7de}&&XMeu_C^QyB} z$nt6*8{sE^+LttP7ZvjkYP@Zd8Wrk2Jfi>K%fxu7e;%>qDyF zwX+}#vv5^JKHw*P!bc^g+{CW7$3o>Fjbj6_7*h1YbgAb4=;+R*Ul^#Q%qVCUj86(0 zBo?6GlI1@)w_3*jwGdjf7BYJ*ae;@5ef$aqwc6;JMiS~>R8mG1v@W8*bWr&@RVmbc zRNX-F*xmxZjUE5Hc5Fsfcn`_r<7_I5q*y{!N<6wJBEV0MiaHupOQNR%|1j1dkwT$K z$jb-{NutL^CC3kF-vA0B5s~kFayUsmRH)?KD1mA(!H{TVx)a?0hlT?sP=RS=PQNeL zYKQ0^1oT*E5-O0X$yuNaR=W~h5!bjfg!pCc{C}Liby$?$8a^r^-J#OmNQZ!wL6?9a zDJY#HB@8f(hzJO%#L(?f(%qnRONW4fbc}SvS%csA?Y+<5*E!ek_m7uMz3;o$v(|dz ze(swDJz53aGO7gm_W#v#0`?ZO2PC(2^o2JTvh(E!ss7Ah3y;Sy0$S8mP_VzhhQ41e zC?4|eoF;6|`M77jp@F@=e`hl|nMq3;^Odod^ztVb*wy{x0}H$d+Dh)rJqf&efe^fq z8*~lbWQVQvp);Xx4@l<0n+efw zFk>(Z_!{4@@n~s?o*Gd*EgBuxiZuPnkI6W|4_1>O=_52aDHvcj#v=77-7ux$O6}Cn zRPl#{=j~nfDe}Z0-jdw%!t6RXe)fw_kOhv$j7AAhr3N>xKnTA6$Te3&>z|v(#CtCT zo+JaA=-P61aRQQCqF@*ldvliLnBa2+XgBZ$PEb?d;Mzxe9}bN4rqnR#`J)rfipL22 z?8w)$fx(0_blum0~%!vHX0_bmN&Tj&(*Ps7#62sITrx#bNS>{jKUoHJ^&?F~AE>e3*mdMOe`b-OZ ztuaS{Q&WILTR65_sTg7 z!|^+boemb^sVDw&)i?lQVr;Q`^J>Rd13N|h316fq0)R3nf@JoXc7bmB)6gX#IPHmS zJzL${5|ce@CN>^%_FXqZ+tvw0;eICoG5C)IBd6-t`dw`FMRs5VPG9$kQ8qxKqHqf~ zGh^xsphZ=Wd_n!bPeL*{wFIzQ&u}Vh3wrd5j#WKS!N0*PfNIasBWKgASw$oy2CV>v zlce$_beyOFP&DY4kCwUG4apxi;jjZDYW`TD8dgn8kAg-i=8h+o^z~_!+I(l;1(Mei z*3!^9pi+X=i;&v719dU;0*OHUFQw?Pm*ij-|9G>?aP_dz_X6nr4M)$c9)Pf~kN5Ou ze0;o`iAfUACo{wM`KI-gk0U#{_;p&z9`|Tk#y~+~!M5RKq%c0;sPe`Tr8}dR5J*2= zrLO<#!UTPh+R`TM6C-v(vqm{VO;w7dFr~V_VKMb$IM3KRr{pH3Z+89YCi|zgoR#Nh z(BByma+>LNaX*6Yl~sx_#uq*<9X#8xmc@4ygxv+tpV?e$Iuq&3*|;NW)W}}e;UI^f?f)+UYoTt2(P+CDOgD& zB7QKuzmDD4_DD4qvI`(zZto|d3(HMBp>lP$B{{aYo*D#Lm1QLOQ?5fGkr@eSiceiI z4J+8OM3%%W2xgJ*!pHQhitCH#C3bEH^~DV=^YrpxS}vzR8$OP$aL67j!?l#p?;iaOSh_rKytD>vRC8od-rFWM!yVle z0dmz~FO5kMI0pwY$*n-l8P6Z)ORaGIgwcw=N3Zs1U$T)MP2WC@i-Cdm!Fh9`C2ppH zgrBwLDjJI`5(F*%-(3sxR+MPgu-M^hR0+xKTO@mh(nP@^Pex~Q>Z0Q*a(N6p#5$1# zQsa0baTi_t`}-Fob};Ifb@idOZfo%il8k^Rj0k?gewn9< z`rlR*gtk<_jOjTB91ME(Ng4?gE_< z{aMvFQH9%WVKmc-dqcnKE69{7HQhkL!wPiOi2zOA@Jg=T20*kU)jxs+#jd5TTBO3Y z-Vf$!iI_^Sbv58FH_<5N!4(-nNmZSd4DtCh8xjYb**`au&QlGN&1YOS5Yk+2nFX5a z1~x%MU$StHt|Y>C2CWDUosIFssS2-ryBiv6?cNX>bA7B_m@a!P^f-3EOH^z$#p5Vp`;1GbrdSEWKM+SAeQz$p#Fr&1?g#ojQcsl5O z5nk@ts#+k@FZhL8?i~gk##MlKvpOAcXmQqYl{NqtnMxnf;;^_+)~joUx*4#KpuhWb zFBD4LVgB%k7oanyS7L9yptdg>RZ2hrDRAqg7jw|_F({RnM+2z%ECEU{LT9lZ0#we| zz=V->geT%h=QYC-CaWQwXmnAETqPyX#SxFB) zOZc$^bTG_OBIvv2g}%3o)Rn^llfH<)$DlGt9H<~zRq)I#-)ua3xO=n^k$_9i9t*mI zL`%c2<3vSXYrfH?3~%@!$9T3kx=HXaXc5D6u7e!g-*DrX!PQ_t>xDO-S-Jx8jW|HQ zKYkbVH7yml)4R62ZMWW}QGZN41-96wHJ;V)Kz$*)X$kP04+BcHaTHF$3J@6yq7*^m zfT@xj!g&qmALs|?ft^t?m5zsm^Cp;;mmfbr{db1~o~ zbi|IWfF!&44wgxo%}26wo8x?1$j4=b>1UDc0bNf-E&LO=FJY8iT#71z+nwQ{E0^+*11X7eAfGF1 zUU}GrB>qh}EFe_+s)6(EPw+timZ*v%h;!U{S8=aG4Lp#Y0>0QduKAs-FA)GF2?){s zW5U^_<>BTae5Jv|b+B8fKge*}R3SK(nP9@?jt0K*=VXaC!ao76 z--wdv^mt)&mv8iVxv_|@jz;zp+~KhnV{=wA!D1WM+IzUF=~E|%2L@jYR`JjutNOQJ z|G$W0AN~Iax%42&EnKC$x85wO&JIF^02;1*#-eku1v_23HvJXdXPcp~8`c0nZF5@s zTZ5(t4oroOq;MD}>~!T6aWD#y;ix$lbQ7?U1zW^)7n2CCUbGL7MP)!v-pD;hdEH$@ zUGjg1iq#nN?suI@YZ1el&WevVhS$OzAq)l2Eo4cuEzlRwb@ha(+OCF#k^tpK-vtI6 zA{$8iuL=AR0Y+9vDC+)|JBmvQG7V+(noQV#;v?{#Utm7tBM27%nX6XdoaP_tbfCe3 z*GWMmjqB!iNN~C(RU7&D!+^~N1tz(FHJOymIIN#?%o!9*Dm}*!JBgUUQ9d6HhY_OP zVpc$hxrCL+nYVOp*xd-r^c>L*%8Z2dp<|vfHD!HfkaKtocfm%Zparq~17?N7mxk4+ z{LAC3Gyv{^vPFl&Z_gU9?&~TDyndB#xr+bczX{N=@N#3ovyCJh5)oV(9;V#2Me2XO z&Ht~r|3|=O$2I^`f^9DHsRY1UyRSv_p~4&*X-u&S55m*t3itj0X_-{OGL4_2!uC*> zRRSqDk0-^L^ORCA|FTO~5V}#P+@cDN){lL!z`XU&exaE3#V-Oh^l16ix|FUivnI~J zLepz7ZWJo#Ge-O*<&M|&a+b3#wLO)ev)q%O|fiUL`JnJsp9}Ddk z4ZLaTV9Reks^*?Gv%wnQDlu3$jHcy^Z7_LMs7%fZGiMa2uI}4pBP*UjZ002*d^_UT zKDZ6C9^u0-K(=W$g;ezFxMGrP#LQGzkPa-FoeaqPxsR+TF1TqZZ{|j7hIh6iIZJ!{ zDvHBXKm~b^+-oGB{0W6Xb_vAre5e?H-N&`;F9W=V3iT96R+r5F3>2O~4J!04*zNs( zmA<)arVGfz=pJV!1pKcUei+2?k)O9pW-lhmzcu#zVL!Z%^Ii)T!|$~S@LpZ`CifdT zi=*~xIuD66>`rviz5J%LK%R+>pTET#Thx*kJN2CI=dpt+10Z5zFD39NB&oi3gw=Y z@|mFAbC73b|3fbW1^hDd<0yU^Glw5l-=iNoe91N*eR^u_$X&kyy;z0{%`c;1?iXjv zvIz(TqQ^W5xELBF(PLj84D$QT;V+?_9xEyQ|G3ySOq8{Jd+;RaPjV~jT@z3~3o{KU zPpe?q@T=6XMTL_niW%s%xR_YYz%{>Z9U2O|pCCjPul(^cbfV6yG!*+}f#Gs|cx0s7 zGgF=1)J`gl)DTJc7#3w8UG*DShETb$Udn zodfz=5T6jPy$gVig>q=9q|FrjHv5dobW4JDe2wj{dhhtgot^;cCX06Vp^LS1_KAbH z$`#H6T#(*2>W$@e^G`VmciV^6XJ`FnUdmSO984VS%x>N?svmicd~{k(Q@Z=;`g&3E z4$fX#gO%hMUkP+<-EL^822^UO)x$vD^uW+D)~o28h^IHd3P3*oo$=<1@WykUm7GSn zSENt!t7&bZwF)a35kS-ONW>yccBbt9_x&6kR8E(bgtNDDOX&_-8#q|6tWEAA`K!I} ziqj&VbZ_cp0n$2d_Z4dhAM{C4Q!CEq%8GyrvnY5q(VeSDozXV1rRBu>eX)37@;<`* z%bf`;ZAber0yUGv8b&tE2khRQ?8QRcZIK#9dxt~ztpg}_8KE4P+W4cEYOm}19H~yG zR^DRL&b0u3I+UO?zFjf3ZRld&{Zf3!X8Ned<3?_f|C^Tlep6>CVCBIi;x4xQO!+`R zK71r179l;e_~5f#p!}CBLJjER(*tdMc0<*QdZ+8($MQSMS~d&43J;CYYz+IPS4xK) zb{d+Nd8}I<1B(TRNagU+a+@6gA+FiKE9@+WwIlr+@w%V%0!Q6=x5qSoV4#U$r#{e}y&R@Z z=+@R<^L3giX+=5sR)97bRol0ksq?zuWz*=McTQ*ajfS!W#P&Ky-S>O$y&u=3MtQDt zO8i$#|shz>IZG==rQ_?o(pJaz&b4=A7~ z>&YBe_g7~ssid;nJ35L2F3Sod6uIiONB%E+B=^Y$XY*jc-+?g879iYo4~9ocnDm4L z6}p=`L;BK1N3HyR-yJ5)I#CSL|IK7m3H34#83T_9b&>(y$}#)cxot%rn})jpcpn=~ z&VkiNW%`}i&VcuZ>#{^$(*7ByH!Hi7@u$Yu(}Ppxpy~V~D<|RDisv;E-}k7ydo}t7 z0N>HTq-MBXGUlBzss>loRUfS)TTd3_TF)^Y@Rn3Nj6YTHjXKX4w{e=QGlPhQKS9rV zP&)wyL^0~QJm+!!5?VZZ6%8-)@`SC%p1bOCA&LM#O0TM}ayEBIZo1^iP}qDfD?iJ` zMDLt_U1AMd_2Dt>buK$2TK(qdT0Hva5&Koc?$UFqnnZ?}>icfh>jERy)ju0pzg83; z1qF|aE&3`A43SZ2;d^$B`e07hoMIw<+*P0^GfBK>vL1}5D9uMlzWP1wh4VITqjRjw zL3o1&#cLY@1c{cnmPrQy!b-XV{PT_6uVOe+s1&}kz##6`sqs!uRW(#%;k|HAulOwO zHRd!;SwB{Vixov#ArxG`2j~QnWM@!XUN)ejRM9cx=OT2I!qJW3FDPH~5@6iG$l&TH z6ZpnnJRFMc1=^%yFVAN$MFE?TOX_5e`+5(G?w8}%T2M0PgwMe&cykG~pX0xO=(s<4 ziyXB&s(IzfQ{Vl^&|;d5 zww%@Zb_QtOb3rJO2a}`Ae^)M)$HP+JG6=|pHkSbWhG5vJ!Xa*|zkJMeeZ~_&cAE~g z3)B+CdfCLehqDTTT0?Kvasz7cCymc+(;5_}DsZ7+yosCMm zS;g+!OY@vUwFbsG>ofbEpG`B@lGO=@J|R{QPK+l)LJx~n-JOckH?(yC4UC+A=Ye=h zYrjo?(t%8Kc;!gL2NYr3>RrbR0gjg7W@|RksIhsZj%C7?oAks!mnSTfYJ{HUX+1m-PI>6@P!!meoUP1lHY&WNKZgbz`3lGu zci9!#3JZDpawk;QVe8~X9Rry3n%@Ji?TF+Z;9S)&0m9op+LuCcB1m5(Bn%bq%m;Vj>t~VFW#C`%g=YTFt)Y+@ss0ZYHejWV)AVDX*$hFm@XcyC|mvImJ z0r52SEAR^206vg@%a~j09F=kIQ`Xds*nK@x9cjIGAcQ-P^d74f8p`<<97?IlSiQKI zYHi5sWM@rpD;*G@G%#DbZ6I~q*9&o9JqV6%>gNAC{jM0K_DjcVX?za00kWs;u;R^l z%cNrByVb{!5(mbsrZ*beUOhx84(<7}r$H+`PMAR6GP##ua!y32)f=7tOOuh*zP}nN zezdCGnCY^;CKgdDKmj3a`xi_)Fk2;m>>cqtaJ~t5qvLs9*?8hlReN}f;TU4z@+{Z6 zreyN{)a7%8$5~&2*AP9jc<8tD3!R@Pu=B|AsvaYPx}erf>M6zEstK6n1*Pvr5p<1l zpt$P-x@V&J@ktUD^k=)ePmk^?T|&w05jPZF%(^%r?tnR`Nx7Zd8rC32 z6rb5@p1cW@u9@(Y5wdEJ9)H$oy&j-UwJ)Lbeo-bA!0(*5jCzYHN{JbZzG9w)k{egK zUmB@SH2T$o3J6zc=4_|jOk5Te$kDpu3^>n$G|_=P(g+F?J#p5_tB9oza@}W%n4`w*+klV=3PK2H5yxraRwhZ1*-U(w zQ<4Ibhz97{b@-xsaftQqAwcotlVvoPf{HShYG=<2szGWoaXGW=j6Tnh-@KC z(QQ3Z;KZo8Cl%khQ#W&L%mN8;W;sOkX|BL+x1ozpr?qpT|d*^z<;TlsaavXF%Vi9Hg1z{u}<=kV7vH!7XGgRCrv$7wY3AG!l z8cIOlJTdz24lg0TcBWU@qB7N@8C{FwM=5J(C)({XNf>%Pmi`xm3n5$7n@0W^o&bNV z2b~mK9Vl)L;`mDtp484R#y@rgK}BvA=k6B%!7LOi>?M@$^=KelR}bCNqla&Oe3kZ3 zDjx?;eeM9Fs^SLS3r(335G^;N4-xWWvFoZ)bK$unFIfcK?y-#k`fdPvs=6|Czh;bo4fh0_%(RE?Idw7gb{@wnF#2*grT_58- znjh;x`0r=Fk#2?cSnt>jm$}Cwx`P%Tdv8+o3qAeBf*X)50C-0N>btwYpE6}XxSMj+ z!coUpI>UBD#b%n`11N;mPTNg>X#3dDdc@5XWgYL4(l4?pn^M234)vYkQi&2VW_K|O z{z%a{Ec;NU*d#h>IP|)t@j}k8^S!s%A3_8^#v{Jm4>>}l+CPsWi5p?AZ3@nBm{;QU z%zpQ~K|}!wd*H#)oZwq{`htNQSw68`+{N+p6Yen)3q9FJITxIh%4vC>8JUhCKFzoP zwkxLQSlQU&R-c^Qw+f|f`%*;_JmQ_VWl}5;aXkC+-k?wxSEq|$w%#jH`H72O_mh26 zCGO+L9WPB%YMjnu@{*vC0!uH}F*POGAA1#^^$p`jZpix1H>)u>%lqehR|Qxb(gDLw zvRPqO6^aN2Px}3ixK;H$F6T_06+vgX^uc)IUrZG1;v#Y4`O?;{T4~E@Yv=TES6B&P z`?-My+mugr_M18wmpXle3gKb7jK6{4hOjHG%Dheod{!9k7Bqp!2~pAqkoYk!t^E6H zvKtg-CJ2tv_~hh%fYM(mnObjOco9Eu#vPNZ<|0d(SNg&Bq{iBB)HLqU^qql>DL8}c zRhveLU0dHo3m|fS@36*PJ3~S*r&_s3D7=Exo|vPX%Ref2wS3jjT^jt6b-ZIC`$;5+ ztuyz>oSCTvsN(P9d99`|hX_Qy>M()zn|>bZz|$f+B84)-_W{wH8gkoYk()H=3-6Yc zX3~2`^#aRsJtys-5`NnBWgTtorP`elmBW@NJ_}@gupHdOaD;8SW5BsLk7IP z0fEzwrW(DkGj7yVpfOtnH(_5&Siwk7*eNE=^2Ah18tSd0$3UUW4ojvTGx_4@Gur$n zlBCnVY)r#8qQzDASLd3Ukv!l#!C<12)RWz4$XZ8i+J8%`n?p>Vw!Si%b=O(4biYY- zIZ!`$od$B4BWJ9|MsNRPTWhFj=wloSg#q0o&FyLrk=3`>AzOz#r(S)a z!Rk^?j!ahKE*{-2+C`emI^cWFEG;Y|GoZBHWOZzQT_Pj8Gp#Dwi|Xy4o|trCRM0-p z9hQOU_}+sq-D0H;ysf{XmJ&-`QCK}7g_0{!!HFoCWSRlwPRV|2y7uf8@$4E{-qn% zoJUE8*sAe0Nb$fAIP4Eg;9rDB`7faf6FbAsUlcj%5ByB0Al0>{-+)Gw(C-~9OG;re z=NnO98N3=RUzQJTa2XWq{ABa@X7`5zB zSps$y=vAqi3C)YuTg-Jy&AE8Rvp+yvwgEIM2l-7MKU6$l>9$lc6}bi7W$1oSHKEh` z)weHRl*d^2z&!%)p}TP*ZFTZJ=Z%tiq!M+w<;?=&PovPC$duT&wlb}>DumyhH*1*d zgD^JxNk`U0iT=4qabu@K{b+AnyDhr&4CX&O6+voCp9g!*fMe$PV-tN8myAPa{=_Vr zDZ^aN7kwhw?>$%X9ph)vPhxMnTq|zR4^W=Im}gp+D9+Z}U#jMa!%7H~fd?)?mvSSy zN>e&Vmz`a77KDLg!Hd>d`B@6*)Fih>(HtNDU^00`><4GR_il1~!cz66c_mgO&-qC7 zfc@Ep_2f}sb8L6+(2&&vpON$WGxcYFpf@2TUuJEs-0h&(hw~8HT-jh=-<2tCCqL`W8OvHqpLf-Z^y`3$M86$abnVrynOGu zA6ntS(3)Qjw6l7G!Fn(OFQs(8;$E0nRqw**^BcknBfG*hU+#SIQwRx$K_^d3EV~uD z1{5~#c;BCK8>Qsze-c1{uRy!XF*(N6#b=}7!0*Rc#Qu!h7vC01mx{@RlBn(<$*-g% zmJ@n^@_7Yc%UuR(&;9><6_Q&f&QlQx0fJje9Xq{J=9*w3^z@hpL z%Sh*eUH9B3?G|slHz2e{^##?pnf4Z!7iNR*9`)0%(-S-&)j3Qx(u6&P%W_6<)p?)5 z^TYF^UM#48p@;V)<`p>xAq)3=%^n;-iBHp-f7P0*ci`)|Qx>`kjM^>?Ry4dy@a~v0 zHtkG+<}X`mb`0qdmfRB?FNqJXO+Wxr$2OygM_!VO>#vEU$P;1``pPwzr)O|?0N51hxfMGMZTT9rO)4e>9E&SMzacIb-Ynt7%?6YbDms&DGmYSeR@+>asZ96;tk z!ryP+dM9tByH>oDNwx&|4{97{PN~h=|GfE*id8rlZZE)UQnMY z$Sf&&gJ8Gkx5RE5&PLBi1q+{6$WyA|tHtP_t#kSa5s(q-8M_P5lo!+~MR4VcN&B%j zCU>jWq2&G0A3RWOkI=%1*ZbplUgG{D3EuI}4U`_oa=BoWb2RV?qm=c%{I;QM$)@;e zIk5|3sF&q6VY}(rjr2$#0i4zG=Fsd*=P)S-OTdP1e5Q^2tXEGiL{zP3_|fQz&H0m| zhWQzg7L4(jun^rB^I*|y#=JL{M1SzBjCRqSRMxsy%Y3ZHTD|)d4aR4|^;TnHzY4;! z0b7c%&AYtO*^P^5vqnmcZWo_UOJq_mybD4XV$A_@K@vujgG0$OhFxdekqD+ z*gmzvHEE209dR8L!ZTZ&v(MpFV<|$03*&o-U1#4S1T}xKkovq>>h;(mV_IBq4V4Ds z)y+UG>AmT(TInm4HV*(NEZ;$DNl+Xi33tjfXwmO8kzBCdjcTv~Xlsy%$cXZPdTk$D zr$6xiknvub!H`?S=`<2r!|f;EPkzs}e7{S?y-oPCp}r)ou0KJXe-}e-JG72trut;K z0#I#ojUh^6R%7dJRa7u_ZQfJATYKfAq1u}lg?$2uL1sC1R)E5iLqYXYykD)~!Gw%U z{y@ZrpA_|N%CE1r!xLTvG3$HTgt!QHJR8N=ExdOlr(fge;$b{=@a5)SfaA*#4OQ);4Y zJ?T(#-uD6NyOw*B~C@1W$?2>0-a$ExUV|B`4~2Q}f9Y7o?uvxvlS%Sa|N| z<;&puIu(_Q;ff6n=aKsM1W$qI9)6*ytbxgUKkDn;}|f%HSs)gVvL{^|EOUkvPp5M*&# zY@|h*S3xwpD=DJ{K_WLdO!9B)kIj_UsqotL3IT#^XZ(p129Vsi1YI|Jh^@O_@Z>k0= zR3|m!P>8&bECw&ulv-eAyls7So;>H9=$Bua zIb4tEpAnQ6n(}U|Od6hP-;z))&sz5H41(~Dm}YD*g$EcX_Y}}2DGn!)St&^)4GNWQ zwR+-7xh6AuJ;;o8v#5lX_{U-4+>+iFGR2-8H7wI zUjf)8Yv~I2N}gUhVXqp!sJrG-@erurK99ksV3gF`ny2!4~SJ#F_Y@hK(ErDA)$Kpd4z7arAB ze5yQh5TrcfacKobS6=8rpCH`&Vl~A}J#QE|C6QB)#-re;k?VQq9ybRqb2R&kZ8weR z6rcYBeSFRsCJ*)<_>OTp`eXv1*7p-=rYrs`@}@98%Y25ctv^gxq~Mh6)04xTh9 z^-ww6fcn)zm0x&|M@uh2rcr*k&6P};b={tCtuQ9^vns{0Yin~MfjI%~pG~YA23G=U zk~apRrf)H@%CP8A6dyFHgOrT>G|$hMo++eh@`~9oO2C#SO<-5vG@rc2ib_F^^asS$~+EfB0RIq5SZX zzCK6I+7ZaoT6BJ5>r5_`&gnaD%Gsrd`U3g37!*u;=XCZ`CNCXOYnw>`LIFq9Bykd`3$0;zEhJt96}-0ex1r;BuL6e3E#{+P^-GU zIDHtTmQ&BZiS}b6X#L&~#p2P4n%tU7_h6e7hN9qihfmf*wr0pmW`|6nneybZFR^vV zx#Uy)agA#r0+!q+mn6h`geps_j)kU?lM5m`60|&iuot~AG+Ero9qmFy9h9V5(=J$s z!reU>b<9NF=WZ^a45uCEJiTR8xE70xQ0PfayG!r5q_=>vA&Yp@cE0cWTE`_S+M%#rJ!TQ%!^r`g%r zN<;}RPR1B|sKREF$tem3OT&5&_%1DF1#B7B7^o>!O>9k;SBFi*CNx^MayYw_K)KQg zWFwwXfN)(svc_vZdYx0(ci28aHBe7>rt2IsP@JQ=r{ z@nO>v>a*cgXUz@#SE1MQB!?I}wh_JToU3#p2{se+wi!h7NT-lWhv-*B-ck>SR+9`Q zUJ`2x3hDDN{@tNiBd1o^JO$w!@2A0e2X!>Ey*V?GmTahD^OFdnm?K{G*{iOT_b}?E z=FIL!bcp|fU8sY>M~8xfC;9h7rt#ZW@Wy5=5T*iS?i>>jL)_44#`cH>g)6NQFWs7% zJ@NLPq5W$DCyq#2Cyh#e0kKzrE4k}Cy?+j@Sgl#uQ`+HXquhEFE?WOO!$i#>A5`U+ zlkb{O;T&w?M^@>i2H}q4di^C_GSI^CSo+5SG#hJOyul|R>(KSH0jBN1d!5AY$o?E@ zjb7P%t4^Bh7Um8VNlhqeSA7TQ>Sy?`wY3l5BU4Vq!vd2=MM|?B>fr9(#VDOm_3}v)0a0dSCZ5qi1cF${#=~!f7X{Uwm)Ml5tz9A!_ym8Yl1I4a0!kO`Xv*zEo?1p+idQG~P&{GArNBQj8K3YLvX42QqsFDUWi2Mb zY?WGPMyTJ5@co)}mCgV`b^je`rWE-U8+;aqbIp_~$B**5F_cV&F*L47TOr6@yq)ZvPeY=(=nTbp1?%PO zn}=U@i`g0F9!FZKR7g$nvreuC~?okS`ca?tvf#1v|h5#l>1%~?{r zrz`v2aQ@0M`-ry9%Gnt#tt8XfW6ba7M>X?R6^cPK>*xB62p@C@UN z2C)uDM4xc({mpKPz+5b)UK{5ZMYKqLXuxiQY5l}rj017&N^&i(Q(=ZQ7xcl4*RvV+>WGmN}{Ah&YhTpO|AD8geYUopBM6 z)grt%uC^{tI+I#8 z1P7t>X$PvX>{Z-+dFu71%Xz$SC3xIB`T*F*oWVe}0XuFl;JLGoqhtQLhs==p zUj6M_t+90xY;V~`H_gD3VebG$kYy951bOe~Ze6Q-znZUw%G})0(Cd`AozL_`$?|D1 zDHVrMo;#M_$j?~9kURGlW)_c2Pp-)8Xk?rvR|r37C~YnF=k{VhD?3ttx` z#M_@;#mNzFX5?5s3k+k(I?Ns3o#ZL78&>16uS|9iJnUFnT-tuLBT08)bf~c;u`Qx= zNk^=Vk8aE8IwT8S_UadlGtE8eh!VLr3Y~AXlx!^~qW~KU6iXt*is%W)Dx~YY;br*x zu76Uv3cg=Wzh@cUmlY0-tRzOG8#9h2W&74bPj89%WWvs@g+SeN@<7vo(iHO2?plM_ z=GR}M)Ehr)j+%e@tts?|V`tSPb$0hg&dtw_f?lnkXZg9r+LP}JP9WmDegwKDO8geR zmitu3zCExr5)gm|6MA#!P=@Lv4zt32|YlE1Tll%G&F-7@fsNd3)^{$&WuC9B1k!lm~g|?-h zJ8>+1dX93$NEuD(&3;04$C7}s6+t+}KNT$*TTklvR z6`mlB&s?dO?L?F3B001*I?Li_uTL>NPD*qMQzpF{H2U<2k{rP z)%h2o;?CV`Wya+?%du&vQkis`Qf5n?`P0WwRvW*;Y_skRaFf_TZtwY$;Kg}q`}$x8 zKgmR2sVuOtEF!z>?juC$yo4#mCH_1T_48%lT335qgq z@Xrx*YTgt5*tDhZ>Z*pz{UXw9&>zbBp@p{uhxRfzBwwvI@VOg7khVpl6|FrBIj){H zLEr4H$bSr{ZHQ^v?RZ4plg!GZ8l|83U68e$+v{9_ANw$#`|YOHTSpIQF(Q2J*DDuX z8~$=8c7f{sdc{`0w)yxqKrsE5qwa8DjgsnRyNC0(*Gm*N~36alfzcnkG8T*lVhGpR}&F#-&>Bj~i6L zT@XLBOi6GaccjKyzaxT=9XdeQXg^D4l^hg|t}=25ZN!>lE26w(GSZdltT$xU%JIQ} zja~K40f`ip;LuX$Pe6iaP}YvdC5t{O^$?=E#zN{Suv_aW2D??Nq&>p8Yy91kMSFQ| zV+Ea1DKJPadLUKVYNpn4>5(&}&9+DzLw|yuj@?+)Nmj}eYWgFJrTB(0+{SS|Bl4$h z#bRXP(e*Ox&j>EzIdWjqM5RJ?RD+juGmo|l;?B|8afP~+_bDN7dgk(jslZmEShqv4VFRu z(BF#%{G=ehn>3S>U}bVe64tJy{?6~hY zL}{ksL}h?n%uAog)f}jR4csRXBmR5;>!|f_lL7QSEUA9*6i*TE2aLq%eK)bavO~C0 zvebB>H3OxPTqs@+Zu}+=JxZdOG$HqoWOXBu$K735cf6ASRRn^^UM&djNZ1{6nVaar z#uWFcaP~;StG>Sim&C5Tvp<*rbM+<$*lqRO7Zg8~bHZ*HYxKRuL&N+5esG%FA(FI} zTxXghI|t8wH)O`T5C_})DGnRsfFkBs-(NCU8NYuX;IALM2@BD(zb0Z<6Bh_%xUuf% z!}_~Od?$Up1i8w1uOE$R$MExrM}B`x9yuV4i3Nr`>JaTpX*(((3{Hu!ZMEQkic!N8 zEzoGdhp$H-_unMnyT?ecxi`O-Y(jloE=k&+OP>jz7>7oKA5|o0qQi>JcMflZ(^;Xh zXeg;$8VRG55n%c_ej19*M=<}gm!=4plR(*2mVPAUk4^pC?ZaN<+N=yAui2#DZd&t{ zzJ5RR8Kw`(i-tSD(`-T_1z|VvK{pByu#cjP;c%3iG_DMok$qBlI!gci22K$eGyHYU zO)v>JaDRfS)?#qsqDF}@i~jSq|FSAJzJXi8i|0h~>Boh_NO9#roL@@7%EB4Buu$K` z3DIOU_EArN7%3=5uDyygtozpCIR@HCt%0-Og4fH*B2ONdjjld_Oc5qG0rEh%6X2g{ z{`yL0dNZO*O`mJe;q2>XXLKj?L{y2yB z>iTCCjDTi=&k}BKfp!B|P9B)u4e-Utk3HnhTlg*H(~WO=NLKQ|qo1zv^!tKVvA+(c zzgC@`24;oG16OSgPdwX@mYT?)cX`ry_pYhpBWhrD9DMQ}8k zDVJJ)7T@)ua9`^V`XcMV1%FxMRtV%ui<^FjH#M!(%*cPRFgJ4hamxC3-g#80924d( z^rtxPYruwX;?#j@>ctn1Mv0Z9`NNgXfpMEPT*u)@E#a?fY**H!g#%_JP%BI0YALXF z{96w9KLfcLLL`kaSeWGN%W>~^cM|kxo zP5~HF54ILB)ZN^~MTy3=}uY5;L;*9mmd7 z>lx=yc%&U4TztW2K}*3qs5>@duqqk8-wDl90z)}FNdc#X%AJ z@tjs4XJO-o_XJ3h`~vBRD9{G<(5fev>3$GVG(bNyS-TI9jdg!iq~}5*pHU4$5oI5!y(8eQT--tR z$6z@|NUm|xocRFp#9ujC$)7^`YLtLN9V5WdI!1JPJlz9q+L|-aOW#AL3et`Qu=~|L z(jooN<#TnM+(noCWfG7QztwQ+B(XoN6B(8Oa)hWX{!767+6mf#SO80=M)LZttv1j% z%-#s9a5WIBW73W|a7MW*D9fXE5B^o0l>txn)2-S z2K%vsXyN|5Iw+y8Bid1%s_$fEFJ2wEEu8XL-_#_$h`Xbe&V$9KdXCoO1vL0!kn;fyJW_F06usA%WcK#{4vpHZ9J>te@`Xr=M%K( ziVpKZt@0ot_rG_+NB9on9&gUp>nR^jD^LgdvyeoTzbhj-bG!SB&krV7rH$UKwJjcX zgZU@fyXp#P>4ZZtTS;e3{}8-uK^IV^Ya6h?s}ijKzJ?FSNoU0)J4ush6h*F^YxbYK6qGq>{oZa2EriGh~1 zT{LU;Y-T#*WH%)sc_T~t8gL0PK}vfDI>@OlU1Tw-dY@Daa4U?O&raA8WZMz$*pCa2 zyKe|iJriD{k@3TZoVh-|qja(CS@~+`J&kz5kQqX%w4f_leb7G_p~YFtFJ$^^?t1Od zj};G>uYtrG0odb0=N_jL{^c@hX~A-H@g}@>-n_1Wb&+=HnSEQzpRi)nW26e-p+oxH zw;qr0^PKnhec$=x{6Hr5z4zK{uf5iFUF+d<%MLP zIAqaIs8_`c34x1Pn248@BH$QJlvWP8``=y>8&6TnbmXlWpUnV=UX6PpL@#s(b@wbm zI8^?5psgL#nJmJO@mUU#CKZukBp{8b^f{KEYxw(0qM+E1K<@&|9&7`H+JwQ)gqo{6A zf1k7B;)XWkqU{n@81ji66(Sl3gh@%a1mY-bavHD-n1m-?krRGL80&@G{;+pMH}UF6j4zIXW+VR< zb+fXnQ+FZGN}O&+P+l{7W(6ErMxE4U9z!bz{f&*;W_xul+;kQ^MI&%n@S0huO;N0b2_ zv}L>`BysO7u&F_iKLtWcOd;d&&v=U%)(d{+9*YYKf`WcSM-YAO=|rsjJG4TPOk|hH#m0Ud2e_iyYbkPI*Gf?vURl`^0$j zZM9%6#m&&o8g#-?oQefDD~lds{Nm%AGtW3mTc-nW1XOID6rs=gWeeH+ zZ2i}-2)+TFVOxmMDruw{AvN*g8}O^dQJ>2<5Ytc-NB`gcD){HG|2Xj9u6*Nun#iU7 zcRQW3l^-7_s5wz#BcP)@zMMbOboW^R(S17zzDqRQYJ|L+lmOO)EQ@7u(XbO8vNb0Lj6sk zE0|x5S``$Po%e?|LGI*rO32@Fvm+EYpVyT56*nt`Pi4y17zYDARuig}`;1P!Zzm2S#EO$#cv&nv|Lv7^SB* z?CyLTj)e1=V4s#puiJK@fK5>o_bfn0bx?_g^=p-ip*8fiui}T+FiyoB`vC&xi1+ZN zj3du_Z}~gV)0Nfx(p7{3%;dmw@py4m{zUJ8{}j+iZT0%&*3=JhSKb%RBM;D^CXBOb zwif=^Y%_ur0XKi_0~dB=TDQ>`oE%aW5U;HeNM3}cDS^2)qKL%C#=H67bJeQKI3vSR z+A)POR6UT;?Z>!3!N&P?V`h7w;*-xx840ui#O=^duWI-(!2ni=n%MUN_?5wLv&!wK zZT?^XD)i^C{~Gd->pfJV2l{?F9}F@!m4iisuyC#bqp*O@k|VJNC{+t#H_2OtY`eCo z)7q2EQ6P@2gx#K|{&k}Lr`J+f#fh06IyI+>zJPy15+vF*6r!G+Ojvap#6oH3a*Jz$ zD~4{F&%xQDFYmj251EdCiO>Wr*I~QT?5}%*JakVOKs)>j#W%G5vHp1<;-B~Z_HcjZ z(VH!>Ae{^ZB1!FW@TCfbb`0EsecTWgoweTC%a@_!=@!w5+F& z138!fd@VMfH)~HremO#~G5V{A8h@d*rkr`}dZPZyEIZB|Bk+z_jb+!sV*VM!&!33d z$S;%yvF=*t+K(51#v5l&Q$_g=0{a3oKPH!>`wQCYpom;4mrIT00jKg`Lu;31qU^Qr2A%F_L%P7oiG_n~ekl2?2gF z_OmfC*wZHeVgf16;GBH*)?sgss{zgwV7*j}+4g@&Tz^gjd}_ine2x=a83X^p>DvW} zFQD*G%-Bf&11C+B^|!6569yb?FLW&SW@5g9UM$MCA3z-rQw6X+IVcy>j}LF@RXJyu zB38V*=DoY|&FeJ*5`*8et1M^waHz8V*r7`N_oIMV1^BCVTj`G9&lZqOEFdi0F;;Gi z$*ixJC|uoL86I~5B1d{-BH|lgU-7nsvItZ7oh|v#+eSvIn;Co(w>{Pj_1-(B3r^O0 zxm=m%tOO`Fvkb4m947S{J6IY$}ePdYR14t3B~87A*8pNgPISM?E+ZeJSJ+VfsGx zz~k2454iV~K!JvDz0%C+;l0YjnECFhnATWOiu8JZB7CWc?os<+*G?+Nb+6mBxa$#I z0ZS6_@KWoSv+7Si!DT?{VtV ztKDol1sf0?d6A1b`<>Y@ zzt6q&{D^mxmc_V1F~dM2y8V8s{fuA&lesi5IcqkjA-hT0s?APA!+@CT+DQq9`eo3u2r*~{eq;;d1(wC*$w#7$Z3>3yzd(0}7DPKT7oUtj_w&-O;VRI_7y%j}W`##^}ue0l%%1 zx_XPRO|_PjFuQ9I)o6pZGRv5U=GAXgVN~zLX#k}4?@Q>^Sks#vpeac=QU#McjIiUw zrM-er!N%BMyH!;-gBEtVo3S!26p;97s(Eta)f)rTr>RPv+L)&9GG=A+0`uKh6yp`w zG12rd0G1>+rW=hfsmb4>%qY0E)REDot>Hy9fcQ~HqF5c6n&*)#enYr^XPV7_db=v% zTaNDLys)F%dB2>YY)r1nUY5WcPsAc-n|o_M$*~<`oOyv8Mpj-PC)z&e zCSD0}{#F$l-=%-~SP0fbb_4?XVn{(?3sCbzcyvmkKHmT{t+{%09+&iXTK~}y(>Kwb zqd{Gh7{I_D$9Du_^yP$GkAkkc_DDHTJZW+BJOR39EMOGj;s~IP((&TrweL_Lu6bujZK^nk7@>c-_g)TOm-8J$W%0uI<^+C03-m|+*2VXBYa6k-V1SV!= z>C^moGQNxGI0O|lYFCh`)Y+T zl(t9)fMgB5(OsvxfxTUX>lr{1#4;SuoZH(1?4+ZTX}_c0GL9l!$`>~3{IOdt&bf+D zsc}0uoE^h^;z6OG#x$roDSdNN-Fi@>gN{Iv^NRk%??Fh8wWQhJ^9pjJmy!|5Hiaj)iT0?unz!Ln9+)zVou?U z5g|-sik%l!!iDV>Qu<8~hFHVfXA;n?;ck|X#^K>gwY{}yLr&A6dfm_zGkBs+w~9w! zb^bk7qv6??s*`2uH~^Q=1K z`h|4Z+Hx(x%grGYZHpqStE<2JF0LHlVr_6}PlL z2kwm@{Z6k4**)00O-uNL8-_LDXx9Gs5gw31?fLa3yA1BDJTBLj4Rdtm2S_-E7Hf65 z*ksJt%2Zu)1hLh|p5ijrEX4pR`1#CnnM6*XAl5wiymzeB=qG!z7Jv;IzzsWked3JV z6WFJLGJ%~Q&Er8yRp@%y?PV#48y(Ex_x^@S{e;NlRLR4|YoN#60Yi=R5kQqLio{iC zic1Kbft(H!Tb;yJ)ZTS##b)EM#}%jnvf`#C6nPZ&S{enMg7c{Y(`y=u+<9MtFga82 zqleNRrM^bV+iR9}mr;*2b>9tRZO7H`u+px&E$9KI0^ctJZwrfiLD6VN0@G1JkP0=L zqkWbFG!_kksD^d%b6)DWC+TBHnhRfD(dw8zi}yzt_%8@EKB0`Z)7!0mDJ)vYK}j)w z9Dpb1*X66AGd>nLfWjG+Mr1qhObI#VwPYs6jhUMBg%dr#r|j6<9tme5v&IrMo9A#t zheW~?)9h?vCgZev+xbK}4oVg;2Yp%6rI5<7O2pP*wAoxeZ|_crW16RMXRlv5JXlwJ z?_Tzj{j+%z(On^*9m6dGx7$8;?Qx}vnfzY&$M1C5vYL-T0JUNW zgqne9w5sxE=$3Rmy}fRcDB)X;AvywL3T@M8rylU~z3TvYG@$3Gf=Eio$lbV9YVQqD z=&Kho+8OrAAYo|5YlvyeLl9bT9N3y1B?=yocKglqd?+mxH{4CpEjH#B!;~Rc8_rHM z#1#Z?7P!e|-uD6x-R%8!;QfcF({0H!P0&uuC@(+1q)vZpbbDYO&ZTwm?odr-5EI}W z>7O9%vhck@nuG=s)LEN)i!#|&{dxk(34sQQDATP>K*!2yR&M$yj@QS;ywQ-|{B0 zkyG2wolradlunBUV!Cm9EPGX=44v>K)CIp~*%)cIL_W3zzA-rRo{DLVmN5>{DSg|d zR>wsrn_+T6Od&A-@XBKKN)hr=n8-bg?i5V&^3DO^H6S5f<@w&J%)_IJ1md3pmM1vd zp~8qM@{`Ft6P=&HM(iFR4$X6a2hdb8{-jAMVje{#e#?fTqTg4dlz0^UZV!K6KDP(j z(8)>sn87^&S}7*9(fw1&iDm2`5yvfv5qqOsCuqr&JQCG0Tzg4@2sA>#3QKDnk|ur^ zAn!rHUqQern423g1cW*Sy%Z6+sbi&@xao-wRdcarSh;)o1Mt2qMB3up!_$*-c(iqK z9|(e9koLMqQT}k-6Z0NQdn*AzxkZn$wBS)GKstraC~EVEn%@>BZTtXIG};7iv8;u= zmQ{DR>-M(`V#liRr#9<-;(J!>W(!>fnxs~J^qk&jqC1lcH1dNfH3%BK!Pz~CUvHl_ zD3ZEji7J`NNpdLpWCKs0aVlC_vYo364E0_QXxHzrDQs^}E6A$>`AS+X-Iv4k$pEep zvZ@fE{#kpjd_NDZl5;tiKwMfdb9HVrb}t3x%{apAYh~F}#-)DW);pkL_=@cVE;F38 zl(;l1kB7S!H2A{^a*w*;5Hm3^2i6Do?VpB-DRun-tiPF&0ns>${ZV7oV<4Nv*z{VwLg(PZ6 zr~Aj6wHB@nc0XhR>ChRYd5{kMyB-PaC0cZHW;e}kx`PRrIdv)`89KNPwXl0UJ}FKm z5TDmvAqeTGSvRSPOMvWtuYbxO$KVbSgc>nUcV}-VJJ~>hz|H$I^u}U9Rq^A-1rpr9wLh}Hn#U~H;>^MzK^41-6Z07#x z=+T7BEiqZHw1AHTRBMvVyAdF?CA(K?t+uenig7a(u?tzq$Qg9>-lKf?LZI0GtWQa8 ztV-Od>Zt{(0cI|*PVU^AcXC#RZbuJRp_ulCsiwLP_m}4v&&0dd<>`DQ5uDcKT1Y>s zf@?x03ZmiFemZ^ApAJRX1g%ZMKVRyf%3#6U0%=hJhV}u_BI2jpmu1}mQXHcda)BwJ zv#@NieLoWhf}t9E&Bdh5u8DfY-uZj(S|E71@z0cjYPx8gRP#esxxI71pU_tY- z;F7vK-SX7^K5JZ2MN#Ex1=Y9Xquq0*^^&fC)srrr1uk7d(#Ut48~NtlILW-joo02( zQ$c0yeAlxZ;=!85+sU18GwYTnIA>{6RwYvM7Lvq95_MEF5WYOEsvWn)6*Plq;9I4d z6ga*1g(e|Yw6yv3Ujad6P4>l({~k@KTvGp*n_HSqw8!R~pm3A7T-7Z!!V25fpGwS| zjHc)_1e>$GcRGtg$`^L0Qwdy!8+PH{>hKfiw3_DK()TI<2w7 zetb2;%D#FJ7B5=(r>wG++81DyvRVOpzvS^#%#um_$}3 zkhf0Q1?oR44j$zX>p&4_yWPWix{{};N3RcIoG*9v$uq0s`zqdrdn|3yqWmkZT$&={ z;yKuA7`=ie5JqOB&dSl|PmDtRN$9H=&IA^|^z6J=SeS7=$Ndd+Cc!taCXy??BHm_L zEY$<+IYMiEX3W;hZBx(>6H{aUiqaiuThO~dC6YV$c+qS*dA{lgK3{!Nk%$P=BZ30q zH2iI^?%^fnWz}V22ep@W#ysv!*SIsYjVtQ*<9)MYGwX!TE#WBGZx80Ac&y6_6-R_t zZ$uJx)*uNa&vlFTzhRy}t!ed5!Sqo%A=+;H{MdLcoGv}&=4WfPs)m~*y5O#c^dsoM z&nUWe56O>GSL^sDidTJY=A>{6#ntu-%76TbcFGUPtP?pTh7yZ zohx7ZATy~@D7{jC*&5!ct!qg}w&%xrvA`fw$L_;mM?Zna7oQZ>5c`vlU6>`(tj|f6 z@i6T}at#lWuH8Fvu`uQ_`U{oZ!Y<|(-rd#HWLdw zPfk{%3%bGmq_cf(T4VCSPO~jd>mRc#&gCJ{nTz6vvY&PX=dUi|WN~(0Jg5-^+rA zX16g0)ZL(GuzI|bBa@pZh3WLab0dwG*qElJtS8?-iy)HpByGw26@aLwlaQG2P-7+m zY-MEyt7j3iK=)rXIctfUz%~JSHv@U2xOhxyvGG0Q6RkVvK?ts>YM^#2t#e zyt5ux3o*2~s`es|zcEx?x3{&F-bC#|8IBAx6)3%M3@E@Js+-ADjGEmd{re))s^ZZ! z$D4bEm>K5o*^LMkg0;fFT+g>4Z`YHVr+4=}rWtw5t%W*z@Iby<+9P%`2*K*|DM?7a zqunya-x{uSv`7Iy(qtIEJ}ODWY1Af!f}V&*rF`ZEFCm8?X9I(!P6n_WJLw zfJo!Oa~-Mn^B*?$;nFQyEm0W*#JXf}G|FAkcpFr(Su4fHdOjgsBd@B^bh^Y1_6kml z6umkJT4@(z$3YTS{a99Vb(}StCTd&x38aYrEQ@}$+Irbvl`9}OB>JVJjgMNMh;6iH zl`$<^R5v}Vb21?`Vz0g6YF(VnP(5Ncp{3uYu#^xvgD3fPX1t5|IQ!=>^!{0MCF>b`vQwrxi@)ME}F6I$qTC2+`}U z2}iYP6GheJ+qd+}{MdSAqX3UZxVM|Ny5~Ec$3*0vZUP)X02AC*igo_G)SAYD?>Z9G zRhj2<+on;SeRsGDkt?bw$KYF5yVcH>v8^Ei6f!F+Q}^EVOp%WzrDuwWqqYRX`7JGl z#PLxpCB8!`^j&ZH3T?y1_ZBz$%sXP;JiEFn*V5h*+^r$@uRqxvZ=3&N7JH*MnD*po zKdytP=1oUIIeHOLxM-6cA#N7b5x}8l;WnluqCSyH8WH~fwfMpSJL}D{e$$6`ir#Azc~=L`pur`Mw6s$5vp-kUc$r6>XSzOR6U*JcVC8YkZ;7{=p_Q zw)(m9kEy8Hg+uc~)s?I$5CZKhLxpsg+WYI zSnuY@X|VTJ)@Tpb5}0-ivL%LxV@g>gQC-tsAb099dc&v@(MIIo6@Tkmfo=7Tvhc@m z3$&B8(JEy*Ct*fhMPldN%J<5!tlobXnZ#J4D&IZ)z97>axhY?FV5So&Tu@_cs(|}8 zPYRu;ca=?(IKe+pqRjVYY@s`;x{<7b=Htu+lL2JsLlKKGbq(wDkM-${rls?mK6_?0 zxnv&J_sq^uDj*`OV^UV3eBqIO{6hSi7Qop%-!Q{X$g@{$i>@#*Yxg(@%N*~#jF(kH z?u3)(%`b}f%)!c4EzZQ&P|-xImUoXHTiMu@q@rwW_|DFRKr0c0*Q%_UUZr$lsVbA- zrJS(kp8aN6G9qE~YAFfM&Y<_ruQ~;SgED~OWKk{y5d5rRU3F2_N0;0lBoIgnTBr3Y z2v0Z#mU2QVUDPz6S7Uj4z#WnZn?Bik*Xy;QL&zdYitPSzbUSe2ovtsSOq4 zI{Yb}U(@!yUHibTVBXGR_l{-P)ltORiM_FXCV>qcm*87ECKuuko7jwy% z{$zimjB5CrmuiV7A5Oxhzd{C!)oc-OrjE`mtr~O7?ikU%sPsJJewKgEvX({DanZBP zEoEHh#`WI9RK;z}GYU*EBCq)y$|di)wVIBa;*uLiq(xPMXt8!N^Im!Vxm3{j{Iz|m zB%)rtDD{yA5Ouz#wk&nBrGcA@^3_{Zmd=uT;5UpWE?NJLTK{`mL-b+6kGp{YaK-)L zCR?n{`lB@@Xx1?{MLg!R4qXc=cWl{E=w%=f%d0%q-*k>4d386hxqX-HDSWM}rHeDa z?wVk$2k5Btg5WgyIc^a^K0fIVh`I45j&^K(yJFlyiAQ+@2q((bgN9z?UQ%w0CsC|Q z-y!)#C=O$5jhf6LOi28G9FVo(P`$(^hG&S|q!d?CB$Kp6vww@EQ?yQga2_ZTBVr*f zP*th4I!O?iJ{bjv&m6cal*eIA=WR}YX1$(bB-qH`>=3wpF0e}>TFuGMWbx9jATBeGLj_Uy10K<#^QIFKYMMz*AKQ@)-&tp5P4LK>Eq}+wrkib5N{qFS)j%vc~mIp+- z$^;JQ%x$Uxlz8sxd7ihaE)xq--tEXse!*avHyOpWlPHJ2EPAwBYCiknxx1^f4QkDK zOOns^KfbJxjg2D*`x(YW4nxv!jObMzrx_+5JjE|2It!F;~V= zUY~dLH4ccZ*4tqtVizPzPVL8=`6}G9Wrk|6znl+Ep(Pv6+0W_s63bk<>`kBN;uWOv zv~V+Lx36T+t$)R(oG96kg}cCHqf5@lhJ*OsNkA%G|h zQNLdI8SlPZym;OnygOXENd_SaReZvAr?!;IK1LPXCk1wTp5Y7k1PS#bh1AeQ@FST1#75wC5y92SdK1m{e&2dtN4XjiNq3$%85UV`j!3_h zUdAjx(!Opk(R^9lXi<}D0^Z#YHlrWi;;EII3jkGUCifh|6m~Hky%C2w)*o2Z6lQ9< zGBtu!vl688d31CF%9?>E$yr|2`hWFm-V`GQyc5%@&X>l9-Pyku9sy8RdW4I|NKsd# z1>8aZSx}9w7G#exE^ev=l?@rtWX!Ud;a0{?X#+Q7Zo0#w?Km4=CCsP9j)92@JGkvK zgUXm~MJ9fzXo05WQyx$mOAKx{8wJC{FvyE@7L$m@1b($z)#`05- zQJ<=e(2b~(OOETkWy)RH{Jd6!?@#)1^tzUuz1M^-617NIPYvU;Ez-`4?Y-yjfmBMQ zcK^BVL$W*?5pM4F9AQ6p?v+cd;ku#8Jb{*|go$Ai3@f7gAapvhaP>5qOouGW%;Z>& zNT=+as+&W%O7SFfq)Lalxkm-R=U$bDppT-ZUi~qj2*-;qxCFUeN{#y6RTqKXPrH8)mqv5Hv0^Q2}vf+y~G*oUm zB#PV$mje(3Rb6h{9w*1hL|1)UOMUu%krTTN7o?a=d`%ChVFdx}9iQrF=NwYs)o%2lSf|Qv z67yEk^V?8rK020cVc81%TOv)`Uq8&f!@NBpcne|$pjX(~z&2(%Uf zJ_p*B?E(K$XmMNZbGv-Pxgus%r6B_i7Sv=6fK~5-E@O7Vc^dnsfUQT}0|f0S8CwD!Go=Yk5a zIzKwM>slgBnP9*WpwDLnXzZ&$b~Im&De>QXd$o1}Azya|zy;g5GG0BZLKYoL0N!r4 zb8FaIv-sFr!?%DO;d!mmy;GA;o3-_0goavUNZ$fpZ48Awgl5Yq3v8@{9yYT0k?o0o{+DuFNv+3;EmCl ztJj9!w{>MLkb$U5MyxV=<9Z^jO&_Tg1E!Qm4i-O$t($_1U&hE2~ zNH7<`YWDC7N6ipRl_!2V=e8|gIvCiVg)W6`R$OpN0lfwYqpJawN&|s!E%|qD2Dl|1 z9QI_Vryrf)%721d=+!LS>%>HHqH$<)duPA-p<%M2&pI=O6>!tXY_)czmP-*h>3-5O z41VmG*((ACrRXB^X_R;Q@WcACag=)`XmLm0`W$?VA%_ORpD*b>4`pjov=`uGGa(@% zDaP&p!*uW=a442*zRW*3!PbUGP&Z?(R&;Sh7Clh=gM;Uw!{)|G%ih-S)%d+vu>y<< zlR*Ss_dZeS57b#voAE(D33?LZvLJ2|3fz0><#>^#Ya$;iLz3!`%N(K)X+< zrTty9mV%fd&5E&IA3}_r)aOfT98~!{5M$fK7c|q!Kbme`KfwRPqR%84yc(9 z%MQ@WZF6j3gcHX1*Z$3OB4!~cXNN86;wEq8cR$f-NK~tW;F87Xi9e`j1K{OlLNbsR zJW$Ue{Rzq>d;y;2>03?CpJ?ds(4$I3=ix%2Od?{yK`CQg)e3|l_q z#u^TU(3sC%1rHa7*9{>|iPhb~)c^A~o!@W!Z8v|je4y?40+t~&roexZg8;n!)5i54 zf`6>~_sy9@FW|HMjdlC*&x8TV7^3UXL@yvW=7G$~N?CxVDd$Wh;$Av?~aJ4V!2YAHHz|#MJ zR)p~PiU1xfH8xgUk2N+8^5$&lbtIPgwplyvc_R{#4gW>_V`j(CZw9s*;4j~E=U7=kI41w=*J zvM@2#J7A8?eow4ue%j-XDzL}(}V4l>%%9!{T#=|By!ruL9rmg^5iT;)W1Cfq2bD!+}E(UYgh+Mhb>EPmauh z$OCtL2mw7L_qixJ2ef79HbtE;|J#Q!2FyRWu`by1RR*>lruv+fi=1aPw%-Zad!ACA z5HJ63`j8GD4#iC}Nn~ErB?i(1gYeXSz&ieJ0H&$_Qy$qr#_@Y9+Z7v$ylv$97Co@a z3Vqng^>rd9F5pN263Yk$ z8|&|%U_|FXQoO9Klt}p|T4^`R3_v6i+66joJ4^lMV`bJ97xs##4FzaoKnvM1xkryO zl~JPFpbZl|!~Dng-O*sA??zw=L^fj~7C|Y5ci7M+6!bV_ zWhEe?VTm*Gt1|SQectwHa?}%qQQrZkBoZpjxr{aj;Lbs9ReeDBx>$9N?;-2-4_c%Br5JHyD*|ww7(t-ZRs-QXHH?h7c^i&PAdIOqXcE*4)GRnX zCf#%^+gbT>!~w-eqPARo}1kkHn5Qfv@;B>#i zvDkwwyWv}y*muxKqh!~*3PNp*afP`MewRt}K#F|NBBDYD)KRHKIAVi4k2TQa+yx+w z5uoIH1K_vbIJ86B!KtMOn~Ry@ikg~8)QV>CU=0w;IsxiZoKQuXdl_I{yK>~W7<+7i zvDUF=VL>J+REFnD+6#d3~GM5$F*-k)qw`{b2<2kSV?PRZ+ zO+Y9!hh!?n;)Ig1neR3ptvB!{3E7u~N`Rn+U+6SKn)F>ux!l4}LaKszuJnT2r+r$- z+FQF?&WKMT)b4Q55xnh@9peldl9h1djJpqE{_h3=!(uZK1jXX-74S!m{@5DY+U8`| zjRNCY=joSh))V>w_VMGD!+HQ?wQT=z3(uJ+qKenKs~{crpXWe8Jx7H6N7k{oqZ8g4 zyi*eUK!5Vxa=PsF771I#42iY>i+l0m+~@ z!HqS=Dw9KI`w(-^Jd|EszaDfo$+j?Kz%~KS`0#^b`v2JEKd#N}FMl{KOVl_j0WkMa z0l_lTQcMTw1l*da>W^_IMd-Ry%$AXiAW~W${+#uB zPIu7O7)8L3m`res_ZCus+uIO7zdQQ4JD#0ByyN+00CKr*W?fzL*%8vyN! zuW3@Lf$8P;&4c@+>sVgLv**0u+;m@kzy~jc=%)DA;Xmo7>x=s0MdvwC&h4CAJO zcDhDxF;-ZhSCa&O^**m?Dt9IiU>i{ho8Tp5S+P(=g68&$_V)Ily}c>1j2isvuKMou zZh_W{uet>Do2LMFy7Tp}6FIgdaNUP_N~6$Wapxk$S+!3r>P+Fp%HMuwLdPrKdif20N~)GE5E}|O4roF5jQ{e) zRN;*vm1#-q0uqT9b_2Cu+pAA7ReE}QKNHrT2F;IJkZZ%VuR?fLNCU6Lv>YH61+eSVEtXU8lR0!!l+xxFY1Xo1`Vmubc zCfFJdGJ(8|w4v`Gsw%|nZ2$7=4^@?z0t!Ix@?KJr{TYv}fltMxT(0_gViP?B4kxzf zm8>Qt41zeDoUdgsnURdIE>^ym&CG3LL8wt&@{qaa z^tFjvmXbPDM1)3Ned}CYV0+ZZMad{Uk&?e)6Zf$rec`PJ7KbJc-dg?Q{rrv=`S8Q8 z&V`hIWjD9Mx=IO;zv9;?@&ThWV3dduX1OqDg;U9ziT#8rF5@Hi?F)okQV+4P*zgr( iq_ir>K4N3DQJtcA^zT0@kH!N3DafkI6x}s_`hNgCs0Yyi diff --git a/docs/articles/simulate_noncanon.html b/docs/articles/simulate_noncanon.html index de71337..5461928 100644 --- a/docs/articles/simulate_noncanon.html +++ b/docs/articles/simulate_noncanon.html @@ -8,8 +8,8 @@ Simulation with non-canonical matrices • mashr - - + + -

    +
    -
    +

    Simulation with non-canonical matrices

    Matthew Stephens

    -

    2018-11-08

    - - Source: vignettes/simulate_noncanon.Rmd - - -
    +

    2019-03-11

    +
    +

    Goal

    @@ -120,6 +113,7 @@

    print(get_loglik(m.c.ed),digits = 10) print(get_loglik(m.true),digits = 10)

    The log-likelihood is much better from data-driven than canonical covariances. This is good! Indeed, here the data-driven fit is very slightly better fit than the true matrices, but only very slightly.

    +
    @@ -149,7 +143,5 @@

    - - diff --git a/docs/authors.html b/docs/authors.html index 9d00aed..b0fa702 100644 --- a/docs/authors.html +++ b/docs/authors.html @@ -6,7 +6,7 @@ -Authors • mashr +Citation and Authors • mashr @@ -21,19 +21,13 @@ - - - + - - - - - - - + + + @@ -46,7 +40,7 @@ -
    +
    @@ -91,8 +80,26 @@ -
    +
    + +
    +

    Citation

    +
    + +

    Urbut S, Wang G, Carbonetto P and Stephens M (2019). +“Flexible statistical methods for estimating and testing effects in genomic studies with multiple conditions.” +Nature Genetics, 51(1), pp. 187–195. +

    + 
    @Article{,
+  title = {Flexible statistical methods for estimating and testing effects in genomic studies with multiple conditions},
+  author = {Sarah Urbut and Gao Wang and Peter Carbonetto and Matthew Stephens},
+  journal = {Nature Genetics},
+  volume = {51},
+  number = {1},
+  pages = {187--195},
+  year = {2019},
+}

    Authors

    @@ -137,8 +144,5 @@

    Authors

    - - - diff --git a/docs/index.html b/docs/index.html index dabd9ac..03ff313 100644 --- a/docs/index.html +++ b/docs/index.html @@ -8,8 +8,8 @@ Multivariate Adaptive Shrinkage • mashr - - + + - - - + - - - + + - - - @@ -60,12 +54,8 @@ - - mashr - 0.2.18.476 - + mashr
    -
    @@ -95,25 +84,21 @@ -
    +

    Compute matrix of conditional likelihoods.

    - Source: R/likelihoods_scaleddata.R -
    -

    computes matrix of condition likelihoods for each of J rows of Bhat for each of P prior covariances.

    - 
    calc_lik_matrix(data, Ulist, log = FALSE, mc.cores = 1,
   algorithm.version = c("Rcpp", "R"))
    -

    Arguments

    +

    Arguments

    @@ -172,8 +157,5 @@

    Contents

    - - - diff --git a/docs/reference/calc_lik_matrix_common_cov.html b/docs/reference/calc_lik_matrix_common_cov.html index 29d0186..dab6173 100644 --- a/docs/reference/calc_lik_matrix_common_cov.html +++ b/docs/reference/calc_lik_matrix_common_cov.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    calc_lik_matrix_common_cov

    - Source: R/likelihoods_scaleddata.R -
    -

    computes matrix of likelihoods for each of J rows of Bhat for each of P prior covariances; special case when standard errors and variances are all same across j;

    - 
    calc_lik_matrix_common_cov(data, Ulist, log = FALSE)
    -

    Arguments

    +

    Arguments

    @@ -163,8 +148,5 @@

    Contents

    - - - diff --git a/docs/reference/calc_lik_vector.html b/docs/reference/calc_lik_vector.html index 5f2c248..69edad2 100644 --- a/docs/reference/calc_lik_vector.html +++ b/docs/reference/calc_lik_vector.html @@ -21,23 +21,17 @@ - - - + - - - + + - - - @@ -60,12 +54,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -95,24 +84,20 @@ -
    +

    Compute conditional likelihoods for bhat vector.

    - Source: R/likelihoods_scaleddata.R -
    -

    Computes vector of likelihoods for bhat for each of P prior covariances.

    - 
    calc_lik_vector(bhat, V, Ulist, log = FALSE)
    -

    Arguments

    +

    Arguments

    @@ -164,8 +149,5 @@

    Contents

    - - - diff --git a/docs/reference/calc_relative_lik_matrix.html b/docs/reference/calc_relative_lik_matrix.html index 1b3fc65..8cfc8f2 100644 --- a/docs/reference/calc_relative_lik_matrix.html +++ b/docs/reference/calc_relative_lik_matrix.html @@ -21,23 +21,17 @@ - - - + - - - + + - - - @@ -60,12 +54,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -95,24 +84,20 @@ -
    +

    Calculate matrix of relative likelihoods.

    - Source: R/likelihoods_scaleddata.R -
    -

    Computes matrix of relative likelihoods for each of J rows of Bhat for each of P prior covariances.

    - 
    calc_relative_lik_matrix(data, Ulist, algorithm.version = c("Rcpp", "R"))
    -

    Arguments

    +

    Arguments

    @@ -169,8 +154,5 @@

    Contents

    - - - diff --git a/docs/reference/compute_alt_loglik_from_matrix_and_pi.html b/docs/reference/compute_alt_loglik_from_matrix_and_pi.html index 4c9476e..ad4c1b2 100644 --- a/docs/reference/compute_alt_loglik_from_matrix_and_pi.html +++ b/docs/reference/compute_alt_loglik_from_matrix_and_pi.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    Compute vector of alternative loglikelihoods from a matrix of log-likelihoods and fitted pi

    - Source: R/likelihoods_origdata.R -
    -

    Compute vector of alternative loglikelihoods from a matrix of log-likelihoods and fitted pi

    - 
    compute_alt_loglik_from_matrix_and_pi(pi_s, lm, Shat_alpha)
    -

    Arguments

    +

    Arguments

    @@ -151,8 +136,5 @@

    Contents

    - - - diff --git a/docs/reference/compute_loglik_from_matrix_and_pi.html b/docs/reference/compute_loglik_from_matrix_and_pi.html index 4b40dcf..acc18cf 100644 --- a/docs/reference/compute_loglik_from_matrix_and_pi.html +++ b/docs/reference/compute_loglik_from_matrix_and_pi.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    Compute total loglikelihood from a matrix of log-likelihoods and fitted pi

    - Source: R/likelihoods_origdata.R -
    -

    Compute total loglikelihood from a matrix of log-likelihoods and fitted pi

    - 
    compute_loglik_from_matrix_and_pi(pi_s, lm, Shat_alpha)
    -

    Arguments

    +

    Arguments

    @@ -150,8 +135,5 @@

    Contents

    - - - diff --git a/docs/reference/compute_null_loglik_from_matrix.html b/docs/reference/compute_null_loglik_from_matrix.html index bf6a752..123a174 100644 --- a/docs/reference/compute_null_loglik_from_matrix.html +++ b/docs/reference/compute_null_loglik_from_matrix.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    Compute vector of null loglikelihoods from a matrix of log-likelihoods

    - Source: R/likelihoods_origdata.R -
    -

    Compute vector of null loglikelihoods from a matrix of log-likelihoods

    - 
    compute_null_loglik_from_matrix(lm, Shat_alpha)
    -

    Arguments

    +

    Arguments

    @@ -147,8 +132,5 @@

    Contents

    - - - diff --git a/docs/reference/compute_posterior_matrices.html b/docs/reference/compute_posterior_matrices.html index e4695b9..a1bd761 100644 --- a/docs/reference/compute_posterior_matrices.html +++ b/docs/reference/compute_posterior_matrices.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,25 +83,21 @@ -
    +

    Compute posterior matrices.

    - Source: R/posterior.R -
    -

    More detailed description of function goes here.

    - 
    compute_posterior_matrices(data, Ulist, posterior_weights,
   algorithm.version = c("Rcpp", "R"), A = NULL,
   output_posterior_cov = FALSE, posterior_samples = 0, seed = 123)
    -

    Arguments

    +

    Arguments

    @@ -199,8 +184,5 @@

    Contents

    - - - diff --git a/docs/reference/compute_posterior_matrices_common_cov_R.html b/docs/reference/compute_posterior_matrices_common_cov_R.html index 8c0350c..5835da4 100644 --- a/docs/reference/compute_posterior_matrices_common_cov_R.html +++ b/docs/reference/compute_posterior_matrices_common_cov_R.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,24 +83,20 @@ -
    +

    Compute posterior matrices (when error covariance V_j is equal for all observations j)

    - Source: R/posterior_common_cov.R -
    -

    Computes posterior matrices without allocating huge memory

    - 
    compute_posterior_matrices_common_cov_R(data, A, Ulist, posterior_weights,
   output_posterior_cov = FALSE, posterior_samples = 0, seed = 123)
    -

    Arguments

    +

    Arguments

    @@ -179,8 +164,5 @@

    Contents

    - - - diff --git a/docs/reference/compute_posterior_matrices_general_R.html b/docs/reference/compute_posterior_matrices_general_R.html index 49bac3f..b83c1f0 100644 --- a/docs/reference/compute_posterior_matrices_general_R.html +++ b/docs/reference/compute_posterior_matrices_general_R.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,24 +83,20 @@ -
    +

    Compute posterior matrices (general version)

    - Source: R/posterior_lowmem.R -
    -

    Computes posterior matrices without allocating huge memory

    - 
    compute_posterior_matrices_general_R(data, A, Ulist, posterior_weights,
   output_posterior_cov = FALSE, posterior_samples = 0, seed = 123)
    -

    Arguments

    +

    Arguments

    @@ -179,8 +164,5 @@

    Contents

    - - - diff --git a/docs/reference/compute_posterior_weights.html b/docs/reference/compute_posterior_weights.html index f9dbf37..0e34c21 100644 --- a/docs/reference/compute_posterior_weights.html +++ b/docs/reference/compute_posterior_weights.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    compute posterior probabilities

    - Source: R/posterior.R -
    -

    computes posterior probabilities that each effect came from each component

    - 
    compute_posterior_weights(pi, lik_mat)
    -

    Arguments

    +

    Arguments

    @@ -152,8 +137,5 @@

    Contents

    - - - diff --git a/docs/reference/compute_vloglik_from_matrix_and_pi.html b/docs/reference/compute_vloglik_from_matrix_and_pi.html index 937c936..f2eda31 100644 --- a/docs/reference/compute_vloglik_from_matrix_and_pi.html +++ b/docs/reference/compute_vloglik_from_matrix_and_pi.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    Compute vector of loglikelihoods from a matrix of log-likelihoods and fitted pi

    - Source: R/likelihoods_origdata.R -
    -

    Compute vector of loglikelihoods from a matrix of log-likelihoods and fitted pi

    - 
    compute_vloglik_from_matrix_and_pi(pi_s, lm, Shat_alpha)
    -

    Arguments

    +

    Arguments

    @@ -150,8 +135,5 @@

    Contents

    - - - diff --git a/docs/reference/contrast_matrix.html b/docs/reference/contrast_matrix.html index a9554d2..a268148 100644 --- a/docs/reference/contrast_matrix.html +++ b/docs/reference/contrast_matrix.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    Create contrast matrix

    - Source: R/set_data.R -
    -

    Create contrast matrix

    - 
    contrast_matrix(R, ref, name = 1:R)
    -

    Arguments

    +

    Arguments

    @@ -150,8 +135,5 @@

    Contents

    - - - diff --git a/docs/reference/cov_all_zeros.html b/docs/reference/cov_all_zeros.html index c905c2c..828dca4 100644 --- a/docs/reference/cov_all_zeros.html +++ b/docs/reference/cov_all_zeros.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    Compute an R by R matrix of all 0s

    - Source: R/compute_covs.R -
    -

    Compute an R by R matrix of all 0s

    - 
    cov_all_zeros(data)
    -

    Arguments

    +

    Arguments

    @@ -148,8 +133,5 @@

    Contents

    - - - diff --git a/docs/reference/cov_canonical.html b/docs/reference/cov_canonical.html index 5e02555..f69abd2 100644 --- a/docs/reference/cov_canonical.html +++ b/docs/reference/cov_canonical.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,24 +83,20 @@ -
    +

    Compute a list of canonical covariance matrices

    - Source: R/compute_covs.R -
    -

    Compute a list of canonical covariance matrices

    - 
    cov_canonical(data, cov_methods = c("identity", "singletons",
   "equal_effects", "simple_het"))
    -

    Arguments

    +

    Arguments

    @@ -226,8 +211,5 @@

    Contents

    - - - diff --git a/docs/reference/cov_ed.html b/docs/reference/cov_ed.html index b66588f..f5ec68d 100644 --- a/docs/reference/cov_ed.html +++ b/docs/reference/cov_ed.html @@ -21,22 +21,16 @@ - - - + - - - + + - + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    Perform "extreme deconvolution" (Bovy et al) on a subset of the data

    - Source: R/data2cov.R -
    -

    Perform "extreme deconvolution" (Bovy et al) on a subset of the data

    - 
    cov_ed(data, Ulist_init, subset = NULL, ...)
    -

    Arguments

    +

    Arguments

    @@ -161,8 +146,5 @@

    Contents

    - - - diff --git a/docs/reference/cov_equal_effects.html b/docs/reference/cov_equal_effects.html index 373b40b..66d0433 100644 --- a/docs/reference/cov_equal_effects.html +++ b/docs/reference/cov_equal_effects.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    Compute an R by R matrix of all 1s

    - Source: R/compute_covs.R -
    -

    Compute an R by R matrix of all 1s

    - 
    cov_equal_effects(data)
    -

    Arguments

    +

    Arguments

    @@ -148,8 +133,5 @@

    Contents

    - - - diff --git a/docs/reference/cov_first_singleton.html b/docs/reference/cov_first_singleton.html index d0e5694..515d80e 100644 --- a/docs/reference/cov_first_singleton.html +++ b/docs/reference/cov_first_singleton.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    Compute all the singleton matrices corresponding to condition-specific effects in first condition only; used for testing purposes

    - Source: R/compute_covs.R -
    -

    Compute all the singleton matrices corresponding to condition-specific effects in first condition only; used for testing purposes

    - 
    cov_first_singleton(data)
    -

    Arguments

    +

    Arguments

    @@ -148,8 +133,5 @@

    Contents

    - - - diff --git a/docs/reference/cov_from_factors.html b/docs/reference/cov_from_factors.html index 48180ef..9b58a1b 100644 --- a/docs/reference/cov_from_factors.html +++ b/docs/reference/cov_from_factors.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    produce list of rank 1 covariance matrices corresponding to rows of f

    - Source: R/data2cov.R -
    -

    produce list of rank 1 covariance matrices corresponding to rows of f

    - 
    cov_from_factors(f, name)
    -

    Arguments

    +

    Arguments

    @@ -152,8 +137,5 @@

    Contents

    - - - diff --git a/docs/reference/cov_pca.html b/docs/reference/cov_pca.html index 14a6466..507e98c 100644 --- a/docs/reference/cov_pca.html +++ b/docs/reference/cov_pca.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    Perform PCA on data and return list of candidate covariance matrices

    - Source: R/data2cov.R -
    -

    Perform PCA on data and return list of candidate covariance matrices

    - 
    cov_pca(data, npc, subset = NULL)
    -

    Arguments

    +

    Arguments

    @@ -157,8 +142,5 @@

    Contents

    - - - diff --git a/docs/reference/cov_simple_het.html b/docs/reference/cov_simple_het.html index 74bb848..cba872b 100644 --- a/docs/reference/cov_simple_het.html +++ b/docs/reference/cov_simple_het.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    Compute covariance matrices with diagonal element 1 and off-diagonal element corr

    - Source: R/compute_covs.R -
    -

    Compute covariance matrices with diagonal element 1 and off-diagonal element corr

    - 
    cov_simple_het(data, corr = c(0.25, 0.5, 0.75))
    -

    Arguments

    +

    Arguments

    @@ -152,8 +137,5 @@

    Contents

    - - - diff --git a/docs/reference/cov_udi.html b/docs/reference/cov_udi.html index 7fe6d5e..2693382 100644 --- a/docs/reference/cov_udi.html +++ b/docs/reference/cov_udi.html @@ -22,24 +22,18 @@ - - - + - - - + + +"Unassociated", "Directly associated" and "Indirectly associated" models" /> - - - @@ -62,12 +56,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -97,25 +86,21 @@ -
    +

    Compute a list of covariance matrices corresponding to the "Unassociated", "Directly associated" and "Indirectly associated" models

    - Source: R/cov_udi.R -
    -

    Compute a list of covariance matrices corresponding to the "Unassociated", "Directly associated" and "Indirectly associated" models

    - 
    cov_udi(data, model = udi_model_matrix(n_conditions(data)))
    -

    Arguments

    +

    Arguments

    @@ -168,8 +153,5 @@

    Contents

    - - - diff --git a/docs/reference/cov_udi_single.html b/docs/reference/cov_udi_single.html index 48cb614..e94ee95 100644 --- a/docs/reference/cov_udi_single.html +++ b/docs/reference/cov_udi_single.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    computes the covariance matrix for a single UDI model

    - Source: R/cov_udi.R -
    -

    computes the covariance matrix for a single UDI model

    - 
    cov_udi_single(data, model)
    -

    Arguments

    +

    Arguments

    @@ -147,8 +132,5 @@

    Contents

    - - - diff --git a/docs/reference/ed_wrapper.html b/docs/reference/ed_wrapper.html index a3baf5f..3af43d4 100644 --- a/docs/reference/ed_wrapper.html +++ b/docs/reference/ed_wrapper.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    Fit extreme deconvolution to mash data

    - Source: R/ed.R -
    -

    Fit extreme deconvolution to mash data

    - 
    ed_wrapper(data, Ulist_init, subset = NULL, ...)
    -

    Arguments

    +

    Arguments

    @@ -167,8 +152,5 @@

    Contents

    - - - diff --git a/docs/reference/estimate_null_correlation.html b/docs/reference/estimate_null_correlation.html index 16f2ce1..1812ba6 100644 --- a/docs/reference/estimate_null_correlation.html +++ b/docs/reference/estimate_null_correlation.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,25 +83,21 @@ -
    +

    Estimate null correlations

    - Source: R/est_cor.R -
    -

    Estimates a null correlation matrix from data

    -
    - 
    estimate_null_correlation(data, Ulist, init, max_iter = 50,
-  tol = 0.001, est_cor = TRUE, track_fit = FALSE,
-  prior = c("nullbiased", "uniform"), ...)
    + 
    estimate_null_correlation(data, Ulist, init, max_iter = 30, tol = 1,
+  est_cor = TRUE, track_fit = FALSE, prior = c("nullbiased",
+  "uniform"), details = FALSE, ...)
    -

    Arguments

    +

    Arguments

    @@ -137,7 +122,7 @@

    Arg

    - + @@ -147,15 +132,32 @@

    Arg

    + + + +
    est_cor

    whether to estimate correlation matrix. If it is False, we estimate the covairance matrix

    whether to estimate correlation matrix (TRUE) or the covariance matrix (FALSE)
    track_fitprior

    indicates what penalty to use on the likelihood, if any
    details

    whether to return details of the model, if it is TRUE, the number of iterations and the value of objective functions will be returned

    ...

    other parameters pass to mash
    +

    Value

    + +

    the estimated correlation (or covariance) matrix and the fitted mash model

    +
    V

    estimated correlation (or covariance) matrix

    +
    mash.model

    fitted mash model

    + +

    Details

    -

    Returns the estimated correlation/covariance matrix of the effects

    +

    Returns the estimated correlation matrix (or covariance matrix) among conditions under the null. +The correlation (or covariance) matrix is estimated by maximum likelihood. +Specifically, the unknown correlation/covariance matrix V and the unknown weights are estimated iteratively. +The unknown correlation/covariance matrix V is estimated using the posterior second moment of the noise. +The unknown weights pi is estimated by maximum likelihood, which is a convex problem.

    +

    Warning: This method could take some time. +The estimate_null_correlation_simple gives a quick approximation for the null correlation (or covariance) matrix.

    @@ -164,6 +166,8 @@

    Contents

    @@ -182,8 +186,5 @@

    Contents

    - - - diff --git a/docs/reference/estimate_null_correlation_simple.html b/docs/reference/estimate_null_correlation_simple.html index ea2086c..0dbd1f5 100644 --- a/docs/reference/estimate_null_correlation_simple.html +++ b/docs/reference/estimate_null_correlation_simple.html @@ -6,7 +6,7 @@ -Estimate null correlations (ad hoc) — estimate_null_correlation_simple • mashr +Estimate null correlations (simple) — estimate_null_correlation_simple • mashr @@ -21,22 +21,16 @@ - - - + - - - - + + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr
    -

    @@ -94,23 +83,19 @@ -
    +
    -

    Estimate null correlations (ad hoc)

    - Source: R/est_cor.R - +

    Estimate null correlations (simple)

    -

    Estimates a null correlation matrix from data using simple z score threshold

    - 
    estimate_null_correlation_simple(data, z_thresh = 2, est_cor = TRUE)
    -

    Arguments

    +

    Arguments

    @@ -123,13 +108,15 @@

    Arg

    - +
    est_cor

    whether to estimate correlation matrix. If it is False, we estimate the covairance matrix

    whether to estimate correlation matrix (TRUE) or the covariance matrix (FALSE).

    Details

    -

    Returns the empirical correlation matrix of the effects that are "null" based on simple z score threshold

    +

    Returns a simple estimate of the correlation matrix (or covariance matrix) among conditions under the null. +Specifically, the simple estimate is the empirical correlation (or covariance) matrix of the z scores +for those effects that have (absolute) z score < z_thresh in all conditions.

    @@ -156,8 +143,5 @@

    Contents

    - - - diff --git a/docs/reference/expand_cov.html b/docs/reference/expand_cov.html index c73a2e3..4f57edd 100644 --- a/docs/reference/expand_cov.html +++ b/docs/reference/expand_cov.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr
    -
    @@ -94,23 +83,19 @@ -
    +

    Create expanded list of covariance matrices expanded by grid, Sigma_lk = omega_l U_k

    - Source: R/mash.R -
    -

    Create expanded list of covariance matrices expanded by grid, Sigma_lk = omega_l U_k

    - 
    expand_cov(Ulist, grid, usepointmass = TRUE)
    -

    Arguments

    +

    Arguments

    @@ -158,8 +143,5 @@

    Contents

    - - - diff --git a/docs/reference/extreme_deconvolution.html b/docs/reference/extreme_deconvolution.html new file mode 100644 index 0000000..6c60ca0 --- /dev/null +++ b/docs/reference/extreme_deconvolution.html @@ -0,0 +1,320 @@ + + + + + + + + +Density estimation using Gaussian mixtures in the presence of noisy, +heterogeneous and incomplete data — extreme_deconvolution • mashr + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
    +
    + + + +
    + +
    +
    +
    +

    Density estimation using Gaussian mixtures in the presence of noisy, +heterogeneous and incomplete data

    +
    + + +

    We present a general algorithm to infer a d-dimensional distribution +function given a set of heterogeneous, noisy observations or samples. This +algorithm reconstructs the error-deconvolved or 'underlying' distribution +function common to all samples, even when the individual samples have unique +error and missing-data properties. The underlying distribution is modeled as +a mixture of Gaussians, which is completely general. Model parameters are +chosen to optimize a justified, scalar objective function: the logarithm of +the probability of the data under the error-convolved model, where the error +convolution is different for each data point. Optimization is performed by +an Expectation Maximization (EM) algorithm, extended by a regularization +technique and 'split-and-merge' procedure. These extensions mitigate +problems with singularities and local maxima, which are often encountered +when using the EM algorithm to estimate Gaussian density mixtures.

    + + + 
    extreme_deconvolution(ydata,ycovar,xamp,xmean,xcovar,
+projection=NULL,weight=NULL, fixamp=NULL,fixmean=NULL,fixcovar=NULL,
+tol=1.e-6,maxiter=1e9,w=0,logfile=NULL,
+splitnmerge=0,maxsnm=FALSE,likeonly=FALSE, logweight=FALSE)
    + +

    Arguments

    +
    + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
    ydata

    [ndata,dy] matrix of observed quantities

    ycovar

    [ndata,dy] / [ndata,dy,dy] / [dy,dy,ndata] matrix, list or 3D +array of observational error covariances (if [ndata,dy] then the error +correlations are assumed to vanish)

    xamp

    [ngauss] array of initial amplitudes (*not* [1,ngauss])

    xmean

    [ngauss,dx] matrix of initial means

    xcovar

    [ngauss,dx,dx] list of matrices of initial covariances

    projection

    [ndata,dy,dx] list of projection matrices

    weight

    [ndata] array of weights to be applied to the data points

    fixamp

    (default=None) None, True/False, or list of bools

    fixmean

    (default=None) None, True/False, or list of bools

    fixcovar

    (default=None) None, True/False, or list of bools

    tol

    (double, default=1.e-6) tolerance for convergence

    maxiter

    (long, default= 10**9) maximum number of iterations to +perform

    w

    (double, default=0.) covariance regularization parameter (of the +conjugate prior)

    logfile

    basename for several logfiles (_c.log has output from the +c-routine; _loglike.log has the log likelihood path of all the accepted +routes, i.e. only parts which increase the likelihood are included, during +splitnmerge)

    splitnmerge

    (int, default=0) depth to go down the splitnmerge path

    maxsnm

    (Bool, default=False) use the maximum number of split 'n' +merge steps, K*(K-1)*(K-2)/2

    likeonly

    (Bool, default=False) only compute the total log likelihood +of the data

    logweight

    (bool, default=False) if True, weight is actually +log(weight)

    + +

    Value

    + + + +
    avgloglikedata

    avgloglikedata after convergence

    +
    xamp

    updated xamp

    xmean

    updated xmean

    xcovar

    updated +xcovar

    + + +

    References

    + +

    Inferring complete distribution functions from noisy, +heterogeneous and incomplete observations Jo Bovy, David W. Hogg, & Sam T. +Roweis, Submitted to AOAS (2009) [arXiv/0905.2979]

    + + +

    Examples

    + 
    ydata <-
+c(2.62434536, 0.38824359, 0.47182825, -0.07296862, 1.86540763,
+  -1.30153870, 2.74481176, 0.23879310, 1.31903910, 0.75062962,
+  2.46210794, -1.06014071, 0.67758280, 0.61594565, 2.13376944,
+  -0.09989127, 0.82757179, 0.12214158, 1.04221375, 1.58281521,
+  -0.10061918, 2.14472371, 1.90159072, 1.50249434, 1.90085595,
+  0.31627214, 0.87710977, 0.06423057, 0.73211192, 1.53035547,
+  0.30833925, 0.60324647, 0.31282730, 0.15479436, 0.32875387,
+  0.98733540, -0.11731035, 1.23441570, 2.65980218, 1.74204416,
+  0.80816445, 0.11237104, 0.25284171, 2.69245460, 1.05080775,
+  0.36300435, 1.19091548, 3.10025514, 1.12015895, 1.61720311,
+  1.30017032, 0.64775015, -0.14251820, 0.65065728, 0.79110577,
+  1.58662319, 1.83898341, 1.93110208, 1.28558733, 1.88514116,
+  0.24560206, 2.25286816, 1.51292982, 0.70190717, 1.48851815,
+  0.92442829, 2.13162939, 2.51981682, 3.18557541, -0.39649633,
+  -0.44411380, 0.49553414, 1.16003707, 1.87616892, 1.31563495,
+  -1.02220122, 0.69379599, 1.82797464, 1.23009474, 1.76201118,
+  0.77767186, 0.79924193, 1.18656139, 1.41005165, 1.19829972,
+  1.11900865, 0.32933771, 1.37756379, 1.12182127, 2.12948391,
+  2.19891788, 1.18515642, 0.62471505, 0.36126959, 1.42349435,
+  1.07734007, 0.65614632, 1.04359686, 0.37999916, 1.69803203,
+  0.55287144, 2.22450770, 1.40349164, 1.59357852, -0.09491185,
+  1.16938243, 1.74055645, 0.04629940, 0.73378149, 1.03261455,
+  -0.37311732, 1.31515939, 1.84616065, 0.14048406, 1.35054598,
+  -0.31228341, 0.96130449, -0.61577236, 2.12141771, 1.40890054,
+  0.97538304, 0.22483838, 2.27375593, 2.96710175, -0.85798186,
+  2.23616403, 2.62765075, 1.33801170, -0.19926803, 1.86334532,
+  0.81907970, 0.39607937, -0.23005814, 1.55053750, 1.79280687,
+  0.37646927, 1.52057634, -0.14434139, 1.80186103, 1.04656730,
+  0.81343023, 0.89825413, 1.86888616, 1.75041164, 1.52946532,
+  1.13770121, 1.07782113, 1.61838026, 1.23249456, 1.68255141,
+  0.68988323, -1.43483776, 2.03882460, 3.18697965, 1.44136444,
+  0.89984477, 0.86355526, 0.88094581, 1.01740941, -0.12201873,
+  0.48290554, 0.00297317, 1.24879916, 0.70335885, 1.49521132,
+  0.82529684, 1.98633519, 1.21353390, 3.19069973, -0.89636092,
+  0.35308331, 1.90148689, 3.52832571, 0.75136522, 1.04366899,
+  0.77368576, 2.33145711, 0.71269214, 1.68006984, 0.68019840,
+  -0.27255875, 1.31354772, 1.50318481, 2.29322588, 0.88955297,
+  0.38263794, 1.56276110, 1.24073709, 1.28066508, 0.92688730,
+  2.16033857, 1.36949272, 2.90465871, 2.11105670, 1.65904980,
+  -0.62743834, 1.60231928, 1.42028220, 1.81095167, 2.04444209)
+
+ydata  <- matrix(ydata,length(ydata),1)
+N      <- dim(ydata)[1]
+ycovar <- ydata*0 + 0.01
+xamp   <- c(0.5,0.5)
+xmean  <- matrix(c(0.86447943, 0.67078879, 0.322681, 0.45087394),2,2)
+xcovar <-
+  list(matrix(c(0.03821028, 0.04014796, 0.04108113, 0.03173839),2,2),
+       matrix(c(0.06219194, 0.09738021, 0.04302473, 0.06778009),2,2))
+projection <- list()
+for (i in 1:N)
+  projection[[i]] = matrix(c(i%%2,(i+1)%%2),1,2)
+res <- extreme_deconvolution(ydata, ycovar, xamp, xmean, xcovar,
+         projection=projection, logfile="ExDeconDemo")
    +
    + +
    + +
    +
    +

    Developed by Matthew Stephens, Sarah Urbut, Gao Wang, Yuxin Zou, Peter Carbonetto.

    +
    + +
    +

    Site built with pkgdown.

    +
    + +
    +
    + + + diff --git a/docs/reference/get_estimated_pi.html b/docs/reference/get_estimated_pi.html index cc75bde..8831587 100644 --- a/docs/reference/get_estimated_pi.html +++ b/docs/reference/get_estimated_pi.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr
    -
    @@ -94,23 +83,19 @@ -
    +

    Return the estimated mixture proportions

    - Source: R/get_functions.R -
    -

    Return the estimated mixture proportions

    - 
    get_estimated_pi(m, dimension = c("cov", "grid", "all"))
    -

    Arguments

    +

    Arguments

    @@ -164,8 +149,5 @@

    Contents

    - - - diff --git a/docs/reference/get_log10bf.html b/docs/reference/get_log10bf.html index e3deab0..644370d 100644 --- a/docs/reference/get_log10bf.html +++ b/docs/reference/get_log10bf.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    Return the Bayes Factor for each effect

    - Source: R/get_functions.R -
    -

    Return the Bayes Factor for each effect

    - 
    get_log10bf(m)
    -

    Arguments

    +

    Arguments

    @@ -150,8 +135,5 @@

    Contents

    - - - diff --git a/docs/reference/get_n_significant_conditions.html b/docs/reference/get_n_significant_conditions.html index c743f4f..5326d2a 100644 --- a/docs/reference/get_n_significant_conditions.html +++ b/docs/reference/get_n_significant_conditions.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,24 +83,20 @@ -
    +

    Count number of conditions each effect is significant in

    - Source: R/get_functions.R -
    -

    Count number of conditions each effect is significant in

    - 
    get_n_significant_conditions(m, thresh = 0.05, conditions = NULL,
   sig_fn = get_lfsr)
    -

    Arguments

    +

    Arguments

    @@ -161,8 +146,5 @@

    Contents

    - - - diff --git a/docs/reference/get_pairwise_sharing.html b/docs/reference/get_pairwise_sharing.html index e81484b..d5763db 100644 --- a/docs/reference/get_pairwise_sharing.html +++ b/docs/reference/get_pairwise_sharing.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,24 +83,20 @@ -
    +

    Compute the proportion of (significant) signals shared by magnitude in each pair of conditions, based on the poterior mean

    - Source: R/get_functions.R -
    -

    Compute the proportion of (significant) signals shared by magnitude in each pair of conditions, based on the poterior mean

    - 
    get_pairwise_sharing(m, factor = 0.5, lfsr_thresh = 0.05,
   FUN = identity)
    -

    Arguments

    +

    Arguments

    @@ -172,8 +157,5 @@

    Contents

    - - - diff --git a/docs/reference/get_pairwise_sharing_from_samples.html b/docs/reference/get_pairwise_sharing_from_samples.html index 57d5ae4..e03e03d 100644 --- a/docs/reference/get_pairwise_sharing_from_samples.html +++ b/docs/reference/get_pairwise_sharing_from_samples.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,24 +83,20 @@ -
    +

    Compute the proportion of (significant) signals shared by magnitude in each pair of conditions

    - Source: R/get_functions.R -
    -

    Compute the proportion of (significant) signals shared by magnitude in each pair of conditions

    - 
    get_pairwise_sharing_from_samples(m, factor = 0.5, lfsr_thresh = 0.05,
   FUN = identity)
    -

    Arguments

    +

    Arguments

    @@ -171,8 +156,5 @@

    Contents

    - - - diff --git a/docs/reference/get_samples.html b/docs/reference/get_samples.html index d2a471b..7362b6c 100644 --- a/docs/reference/get_samples.html +++ b/docs/reference/get_samples.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,27 +83,33 @@ -
    +

    Return samples from a mash object

    - Source: R/get_functions.R -
    -

    Return samples from a mash object

    - 
    get_samples(m)
    - + +

    Arguments

    +
    + + + + + +
    m

    The mash fit.

    +
    @@ -132,8 +127,5 @@

    Contents

    - - - diff --git a/docs/reference/get_significant_results.html b/docs/reference/get_significant_results.html index 0a1d2e1..f50538a 100644 --- a/docs/reference/get_significant_results.html +++ b/docs/reference/get_significant_results.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr
    -
    @@ -94,24 +83,20 @@ -
    +

    Find effects that are significant in at least one condition

    - Source: R/get_functions.R -
    -

    Find effects that are significant in at least one condition

    - 
    get_significant_results(m, thresh = 0.05, conditions = NULL,
   sig_fn = get_lfsr)
    -

    Arguments

    +

    Arguments

    @@ -161,8 +146,5 @@

    Contents

    - - - diff --git a/docs/reference/index.html b/docs/reference/index.html index 15ed9f4..7e98735 100644 --- a/docs/reference/index.html +++ b/docs/reference/index.html @@ -21,19 +21,13 @@ - - - + - - - + + - - - @@ -56,12 +50,8 @@ - - mashr - 0.2.18.495 - + mashr - @@ -91,376 +80,387 @@ -
    -
    +
    +
    -

    Reference

    +

    + Reference + version 0.2.20.0583 +

    -
    +
    +
    - - - - - + + + + - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + - - - - + + + - - - - - - - - - - - - + + + + + + + + + + + - - -
    -

    All functions

    -

    -
    -

    calc_lik_matrix()

    -

    Compute matrix of conditional likelihoods.

    -

    calc_lik_matrix_common_cov()

    -

    calc_lik_matrix_common_cov

    -

    calc_lik_vector()

    -

    Compute conditional likelihoods for bhat vector.

    -

    calc_relative_lik_matrix()

    -

    Calculate matrix of relative likelihoods.

    -

    compute_alt_loglik_from_matrix_and_pi()

    -

    Compute vector of alternative loglikelihoods from a matrix of log-likelihoods and fitted pi

    -

    compute_loglik_from_matrix_and_pi()

    -

    Compute total loglikelihood from a matrix of log-likelihoods and fitted pi

    -

    compute_null_loglik_from_matrix()

    -

    Compute vector of null loglikelihoods from a matrix of log-likelihoods

    -

    compute_posterior_matrices()

    -

    Compute posterior matrices.

    -

    compute_posterior_matrices_common_cov_R()

    -

    Compute posterior matrices (when error covariance V_j is equal for all observations j)

    -

    compute_posterior_matrices_general_R()

    -

    Compute posterior matrices (general version)

    -

    compute_posterior_weights()

    -

    compute posterior probabilities

    -

    compute_vloglik_from_matrix_and_pi()

    -

    Compute vector of loglikelihoods from a matrix of log-likelihoods and fitted pi

    -

    contrast_matrix()

    -

    Create contrast matrix

    -

    cov_all_zeros()

    -

    Compute an R by R matrix of all 0s

    -

    cov_canonical()

    -

    Compute a list of canonical covariance matrices

    -

    cov_ed()

    -

    Perform "extreme deconvolution" (Bovy et al) on a subset of the data

    -

    cov_equal_effects()

    -

    Compute an R by R matrix of all 1s

    -

    cov_first_singleton()

    -

    Compute all the singleton matrices corresponding to condition-specific effects in first condition only; used for testing purposes

    -

    cov_from_factors()

    -

    produce list of rank 1 covariance matrices corresponding to rows of f

    -

    cov_pca()

    -

    Perform PCA on data and return list of candidate covariance matrices

    -

    cov_simple_het()

    -

    Compute covariance matrices with diagonal element 1 and off-diagonal element corr

    -

    cov_udi()

    -

    Compute a list of covariance matrices corresponding to the +

    +

    All functions

    +

    +
    +

    calc_lik_matrix_common_cov

    +

    calc_lik_matrix_common_cov

    +

    calc_lik_matrix

    +

    Compute matrix of conditional likelihoods.

    +

    calc_lik_vector

    +

    Compute conditional likelihoods for bhat vector.

    +

    calc_relative_lik_matrix

    +

    Calculate matrix of relative likelihoods.

    +

    compute_alt_loglik_from_matrix_and_pi

    +

    Compute vector of alternative loglikelihoods from a matrix of log-likelihoods and fitted pi

    +

    compute_loglik_from_matrix_and_pi

    +

    Compute total loglikelihood from a matrix of log-likelihoods and fitted pi

    +

    compute_null_loglik_from_matrix

    +

    Compute vector of null loglikelihoods from a matrix of log-likelihoods

    +

    compute_posterior_matrices_common_cov_R

    +

    Compute posterior matrices (when error covariance V_j is equal for all observations j)

    +

    compute_posterior_matrices_general_R

    +

    Compute posterior matrices (general version)

    +

    compute_posterior_matrices

    +

    Compute posterior matrices.

    +

    compute_posterior_weights

    +

    compute posterior probabilities

    +

    compute_vloglik_from_matrix_and_pi

    +

    Compute vector of loglikelihoods from a matrix of log-likelihoods and fitted pi

    +

    contrast_matrix

    +

    Create contrast matrix

    +

    cov_all_zeros

    +

    Compute an R by R matrix of all 0s

    +

    cov_canonical

    +

    Compute a list of canonical covariance matrices

    +

    cov_ed

    +

    Perform "extreme deconvolution" (Bovy et al) on a subset of the data

    +

    cov_equal_effects

    +

    Compute an R by R matrix of all 1s

    +

    cov_first_singleton

    +

    Compute all the singleton matrices corresponding to condition-specific effects in first condition only; used for testing purposes

    +

    cov_from_factors

    +

    produce list of rank 1 covariance matrices corresponding to rows of f

    +

    cov_pca

    +

    Perform PCA on data and return list of candidate covariance matrices

    +

    cov_simple_het

    +

    Compute covariance matrices with diagonal element 1 and off-diagonal element corr

    +

    cov_udi_single

    +

    computes the covariance matrix for a single UDI model

    +

    cov_udi

    +

    Compute a list of covariance matrices corresponding to the "Unassociated", "Directly associated" and "Indirectly associated" models

    -

    cov_udi_single()

    -

    computes the covariance matrix for a single UDI model

    -

    ed_wrapper()

    -

    Fit extreme deconvolution to mash data

    -

    estimate_null_correlation()

    -

    Estimate null correlations

    -

    estimate_null_correlation_simple()

    -

    Estimate null correlations (ad hoc)

    -

    expand_cov()

    -

    Create expanded list of covariance matrices expanded by grid, Sigma_lk = omega_l U_k

    -

    get_estimated_pi()

    -

    Return the estimated mixture proportions

    -

    get_log10bf()

    -

    Return the Bayes Factor for each effect

    -

    get_n_significant_conditions()

    -

    Count number of conditions each effect is significant in

    -

    get_pairwise_sharing()

    -

    Compute the proportion of (significant) signals shared by magnitude in each pair of conditions, based on the poterior mean

    -

    get_pairwise_sharing_from_samples()

    -

    Compute the proportion of (significant) signals shared by magnitude in each pair of conditions

    -

    get_samples()

    -

    Return samples from a mash object

    -

    get_significant_results()

    -

    Find effects that are significant in at least one condition

    -

    initialize_pi()

    -

    Initialize mixture proportions - currently by making them all equal

    -

    is_common_cov_Shat()

    -

    Check that all covariances are equal (Shat).

    -

    is_common_cov_Shat_alpha()

    -

    Check that all rows of Shat_alpha are the same.

    -

    make_names()

    -

    Create names for covariance matrices

    -

    mash()

    -

    Apply mash method to data

    -

    mash_1by1()

    -

    Perform condition-by-condition analyses

    -

    mash_compute_loglik()

    -

    Compute loglikelihood for fitted mash object on new data.

    -

    mash_compute_posterior_matrices()

    -

    Compute posterior matrices for fitted mash object on new data

    -

    mash_compute_vloglik()

    -

    Compute vector of loglikelihood for fitted mash object on new data.

    -

    mash_plot_meta()

    -

    Plot metaplot for an effect based on posterior from mash

    -

    mash_set_data()

    -

    Create a data object for mash analysis.

    -

    mash_set_data_contrast()

    -

    Create a data object for mash contrast analysis

    -

    mash_update_data()

    -

    Update the data object for mash analysis.

    -

    optimize_pi()

    -

    Estimate the mixture weights by maximum (penalized) likelihood

    -

    posterior_cov()

    -

    posterior_cov

    -

    posterior_mean()

    -

    posterior_mean

    -

    posterior_mean_matrix()

    -

    posterior_mean_matrix

    -

    scale_cov()

    -

    Scale each covariance matrix in list Ulist by a scalar in vector grid

    -

    sim_contrast1()

    -

    Create simplest simulation, cj = mu 1 +

    +

    ed_wrapper

    +

    Fit extreme deconvolution to mash data

    +

    estimate_null_correlation_simple

    +

    Estimate null correlations (simple)

    +

    estimate_null_correlation

    +

    Estimate null correlations

    +

    expand_cov

    +

    Create expanded list of covariance matrices expanded by grid, Sigma_lk = omega_l U_k

    +

    extreme_deconvolution

    +

    Density estimation using Gaussian mixtures in the presence of noisy, +heterogeneous and incomplete data

    +

    get_estimated_pi

    +

    Return the estimated mixture proportions

    +

    get_log10bf

    +

    Return the Bayes Factor for each effect

    +

    get_n_significant_conditions

    +

    Count number of conditions each effect is significant in

    +

    get_pairwise_sharing_from_samples

    +

    Compute the proportion of (significant) signals shared by magnitude in each pair of conditions

    +

    get_pairwise_sharing

    +

    Compute the proportion of (significant) signals shared by magnitude in each pair of conditions, based on the poterior mean

    +

    get_samples

    +

    Return samples from a mash object

    +

    get_significant_results

    +

    Find effects that are significant in at least one condition

    +

    initialize_pi

    +

    Initialize mixture proportions - currently by making them all equal

    +

    is_common_cov_Shat_alpha

    +

    Check that all rows of Shat_alpha are the same.

    +

    is_common_cov_Shat

    +

    Check that all covariances are equal (Shat).

    +

    make_names

    +

    Create names for covariance matrices

    +

    mash_1by1

    +

    Perform condition-by-condition analyses

    +

    mash_compute_loglik

    +

    Compute loglikelihood for fitted mash object on new data.

    +

    mash_compute_posterior_matrices

    +

    Compute posterior matrices for fitted mash object on new data

    +

    mash_compute_vloglik

    +

    Compute vector of loglikelihood for fitted mash object on new data.

    +

    mash_plot_meta

    +

    Plot metaplot for an effect based on posterior from mash

    +

    mash_set_data_contrast

    +

    Create a data object for mash contrast analysis

    +

    mash_set_data

    +

    Create a data object for mash analysis.

    +

    mash_update_data

    +

    Update the data object for mash analysis.

    +

    mash

    +

    Apply mash method to data

    +

    optimize_pi

    +

    Estimate the mixture weights by maximum (penalized) likelihood

    +

    posterior_cov

    +

    posterior_cov

    +

    posterior_mean_matrix

    +

    posterior_mean_matrix

    +

    posterior_mean

    +

    posterior_mean

    +

    scale_cov

    +

    Scale each covariance matrix in list Ulist by a scalar in vector grid

    +

    sim_contrast1

    +

    Create simplest simulation, cj = mu 1 data used for contrast analysis

    -

    sim_contrast2()

    -

    Create simulation with signal data used for contrast +

    +

    sim_contrast2

    +

    Create simulation with signal data used for contrast analysis.

    -

    simple_sims()

    -

    Create some simple simulated data for testing purposes

    -

    simple_sims2()

    -

    Create some more simple simulated data for testing purposes

    -

    udi_model_matrix()

    -

    Create a matrix whose rows contain all possible +

    +

    simple_sims

    +

    Create some simple simulated data for testing purposes

    +

    simple_sims2

    +

    Create some more simple simulated data for testing purposes

    +

    udi_model_matrix

    +

    Create a matrix whose rows contain all possible combinations of the U,D,I models that are allowed.

    + + + +
    - - - diff --git a/docs/reference/initialize_pi.html b/docs/reference/initialize_pi.html index 874efef..1525fb7 100644 --- a/docs/reference/initialize_pi.html +++ b/docs/reference/initialize_pi.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr
    -
    @@ -94,23 +83,19 @@ -
    +

    Initialize mixture proportions - currently by making them all equal

    - Source: R/mash.R -
    -

    Initialize mixture proportions - currently by making them all equal

    - 
    initialize_pi(K)
    -

    Arguments

    +

    Arguments

    @@ -148,8 +133,5 @@

    Contents

    - - - diff --git a/docs/reference/is_common_cov_Shat.html b/docs/reference/is_common_cov_Shat.html index 17c0ca7..f2a2b3c 100644 --- a/docs/reference/is_common_cov_Shat.html +++ b/docs/reference/is_common_cov_Shat.html @@ -21,23 +21,17 @@ - - - + - - - + + - - - @@ -60,12 +54,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -95,24 +84,20 @@ -
    +

    Check that all covariances are equal (Shat).

    - Source: R/set_data.R -
    -

    checks if all rows of Shat are the same - if so covariances are equal

    - 
    is_common_cov_Shat(data)
    -

    Arguments

    +

    Arguments

    @@ -144,8 +129,5 @@

    Contents

    - - - diff --git a/docs/reference/is_common_cov_Shat_alpha.html b/docs/reference/is_common_cov_Shat_alpha.html index 4907ed1..adfbf88 100644 --- a/docs/reference/is_common_cov_Shat_alpha.html +++ b/docs/reference/is_common_cov_Shat_alpha.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    Check that all rows of Shat_alpha are the same.

    - Source: R/set_data.R -
    -

    checks if all rows of Shat_alpha are the same

    - 
    is_common_cov_Shat_alpha(data)
    -

    Arguments

    +

    Arguments

    @@ -142,8 +127,5 @@

    Contents

    - - - diff --git a/docs/reference/make_names.html b/docs/reference/make_names.html index 76aa642..2af3d9e 100644 --- a/docs/reference/make_names.html +++ b/docs/reference/make_names.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    Create names for covariance matrices

    - Source: R/compute_covs.R -
    -

    Create names for covariance matrices

    - 
    make_names(names, suffixes)
    -

    Arguments

    +

    Arguments

    @@ -146,8 +131,5 @@

    Contents

    - - - diff --git a/docs/reference/mash.html b/docs/reference/mash.html index 667c12d..1cfd46a 100644 --- a/docs/reference/mash.html +++ b/docs/reference/mash.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,28 +83,25 @@ -
    +

    Apply mash method to data

    - Source: R/mash.R -
    -

    Apply mash method to data

    - 
    mash(data, Ulist = NULL, gridmult = sqrt(2), grid = NULL,
   normalizeU = TRUE, usepointmass = TRUE, g = NULL, fixg = FALSE,
-  prior = c("nullbiased", "uniform"), optmethod = c("mixIP", "mixEM",
-  "mixSQP", "cxxMixSquarem"), verbose = TRUE, add.mem.profile = FALSE,
-  algorithm.version = c("Rcpp", "R"), pi_thresh = 1e-10, A = NULL,
-  posterior_samples = 0, seed = 123, outputlevel = 2)
    + prior = c("nullbiased", "uniform"), optmethod = c("mixSQP", "mixIP", + "mixEM", "cxxMixSquarem"), control = list(), verbose = TRUE, + add.mem.profile = FALSE, algorithm.version = c("Rcpp", "R"), + pi_thresh = 1e-10, A = NULL, posterior_samples = 0, seed = 123, + outputlevel = 2) -

    Arguments

    +

    Arguments

    @@ -158,6 +144,10 @@

    Arg

    + + + + @@ -188,7 +178,7 @@

    Arg

    - +
    optmethod

    name of optimization method to use
    control

    A list of control parameters passed to optmethod.

    verbose

    If TRUE, print progress to R console.

    outputlevel

    controls amount of computation / output; 1: output only estimated mixture component proportions, 2: and posterior estimates, 3: and posterior covariance matrices, 4: and likelihood matrices and posterior weights

    controls amount of computation / output; 1: output only estimated mixture component proportions, 2: and posterior estimates, 3: and posterior covariance matrices, 4: and likelihood matrices
    @@ -234,8 +224,5 @@

    Contents

    - - - diff --git a/docs/reference/mash_1by1.html b/docs/reference/mash_1by1.html index c4e4b3a..8fcc91e 100644 --- a/docs/reference/mash_1by1.html +++ b/docs/reference/mash_1by1.html @@ -21,25 +21,19 @@ - - - + - - - + + - - - - @@ -62,12 +56,8 @@ - - mashr - 0.2.18.476 - + mashr
    -
    @@ -97,26 +86,22 @@ -
    +

    Perform condition-by-condition analyses

    - Source: R/mash.R -
    -

    Performs simple "condition-by-condition" analysis by running ash from package ashr on data from each condition, one at a time. May be a useful first step to identify top hits in each condition before a mash analysis.

    - 
    mash_1by1(data, alpha = 0)
    -

    Arguments

    +

    Arguments

    @@ -162,8 +147,5 @@

    Contents

    - - - diff --git a/docs/reference/mash_compute_loglik.html b/docs/reference/mash_compute_loglik.html index 2ebb24a..dea44b6 100644 --- a/docs/reference/mash_compute_loglik.html +++ b/docs/reference/mash_compute_loglik.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    Compute loglikelihood for fitted mash object on new data.

    - Source: R/likelihoods_origdata.R -
    -

    Compute loglikelihood for fitted mash object on new data.

    - 
    mash_compute_loglik(g, data, algorithm.version = c("Rcpp", "R"))
    -

    Arguments

    +

    Arguments

    @@ -164,8 +149,5 @@

    Contents

    - - - diff --git a/docs/reference/mash_compute_posterior_matrices.html b/docs/reference/mash_compute_posterior_matrices.html index d495d34..b9d9ddb 100644 --- a/docs/reference/mash_compute_posterior_matrices.html +++ b/docs/reference/mash_compute_posterior_matrices.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,25 +83,21 @@ -
    +

    Compute posterior matrices for fitted mash object on new data

    - Source: R/mash.R -
    -

    Compute posterior matrices for fitted mash object on new data

    - 
    mash_compute_posterior_matrices(g, data, pi_thresh = 1e-10,
   algorithm.version = c("Rcpp", "R"), A = NULL,
   output_posterior_cov = FALSE, posterior_samples = 0, seed = 123)
    -

    Arguments

    +

    Arguments

    @@ -178,8 +163,5 @@

    Contents

    - - - diff --git a/docs/reference/mash_compute_vloglik.html b/docs/reference/mash_compute_vloglik.html index 20236fb..e223bb4 100644 --- a/docs/reference/mash_compute_vloglik.html +++ b/docs/reference/mash_compute_vloglik.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    Compute vector of loglikelihood for fitted mash object on new data.

    - Source: R/likelihoods_origdata.R -
    -

    Compute vector of loglikelihood for fitted mash object on new data.

    - 
    mash_compute_vloglik(g, data, algorithm.version = c("Rcpp", "R"))
    -

    Arguments

    +

    Arguments

    @@ -173,8 +158,5 @@

    Contents

    - - - diff --git a/docs/reference/mash_plot_meta.html b/docs/reference/mash_plot_meta.html index a859edf..b8fba5c 100644 --- a/docs/reference/mash_plot_meta.html +++ b/docs/reference/mash_plot_meta.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    Plot metaplot for an effect based on posterior from mash

    - Source: R/plots.R -
    -

    Plot metaplot for an effect based on posterior from mash

    - 
    mash_plot_meta(m, i, xlab = "Effect size", ylab = "Condition", ...)
    -

    Arguments

    +

    Arguments

    @@ -158,8 +143,5 @@

    Contents

    - - - diff --git a/docs/reference/mash_set_data.html b/docs/reference/mash_set_data.html index b7bdcd3..f736b3e 100644 --- a/docs/reference/mash_set_data.html +++ b/docs/reference/mash_set_data.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,24 +83,20 @@ -
    +

    Create a data object for mash analysis.

    - Source: R/set_data.R -
    -

    Create a data object for mash analysis.

    - 
    mash_set_data(Bhat, Shat = NULL, alpha = 0, df = Inf, pval = NULL,
   V = diag(ncol(Bhat)))
    -

    Arguments

    +

    Arguments

    @@ -146,9 +131,9 @@

    Arg

    -
    V

    an R by R correlation matrix of error correlations; must +

    an R by R matrix / [R x R x J] array of effect specific correlation matrix of error correlations; must be positive definite. [So Bhat_j distributed as N(B_j,diag(Shat_j) -V diag(Shat_j)) where _j denotes the jth row of a matrix]. +V_j diag(Shat_j)) where _j denotes the jth row of a matrix]. Defaults to identity.
    @@ -182,8 +167,5 @@

    Contents

    - - - diff --git a/docs/reference/mash_set_data_contrast.html b/docs/reference/mash_set_data_contrast.html index ed5fef3..7942b10 100644 --- a/docs/reference/mash_set_data_contrast.html +++ b/docs/reference/mash_set_data_contrast.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr
    -
    @@ -94,23 +83,19 @@ -
    +

    Create a data object for mash contrast analysis

    - Source: R/set_data.R -
    -

    Create a data object for mash contrast analysis

    - 
    mash_set_data_contrast(mashdata, L)
    -

    Arguments

    +

    Arguments

    @@ -152,8 +137,5 @@

    Contents

    - - - diff --git a/docs/reference/mash_update_data.html b/docs/reference/mash_update_data.html index c8f4299..f22df72 100644 --- a/docs/reference/mash_update_data.html +++ b/docs/reference/mash_update_data.html @@ -21,22 +21,17 @@ - - - + - - - + + - + - - - @@ -59,12 +54,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +84,20 @@ -
    +

    Update the data object for mash analysis.

    - Source: R/set_data.R -
    -
    -

    Update the data object for mash analysis.

    +

    This function can update two parts of the mash data. The first one is setting the reference group, so the mash data +can be used for commonbaseline analysis. The other one is updating the null correlation matrix.

    - 
    mash_update_data(mashdata, ref = NULL, V = NULL)
    -

    Arguments

    +

    Arguments

    @@ -123,7 +110,7 @@

    Arg

    - +
    V

    a correlation matrix for the null effects

    an R by R matrix / [R x R x J] array of correlation matrix of error correlations
    @@ -156,8 +143,5 @@

    Contents

    - - - diff --git a/docs/reference/optimize_pi.html b/docs/reference/optimize_pi.html index 7a43b37..6f0a031 100644 --- a/docs/reference/optimize_pi.html +++ b/docs/reference/optimize_pi.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr
    -
    @@ -94,25 +83,21 @@ -
    +

    Estimate the mixture weights by maximum (penalized) likelihood

    - Source: R/opt.R -
    -

    Estimate the mixture weights by maximum (penalized) likelihood

    - 
    optimize_pi(matrix_lik, pi_init = NULL, prior = NULL,
-  optmethod = c("mixIP", "mixEM", "mixSQP", "cxxMixSquarem"),
+  optmethod = c("mixSQP", "mixIP", "mixEM", "cxxMixSquarem"),
   control = list())
    -

    Arguments

    +

    Arguments

    @@ -166,8 +151,5 @@

    Contents

    - - - diff --git a/docs/reference/posterior_cov.html b/docs/reference/posterior_cov.html index 5d2f1ad..dc55620 100644 --- a/docs/reference/posterior_cov.html +++ b/docs/reference/posterior_cov.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    posterior_cov

    - Source: R/posterior.R -
    -

    If bhat is N(b,V) and b is N(0,U) then b|bhat N(mu1,U1). This function returns U1.

    - 
    posterior_cov(Vinv, U)
    -

    Arguments

    +

    Arguments

    @@ -152,8 +137,5 @@

    Contents

    - - - diff --git a/docs/reference/posterior_mean.html b/docs/reference/posterior_mean.html index 03be869..f715e64 100644 --- a/docs/reference/posterior_mean.html +++ b/docs/reference/posterior_mean.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    posterior_mean

    - Source: R/posterior.R -
    -

    If bhat is N(b,V) and b is N(0,U) then b|bhat N(mu1,U1). This function returns mu1.

    - 
    posterior_mean(bhat, Vinv, U1)
    -

    Arguments

    +

    Arguments

    @@ -156,8 +141,5 @@

    Contents

    - - - diff --git a/docs/reference/posterior_mean_matrix.html b/docs/reference/posterior_mean_matrix.html index ddd6ac6..26bfd7d 100644 --- a/docs/reference/posterior_mean_matrix.html +++ b/docs/reference/posterior_mean_matrix.html @@ -21,24 +21,18 @@ - - - + - - - + + - - - @@ -61,12 +55,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -96,25 +85,21 @@ -
    +

    posterior_mean_matrix

    - Source: R/posterior.R -
    -

    Computes posterior mean under multivariate normal model for each row of matrix Bhat. Note that if bhat is N_R(b,V) and b is N_R(0,U) then b|bhat N_R(mu1,U1). This function returns a matrix with jth row equal to mu1(bhat) for bhat= Bhat[j,].

    - 
    posterior_mean_matrix(Bhat, Vinv, U1)
    -

    Arguments

    +

    Arguments

    @@ -160,8 +145,5 @@

    Contents

    - - - diff --git a/docs/reference/scale_cov.html b/docs/reference/scale_cov.html index 64408c9..0c8dee6 100644 --- a/docs/reference/scale_cov.html +++ b/docs/reference/scale_cov.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    Scale each covariance matrix in list Ulist by a scalar in vector grid

    - Source: R/compute_covs.R -
    -

    Scale each covariance matrix in list Ulist by a scalar in vector grid

    - 
    scale_cov(Ulist, grid)
    -

    Arguments

    +

    Arguments

    @@ -152,8 +137,5 @@

    Contents

    - - - diff --git a/docs/reference/sim_contrast1.html b/docs/reference/sim_contrast1.html index 53d3165..020e10a 100644 --- a/docs/reference/sim_contrast1.html +++ b/docs/reference/sim_contrast1.html @@ -22,24 +22,18 @@ - - - + - - - + + - - - @@ -62,12 +56,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -97,25 +86,21 @@ -
    +

    Create simplest simulation, cj = mu 1 data used for contrast analysis

    - Source: R/simulations_contrast.R -
    -

    Create simplest simulation, cj = mu 1 data used for contrast analysis

    - 
    sim_contrast1(nsamp = 100, ncond = 5, err_sd = sqrt(0.5))
    -

    Arguments

    +

    Arguments

    @@ -161,8 +146,5 @@

    Contents

    - - - diff --git a/docs/reference/sim_contrast2.html b/docs/reference/sim_contrast2.html index 92bbdc6..122dd14 100644 --- a/docs/reference/sim_contrast2.html +++ b/docs/reference/sim_contrast2.html @@ -22,24 +22,18 @@ - - - + - - - + + - - - @@ -62,12 +56,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -97,25 +86,21 @@ -
    +

    Create simulation with signal data used for contrast analysis.

    - Source: R/simulations_contrast.R -
    -

    Create simulation with signal data used for contrast analysis.

    - 
    sim_contrast2(nsamp = 1000, ncond = 5, err_sd = sqrt(0.5))
    -

    Arguments

    +

    Arguments

    @@ -166,8 +151,5 @@

    Contents

    - - - diff --git a/docs/reference/simple_sims.html b/docs/reference/simple_sims.html index 2dcc546..c7d4f71 100644 --- a/docs/reference/simple_sims.html +++ b/docs/reference/simple_sims.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    Create some simple simulated data for testing purposes

    - Source: R/simulations.R -
    -

    Create some simple simulated data for testing purposes

    - 
    simple_sims(nsamp = 100, ncond = 5, err_sd = 0.01)
    -

    Arguments

    +

    Arguments

    @@ -156,8 +141,5 @@

    Contents

    - - - diff --git a/docs/reference/simple_sims2.html b/docs/reference/simple_sims2.html index 10d1b75..3922be5 100644 --- a/docs/reference/simple_sims2.html +++ b/docs/reference/simple_sims2.html @@ -21,22 +21,16 @@ - - - + - - - + + - - - @@ -59,12 +53,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -94,23 +83,19 @@ -
    +

    Create some more simple simulated data for testing purposes

    - Source: R/simulations.R -
    -

    Create some more simple simulated data for testing purposes

    - 
    simple_sims2(nsamp = 100, err_sd = 0.01)
    -

    Arguments

    +

    Arguments

    @@ -153,8 +138,5 @@

    Contents

    - - - diff --git a/docs/reference/udi_model_matrix.html b/docs/reference/udi_model_matrix.html index 7cc5ba7..e63945f 100644 --- a/docs/reference/udi_model_matrix.html +++ b/docs/reference/udi_model_matrix.html @@ -22,24 +22,18 @@ - - - + - - - + + - - - @@ -62,12 +56,8 @@ - - mashr - 0.2.18.476 - + mashr - @@ -97,25 +86,21 @@ -
    +

    Create a matrix whose rows contain all possible combinations of the U,D,I models that are allowed.

    - Source: R/cov_udi.R -
    -

    Create a matrix whose rows contain all possible combinations of the U,D,I models that are allowed.

    - 
    udi_model_matrix(R)
    -

    Arguments

    +

    Arguments

    @@ -155,8 +140,5 @@

    Contents

    - - -