-
Notifications
You must be signed in to change notification settings - Fork 196
/
train_mnist_99.57.log
1279 lines (1279 loc) · 68.9 KB
/
train_mnist_99.57.log
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
I0805 14:52:42.048267 1392 caffe.cpp:113] Use GPU with device ID 0
I0805 14:52:42.416288 1392 common.cpp:24] System entropy source not available, using fallback algorithm to generate seed instead.
I0805 14:52:42.417289 1392 caffe.cpp:121] Starting Optimization
I0805 14:52:42.417289 1392 solver.cpp:32] Initializing solver from parameters:
test_iter: 100
test_interval: 1000
base_lr: 0.04
display: 100
max_iter: 15000
lr_policy: "inv"
gamma: 0.0002
power: 0.9
momentum: 0.8
weight_decay: 0.0001
snapshot: 5000
snapshot_prefix: "lenet"
solver_mode: GPU
net: "lenet_train_test.prototxt"
I0805 14:52:42.417289 1392 solver.cpp:70] Creating training net from net file: lenet_train_test.prototxt
I0805 14:52:42.418288 1392 net.cpp:287] The NetState phase (0) differed from the phase (1) specified by a rule in layer mnist
I0805 14:52:42.418288 1392 net.cpp:287] The NetState phase (0) differed from the phase (1) specified by a rule in layer accuracy
I0805 14:52:42.418288 1392 net.cpp:42] Initializing net from parameters:
name: "LeNet"
state {
phase: TRAIN
}
layer {
name: "mnist"
type: "Data"
top: "data"
top: "label"
include {
phase: TRAIN
}
transform_param {
scale: 0.00390625
}
data_param {
source: "mnist-train-leveldb"
batch_size: 64
backend: LEVELDB
}
}
layer {
name: "conv1"
type: "Convolution"
bottom: "data"
top: "conv1"
param {
lr_mult: 1
}
param {
lr_mult: 2
}
convolution_param {
num_output: 32
kernel_size: 6
stride: 1
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "relu1"
type: "ReLU"
bottom: "conv1"
top: "conv1"
}
layer {
name: "conv1_bn"
type: "BN"
bottom: "conv1"
top: "conv1_bn"
param {
lr_mult: 1
decay_mult: 0
}
param {
lr_mult: 1
decay_mult: 0
}
bn_param {
scale_filler {
type: "constant"
value: 1
}
shift_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "drop1"
type: "Dropout"
bottom: "conv1_bn"
top: "conv1_bn"
dropout_param {
dropout_ratio: 0
}
}
layer {
name: "pool1"
type: "Pooling"
bottom: "conv1_bn"
top: "pool1"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "conv2"
type: "Convolution"
bottom: "pool1"
top: "conv2"
param {
lr_mult: 1
}
param {
lr_mult: 2
}
convolution_param {
num_output: 72
kernel_size: 3
stride: 1
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "relu2"
type: "ReLU"
bottom: "conv2"
top: "conv2"
}
layer {
name: "conv2_bn"
type: "BN"
bottom: "conv2"
top: "conv2_bn"
param {
lr_mult: 1
decay_mult: 0
}
param {
lr_mult: 1
decay_mult: 0
}
bn_param {
scale_filler {
type: "constant"
value: 1
}
shift_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "drop2"
type: "Dropout"
bottom: "conv2_bn"
top: "conv2_bn"
dropout_param {
dropout_ratio: 0.1
}
}
layer {
name: "pool2"
type: "Pooling"
bottom: "conv2_bn"
top: "pool2"
pooling_param {
pool: MAX
kernel_size: 2
stride: 2
}
}
layer {
name: "ip1"
type: "InnerProduct"
bottom: "pool2"
top: "ip1"
param {
lr_mult: 1
}
param {
lr_mult: 2
}
inner_product_param {
num_output: 320
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "relu3"
type: "ReLU"
bottom: "ip1"
top: "ip1"
}
layer {
name: "drop3"
type: "Dropout"
bottom: "ip1"
top: "ip1"
dropout_param {
dropout_ratio: 0.3
}
}
layer {
name: "ip2"
type: "InnerProduct"
bottom: "ip1"
top: "ip2"
param {
lr_mult: 1
}
param {
lr_mult: 2
}
inner_product_param {
num_output: 10
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "loss"
type: "SoftmaxWithLoss"
bottom: "ip2"
bottom: "label"
top: "loss"
}
I0805 14:52:42.457290 1392 layer_factory.hpp:74] Creating layer mnist
I0805 14:52:42.459291 1392 net.cpp:90] Creating Layer mnist
I0805 14:52:42.460291 1392 net.cpp:368] mnist -> data
I0805 14:52:42.460291 1392 net.cpp:368] mnist -> label
I0805 14:52:42.461292 1392 net.cpp:120] Setting up mnist
I0805 14:52:42.468291 1392 db.cpp:20] Opened leveldb mnist-train-leveldb
I0805 14:52:42.469291 1392 data_layer.cpp:52] output data size: 64,1,28,28
I0805 14:52:42.469291 1392 net.cpp:127] Top shape: 64 1 28 28 (50176)
I0805 14:52:42.470291 1392 net.cpp:127] Top shape: 64 (64)
I0805 14:52:42.470291 1392 layer_factory.hpp:74] Creating layer conv1
I0805 14:52:42.471292 1392 net.cpp:90] Creating Layer conv1
I0805 14:52:42.472291 1392 net.cpp:410] conv1 <- data
I0805 14:52:42.473291 1392 net.cpp:368] conv1 -> conv1
I0805 14:52:42.473291 1392 net.cpp:120] Setting up conv1
I0805 14:52:42.474292 1392 common.cpp:24] System entropy source not available, using fallback algorithm to generate seed instead.
I0805 14:52:42.539295 1392 net.cpp:127] Top shape: 64 32 23 23 (1083392)
I0805 14:52:42.540295 1392 layer_factory.hpp:74] Creating layer relu1
I0805 14:52:42.541296 1392 net.cpp:90] Creating Layer relu1
I0805 14:52:42.541296 1392 net.cpp:410] relu1 <- conv1
I0805 14:52:42.542295 1392 net.cpp:357] relu1 -> conv1 (in-place)
I0805 14:52:42.542295 1392 net.cpp:120] Setting up relu1
I0805 14:52:42.543295 1392 net.cpp:127] Top shape: 64 32 23 23 (1083392)
I0805 14:52:42.544296 1392 layer_factory.hpp:74] Creating layer conv1_bn
I0805 14:52:42.544296 1392 net.cpp:90] Creating Layer conv1_bn
I0805 14:52:42.544296 1392 net.cpp:410] conv1_bn <- conv1
I0805 14:52:42.545296 1392 net.cpp:368] conv1_bn -> conv1_bn
I0805 14:52:42.545296 1392 net.cpp:120] Setting up conv1_bn
I0805 14:52:42.546296 1392 net.cpp:127] Top shape: 64 32 23 23 (1083392)
I0805 14:52:42.546296 1392 layer_factory.hpp:74] Creating layer drop1
I0805 14:52:42.547297 1392 net.cpp:90] Creating Layer drop1
I0805 14:52:42.547297 1392 net.cpp:410] drop1 <- conv1_bn
I0805 14:52:42.548296 1392 net.cpp:357] drop1 -> conv1_bn (in-place)
I0805 14:52:42.548296 1392 net.cpp:120] Setting up drop1
I0805 14:52:42.549296 1392 net.cpp:127] Top shape: 64 32 23 23 (1083392)
I0805 14:52:42.550297 1392 layer_factory.hpp:74] Creating layer pool1
I0805 14:52:42.550297 1392 net.cpp:90] Creating Layer pool1
I0805 14:52:42.551296 1392 net.cpp:410] pool1 <- conv1_bn
I0805 14:52:42.551296 1392 net.cpp:368] pool1 -> pool1
I0805 14:52:42.552296 1392 net.cpp:120] Setting up pool1
I0805 14:52:42.552296 1392 net.cpp:127] Top shape: 64 32 11 11 (247808)
I0805 14:52:42.553297 1392 layer_factory.hpp:74] Creating layer conv2
I0805 14:52:42.553297 1392 net.cpp:90] Creating Layer conv2
I0805 14:52:42.554296 1392 net.cpp:410] conv2 <- pool1
I0805 14:52:42.554296 1392 net.cpp:368] conv2 -> conv2
I0805 14:52:42.554296 1392 net.cpp:120] Setting up conv2
I0805 14:52:42.555296 1392 net.cpp:127] Top shape: 64 72 9 9 (373248)
I0805 14:52:42.555296 1392 layer_factory.hpp:74] Creating layer relu2
I0805 14:52:42.556296 1392 net.cpp:90] Creating Layer relu2
I0805 14:52:42.556296 1392 net.cpp:410] relu2 <- conv2
I0805 14:52:42.556296 1392 net.cpp:357] relu2 -> conv2 (in-place)
I0805 14:52:42.556296 1392 net.cpp:120] Setting up relu2
I0805 14:52:42.557296 1392 net.cpp:127] Top shape: 64 72 9 9 (373248)
I0805 14:52:42.557296 1392 layer_factory.hpp:74] Creating layer conv2_bn
I0805 14:52:42.557296 1392 net.cpp:90] Creating Layer conv2_bn
I0805 14:52:42.558296 1392 net.cpp:410] conv2_bn <- conv2
I0805 14:52:42.558296 1392 net.cpp:368] conv2_bn -> conv2_bn
I0805 14:52:42.558296 1392 net.cpp:120] Setting up conv2_bn
I0805 14:52:42.558296 1392 net.cpp:127] Top shape: 64 72 9 9 (373248)
I0805 14:52:42.559296 1392 layer_factory.hpp:74] Creating layer drop2
I0805 14:52:42.559296 1392 net.cpp:90] Creating Layer drop2
I0805 14:52:42.559296 1392 net.cpp:410] drop2 <- conv2_bn
I0805 14:52:42.559296 1392 net.cpp:357] drop2 -> conv2_bn (in-place)
I0805 14:52:42.559296 1392 net.cpp:120] Setting up drop2
I0805 14:52:42.560297 1392 net.cpp:127] Top shape: 64 72 9 9 (373248)
I0805 14:52:42.560297 1392 layer_factory.hpp:74] Creating layer pool2
I0805 14:52:42.560297 1392 net.cpp:90] Creating Layer pool2
I0805 14:52:42.561296 1392 net.cpp:410] pool2 <- conv2_bn
I0805 14:52:42.561296 1392 net.cpp:368] pool2 -> pool2
I0805 14:52:42.561296 1392 net.cpp:120] Setting up pool2
I0805 14:52:42.562296 1392 net.cpp:127] Top shape: 64 72 5 5 (115200)
I0805 14:52:42.562296 1392 layer_factory.hpp:74] Creating layer ip1
I0805 14:52:42.562296 1392 net.cpp:90] Creating Layer ip1
I0805 14:52:42.563297 1392 net.cpp:410] ip1 <- pool2
I0805 14:52:42.563297 1392 net.cpp:368] ip1 -> ip1
I0805 14:52:42.563297 1392 net.cpp:120] Setting up ip1
I0805 14:52:42.567297 1392 net.cpp:127] Top shape: 64 320 (20480)
I0805 14:52:42.568297 1392 layer_factory.hpp:74] Creating layer relu3
I0805 14:52:42.568297 1392 net.cpp:90] Creating Layer relu3
I0805 14:52:42.569298 1392 net.cpp:410] relu3 <- ip1
I0805 14:52:42.569298 1392 net.cpp:357] relu3 -> ip1 (in-place)
I0805 14:52:42.569298 1392 net.cpp:120] Setting up relu3
I0805 14:52:42.570297 1392 net.cpp:127] Top shape: 64 320 (20480)
I0805 14:52:42.570297 1392 layer_factory.hpp:74] Creating layer drop3
I0805 14:52:42.571297 1392 net.cpp:90] Creating Layer drop3
I0805 14:52:42.571297 1392 net.cpp:410] drop3 <- ip1
I0805 14:52:42.571297 1392 net.cpp:357] drop3 -> ip1 (in-place)
I0805 14:52:42.571297 1392 net.cpp:120] Setting up drop3
I0805 14:52:42.572298 1392 net.cpp:127] Top shape: 64 320 (20480)
I0805 14:52:42.572298 1392 layer_factory.hpp:74] Creating layer ip2
I0805 14:52:42.572298 1392 net.cpp:90] Creating Layer ip2
I0805 14:52:42.572298 1392 net.cpp:410] ip2 <- ip1
I0805 14:52:42.572298 1392 net.cpp:368] ip2 -> ip2
I0805 14:52:42.573297 1392 net.cpp:120] Setting up ip2
I0805 14:52:42.573297 1392 net.cpp:127] Top shape: 64 10 (640)
I0805 14:52:42.573297 1392 layer_factory.hpp:74] Creating layer loss
I0805 14:52:42.573297 1392 net.cpp:90] Creating Layer loss
I0805 14:52:42.574297 1392 net.cpp:410] loss <- ip2
I0805 14:52:42.574297 1392 net.cpp:410] loss <- label
I0805 14:52:42.574297 1392 net.cpp:368] loss -> loss
I0805 14:52:42.574297 1392 net.cpp:120] Setting up loss
I0805 14:52:42.575297 1392 layer_factory.hpp:74] Creating layer loss
I0805 14:52:42.575297 1392 net.cpp:127] Top shape: (1)
I0805 14:52:42.575297 1392 net.cpp:129] with loss weight 1
I0805 14:52:42.575297 1392 net.cpp:192] loss needs backward computation.
I0805 14:52:42.576297 1392 net.cpp:192] ip2 needs backward computation.
I0805 14:52:42.576297 1392 net.cpp:192] drop3 needs backward computation.
I0805 14:52:42.576297 1392 net.cpp:192] relu3 needs backward computation.
I0805 14:52:42.576297 1392 net.cpp:192] ip1 needs backward computation.
I0805 14:52:42.576297 1392 net.cpp:192] pool2 needs backward computation.
I0805 14:52:42.577297 1392 net.cpp:192] drop2 needs backward computation.
I0805 14:52:42.577297 1392 net.cpp:192] conv2_bn needs backward computation.
I0805 14:52:42.577297 1392 net.cpp:192] relu2 needs backward computation.
I0805 14:52:42.577297 1392 net.cpp:192] conv2 needs backward computation.
I0805 14:52:42.578297 1392 net.cpp:192] pool1 needs backward computation.
I0805 14:52:42.578297 1392 net.cpp:192] drop1 needs backward computation.
I0805 14:52:42.578297 1392 net.cpp:192] conv1_bn needs backward computation.
I0805 14:52:42.578297 1392 net.cpp:192] relu1 needs backward computation.
I0805 14:52:42.579298 1392 net.cpp:192] conv1 needs backward computation.
I0805 14:52:42.579298 1392 net.cpp:194] mnist does not need backward computation.
I0805 14:52:42.579298 1392 net.cpp:235] This network produces output loss
I0805 14:52:42.580298 1392 net.cpp:482] Collecting Learning Rate and Weight Decay.
I0805 14:52:42.580298 1392 net.cpp:247] Network initialization done.
I0805 14:52:42.580298 1392 net.cpp:248] Memory required for data: 25207556
I0805 14:52:42.581298 1392 solver.cpp:154] Creating test net (#0) specified by net file: lenet_train_test.prototxt
I0805 14:52:42.582298 1392 net.cpp:287] The NetState phase (1) differed from the phase (0) specified by a rule in layer mnist
I0805 14:52:42.582298 1392 net.cpp:42] Initializing net from parameters:
name: "LeNet"
state {
phase: TEST
}
layer {
name: "mnist"
type: "Data"
top: "data"
top: "label"
include {
phase: TEST
}
transform_param {
scale: 0.00390625
}
data_param {
source: "mnist-test-leveldb"
batch_size: 100
backend: LEVELDB
}
}
layer {
name: "conv1"
type: "Convolution"
bottom: "data"
top: "conv1"
param {
lr_mult: 1
}
param {
lr_mult: 2
}
convolution_param {
num_output: 32
kernel_size: 6
stride: 1
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "relu1"
type: "ReLU"
bottom: "conv1"
top: "conv1"
}
layer {
name: "conv1_bn"
type: "BN"
bottom: "conv1"
top: "conv1_bn"
param {
lr_mult: 1
decay_mult: 0
}
param {
lr_mult: 1
decay_mult: 0
}
bn_param {
scale_filler {
type: "constant"
value: 1
}
shift_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "drop1"
type: "Dropout"
bottom: "conv1_bn"
top: "conv1_bn"
dropout_param {
dropout_ratio: 0
}
}
layer {
name: "pool1"
type: "Pooling"
bottom: "conv1_bn"
top: "pool1"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "conv2"
type: "Convolution"
bottom: "pool1"
top: "conv2"
param {
lr_mult: 1
}
param {
lr_mult: 2
}
convolution_param {
num_output: 72
kernel_size: 3
stride: 1
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "relu2"
type: "ReLU"
bottom: "conv2"
top: "conv2"
}
layer {
name: "conv2_bn"
type: "BN"
bottom: "conv2"
top: "conv2_bn"
param {
lr_mult: 1
decay_mult: 0
}
param {
lr_mult: 1
decay_mult: 0
}
bn_param {
scale_filler {
type: "constant"
value: 1
}
shift_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "drop2"
type: "Dropout"
bottom: "conv2_bn"
top: "conv2_bn"
dropout_param {
dropout_ratio: 0.1
}
}
layer {
name: "pool2"
type: "Pooling"
bottom: "conv2_bn"
top: "pool2"
pooling_param {
pool: MAX
kernel_size: 2
stride: 2
}
}
layer {
name: "ip1"
type: "InnerProduct"
bottom: "pool2"
top: "ip1"
param {
lr_mult: 1
}
param {
lr_mult: 2
}
inner_product_param {
num_output: 320
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "relu3"
type: "ReLU"
bottom: "ip1"
top: "ip1"
}
layer {
name: "drop3"
type: "Dropout"
bottom: "ip1"
top: "ip1"
dropout_param {
dropout_ratio: 0.3
}
}
layer {
name: "ip2"
type: "InnerProduct"
bottom: "ip1"
top: "ip2"
param {
lr_mult: 1
}
param {
lr_mult: 2
}
inner_product_param {
num_output: 10
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "accuracy"
type: "Accuracy"
bottom: "ip2"
bottom: "label"
top: "accuracy"
include {
phase: TEST
}
}
layer {
name: "loss"
type: "SoftmaxWithLoss"
bottom: "ip2"
bottom: "label"
top: "loss"
}
I0805 14:52:42.609299 1392 layer_factory.hpp:74] Creating layer mnist
I0805 14:52:42.610299 1392 net.cpp:90] Creating Layer mnist
I0805 14:52:42.610299 1392 net.cpp:368] mnist -> data
I0805 14:52:42.610299 1392 net.cpp:368] mnist -> label
I0805 14:52:42.611299 1392 net.cpp:120] Setting up mnist
I0805 14:52:42.618300 1392 db.cpp:20] Opened leveldb mnist-test-leveldb
I0805 14:52:42.619300 1392 data_layer.cpp:52] output data size: 100,1,28,28
I0805 14:52:42.619300 1392 net.cpp:127] Top shape: 100 1 28 28 (78400)
I0805 14:52:42.620301 1392 net.cpp:127] Top shape: 100 (100)
I0805 14:52:42.620301 1392 layer_factory.hpp:74] Creating layer label_mnist_1_split
I0805 14:52:42.620301 1392 net.cpp:90] Creating Layer label_mnist_1_split
I0805 14:52:42.620301 1392 net.cpp:410] label_mnist_1_split <- label
I0805 14:52:42.621300 1392 net.cpp:368] label_mnist_1_split -> label_mnist_1_split_0
I0805 14:52:42.621300 1392 net.cpp:368] label_mnist_1_split -> label_mnist_1_split_1
I0805 14:52:42.621300 1392 net.cpp:120] Setting up label_mnist_1_split
I0805 14:52:42.622300 1392 net.cpp:127] Top shape: 100 (100)
I0805 14:52:42.622300 1392 net.cpp:127] Top shape: 100 (100)
I0805 14:52:42.622300 1392 layer_factory.hpp:74] Creating layer conv1
I0805 14:52:42.622300 1392 net.cpp:90] Creating Layer conv1
I0805 14:52:42.623301 1392 net.cpp:410] conv1 <- data
I0805 14:52:42.623301 1392 net.cpp:368] conv1 -> conv1
I0805 14:52:42.623301 1392 net.cpp:120] Setting up conv1
I0805 14:52:42.624300 1392 net.cpp:127] Top shape: 100 32 23 23 (1692800)
I0805 14:52:42.625300 1392 layer_factory.hpp:74] Creating layer relu1
I0805 14:52:42.625300 1392 net.cpp:90] Creating Layer relu1
I0805 14:52:42.625300 1392 net.cpp:410] relu1 <- conv1
I0805 14:52:42.625300 1392 net.cpp:357] relu1 -> conv1 (in-place)
I0805 14:52:42.626301 1392 net.cpp:120] Setting up relu1
I0805 14:52:42.626301 1392 net.cpp:127] Top shape: 100 32 23 23 (1692800)
I0805 14:52:42.626301 1392 layer_factory.hpp:74] Creating layer conv1_bn
I0805 14:52:42.626301 1392 net.cpp:90] Creating Layer conv1_bn
I0805 14:52:42.627300 1392 net.cpp:410] conv1_bn <- conv1
I0805 14:52:42.627300 1392 net.cpp:368] conv1_bn -> conv1_bn
I0805 14:52:42.627300 1392 net.cpp:120] Setting up conv1_bn
I0805 14:52:42.627300 1392 net.cpp:127] Top shape: 100 32 23 23 (1692800)
I0805 14:52:42.628300 1392 layer_factory.hpp:74] Creating layer drop1
I0805 14:52:42.628300 1392 net.cpp:90] Creating Layer drop1
I0805 14:52:42.628300 1392 net.cpp:410] drop1 <- conv1_bn
I0805 14:52:42.628300 1392 net.cpp:357] drop1 -> conv1_bn (in-place)
I0805 14:52:42.629300 1392 net.cpp:120] Setting up drop1
I0805 14:52:42.629300 1392 net.cpp:127] Top shape: 100 32 23 23 (1692800)
I0805 14:52:42.629300 1392 layer_factory.hpp:74] Creating layer pool1
I0805 14:52:42.629300 1392 net.cpp:90] Creating Layer pool1
I0805 14:52:42.630300 1392 net.cpp:410] pool1 <- conv1_bn
I0805 14:52:42.630300 1392 net.cpp:368] pool1 -> pool1
I0805 14:52:42.630300 1392 net.cpp:120] Setting up pool1
I0805 14:52:42.631300 1392 net.cpp:127] Top shape: 100 32 11 11 (387200)
I0805 14:52:42.631300 1392 layer_factory.hpp:74] Creating layer conv2
I0805 14:52:42.631300 1392 net.cpp:90] Creating Layer conv2
I0805 14:52:42.632300 1392 net.cpp:410] conv2 <- pool1
I0805 14:52:42.632300 1392 net.cpp:368] conv2 -> conv2
I0805 14:52:42.632300 1392 net.cpp:120] Setting up conv2
I0805 14:52:42.633301 1392 net.cpp:127] Top shape: 100 72 9 9 (583200)
I0805 14:52:42.633301 1392 layer_factory.hpp:74] Creating layer relu2
I0805 14:52:42.633301 1392 net.cpp:90] Creating Layer relu2
I0805 14:52:42.634301 1392 net.cpp:410] relu2 <- conv2
I0805 14:52:42.634301 1392 net.cpp:357] relu2 -> conv2 (in-place)
I0805 14:52:42.634301 1392 net.cpp:120] Setting up relu2
I0805 14:52:42.635301 1392 net.cpp:127] Top shape: 100 72 9 9 (583200)
I0805 14:52:42.635301 1392 layer_factory.hpp:74] Creating layer conv2_bn
I0805 14:52:42.635301 1392 net.cpp:90] Creating Layer conv2_bn
I0805 14:52:42.635301 1392 net.cpp:410] conv2_bn <- conv2
I0805 14:52:42.636301 1392 net.cpp:368] conv2_bn -> conv2_bn
I0805 14:52:42.636301 1392 net.cpp:120] Setting up conv2_bn
I0805 14:52:42.636301 1392 net.cpp:127] Top shape: 100 72 9 9 (583200)
I0805 14:52:42.636301 1392 layer_factory.hpp:74] Creating layer drop2
I0805 14:52:42.637301 1392 net.cpp:90] Creating Layer drop2
I0805 14:52:42.637301 1392 net.cpp:410] drop2 <- conv2_bn
I0805 14:52:42.637301 1392 net.cpp:357] drop2 -> conv2_bn (in-place)
I0805 14:52:42.637301 1392 net.cpp:120] Setting up drop2
I0805 14:52:42.638301 1392 net.cpp:127] Top shape: 100 72 9 9 (583200)
I0805 14:52:42.638301 1392 layer_factory.hpp:74] Creating layer pool2
I0805 14:52:42.638301 1392 net.cpp:90] Creating Layer pool2
I0805 14:52:42.638301 1392 net.cpp:410] pool2 <- conv2_bn
I0805 14:52:42.639302 1392 net.cpp:368] pool2 -> pool2
I0805 14:52:42.639302 1392 net.cpp:120] Setting up pool2
I0805 14:52:42.639302 1392 net.cpp:127] Top shape: 100 72 5 5 (180000)
I0805 14:52:42.639302 1392 layer_factory.hpp:74] Creating layer ip1
I0805 14:52:42.640301 1392 net.cpp:90] Creating Layer ip1
I0805 14:52:42.640301 1392 net.cpp:410] ip1 <- pool2
I0805 14:52:42.640301 1392 net.cpp:368] ip1 -> ip1
I0805 14:52:42.641301 1392 net.cpp:120] Setting up ip1
I0805 14:52:42.645301 1392 net.cpp:127] Top shape: 100 320 (32000)
I0805 14:52:42.645301 1392 layer_factory.hpp:74] Creating layer relu3
I0805 14:52:42.646301 1392 net.cpp:90] Creating Layer relu3
I0805 14:52:42.646301 1392 net.cpp:410] relu3 <- ip1
I0805 14:52:42.646301 1392 net.cpp:357] relu3 -> ip1 (in-place)
I0805 14:52:42.647301 1392 net.cpp:120] Setting up relu3
I0805 14:52:42.647301 1392 net.cpp:127] Top shape: 100 320 (32000)
I0805 14:52:42.648301 1392 layer_factory.hpp:74] Creating layer drop3
I0805 14:52:42.648301 1392 net.cpp:90] Creating Layer drop3
I0805 14:52:42.648301 1392 net.cpp:410] drop3 <- ip1
I0805 14:52:42.649302 1392 net.cpp:357] drop3 -> ip1 (in-place)
I0805 14:52:42.649302 1392 net.cpp:120] Setting up drop3
I0805 14:52:42.649302 1392 net.cpp:127] Top shape: 100 320 (32000)
I0805 14:52:42.649302 1392 layer_factory.hpp:74] Creating layer ip2
I0805 14:52:42.650302 1392 net.cpp:90] Creating Layer ip2
I0805 14:52:42.650302 1392 net.cpp:410] ip2 <- ip1
I0805 14:52:42.650302 1392 net.cpp:368] ip2 -> ip2
I0805 14:52:42.650302 1392 net.cpp:120] Setting up ip2
I0805 14:52:42.651303 1392 net.cpp:127] Top shape: 100 10 (1000)
I0805 14:52:42.651303 1392 layer_factory.hpp:74] Creating layer ip2_ip2_0_split
I0805 14:52:42.651303 1392 net.cpp:90] Creating Layer ip2_ip2_0_split
I0805 14:52:42.651303 1392 net.cpp:410] ip2_ip2_0_split <- ip2
I0805 14:52:42.652302 1392 net.cpp:368] ip2_ip2_0_split -> ip2_ip2_0_split_0
I0805 14:52:42.652302 1392 net.cpp:368] ip2_ip2_0_split -> ip2_ip2_0_split_1
I0805 14:52:42.652302 1392 net.cpp:120] Setting up ip2_ip2_0_split
I0805 14:52:42.652302 1392 net.cpp:127] Top shape: 100 10 (1000)
I0805 14:52:42.653302 1392 net.cpp:127] Top shape: 100 10 (1000)
I0805 14:52:42.653302 1392 layer_factory.hpp:74] Creating layer accuracy
I0805 14:52:42.653302 1392 net.cpp:90] Creating Layer accuracy
I0805 14:52:42.654302 1392 net.cpp:410] accuracy <- ip2_ip2_0_split_0
I0805 14:52:42.654302 1392 net.cpp:410] accuracy <- label_mnist_1_split_0
I0805 14:52:42.654302 1392 net.cpp:368] accuracy -> accuracy
I0805 14:52:42.655303 1392 net.cpp:120] Setting up accuracy
I0805 14:52:42.655303 1392 net.cpp:127] Top shape: (1)
I0805 14:52:42.655303 1392 layer_factory.hpp:74] Creating layer loss
I0805 14:52:42.655303 1392 net.cpp:90] Creating Layer loss
I0805 14:52:42.656302 1392 net.cpp:410] loss <- ip2_ip2_0_split_1
I0805 14:52:42.656302 1392 net.cpp:410] loss <- label_mnist_1_split_1
I0805 14:52:42.656302 1392 net.cpp:368] loss -> loss
I0805 14:52:42.657302 1392 net.cpp:120] Setting up loss
I0805 14:52:42.657302 1392 layer_factory.hpp:74] Creating layer loss
I0805 14:52:42.658303 1392 net.cpp:127] Top shape: (1)
I0805 14:52:42.658303 1392 net.cpp:129] with loss weight 1
I0805 14:52:42.659302 1392 net.cpp:192] loss needs backward computation.
I0805 14:52:42.659302 1392 net.cpp:194] accuracy does not need backward computation.
I0805 14:52:42.659302 1392 net.cpp:192] ip2_ip2_0_split needs backward computation.
I0805 14:52:42.659302 1392 net.cpp:192] ip2 needs backward computation.
I0805 14:52:42.660302 1392 net.cpp:192] drop3 needs backward computation.
I0805 14:52:42.660302 1392 net.cpp:192] relu3 needs backward computation.
I0805 14:52:42.660302 1392 net.cpp:192] ip1 needs backward computation.
I0805 14:52:42.660302 1392 net.cpp:192] pool2 needs backward computation.
I0805 14:52:42.661303 1392 net.cpp:192] drop2 needs backward computation.
I0805 14:52:42.661303 1392 net.cpp:192] conv2_bn needs backward computation.
I0805 14:52:42.661303 1392 net.cpp:192] relu2 needs backward computation.
I0805 14:52:42.661303 1392 net.cpp:192] conv2 needs backward computation.
I0805 14:52:42.662302 1392 net.cpp:192] pool1 needs backward computation.
I0805 14:52:42.662302 1392 net.cpp:192] drop1 needs backward computation.
I0805 14:52:42.662302 1392 net.cpp:192] conv1_bn needs backward computation.
I0805 14:52:42.663302 1392 net.cpp:192] relu1 needs backward computation.
I0805 14:52:42.663302 1392 net.cpp:192] conv1 needs backward computation.
I0805 14:52:42.663302 1392 net.cpp:194] label_mnist_1_split does not need backward computation.
I0805 14:52:42.664302 1392 net.cpp:194] mnist does not need backward computation.
I0805 14:52:42.664302 1392 net.cpp:235] This network produces output accuracy
I0805 14:52:42.664302 1392 net.cpp:235] This network produces output loss
I0805 14:52:42.664302 1392 net.cpp:482] Collecting Learning Rate and Weight Decay.
I0805 14:52:42.665302 1392 net.cpp:247] Network initialization done.
I0805 14:52:42.665302 1392 net.cpp:248] Memory required for data: 39395608
I0805 14:52:42.665302 1392 solver.cpp:42] Solver scaffolding done.
I0805 14:52:42.665302 1392 solver.cpp:250] Solving LeNet
I0805 14:52:42.666302 1392 solver.cpp:251] Learning Rate Policy: inv
I0805 14:52:42.667302 1392 solver.cpp:294] Iteration 0, Testing net (#0)
I0805 14:52:43.083326 1392 solver.cpp:343] Test net output #0: accuracy = 0.1007
I0805 14:52:43.083326 1392 solver.cpp:343] Test net output #1: loss = 78.5417 (* 1 = 78.5417 loss)
I0805 14:52:43.106328 1392 solver.cpp:214] Iteration 0, loss = 2.87422
I0805 14:52:43.107328 1392 solver.cpp:229] Train net output #0: loss = 2.87422 (* 1 = 2.87422 loss)
I0805 14:52:43.108328 1392 solver.cpp:486] Iteration 0, lr = 0.04
I0805 14:52:43.809368 1392 solver.cpp:214] Iteration 100, loss = 0.220041
I0805 14:52:43.810369 1392 solver.cpp:229] Train net output #0: loss = 0.220041 (* 1 = 0.220041 loss)
I0805 14:52:43.810369 1392 solver.cpp:486] Iteration 100, lr = 0.0392934
I0805 14:52:44.502408 1392 solver.cpp:214] Iteration 200, loss = 0.126289
I0805 14:52:44.503408 1392 solver.cpp:229] Train net output #0: loss = 0.126289 (* 1 = 0.126289 loss)
I0805 14:52:44.504408 1392 solver.cpp:486] Iteration 200, lr = 0.0386127
I0805 14:52:45.194447 1392 solver.cpp:214] Iteration 300, loss = 0.139079
I0805 14:52:45.195447 1392 solver.cpp:229] Train net output #0: loss = 0.139079 (* 1 = 0.139079 loss)
I0805 14:52:45.196447 1392 solver.cpp:486] Iteration 300, lr = 0.0379564
I0805 14:52:45.889487 1392 solver.cpp:214] Iteration 400, loss = 0.0402969
I0805 14:52:45.890487 1392 solver.cpp:229] Train net output #0: loss = 0.040297 (* 1 = 0.040297 loss)
I0805 14:52:45.891487 1392 solver.cpp:486] Iteration 400, lr = 0.0373232
I0805 14:52:46.586527 1392 solver.cpp:214] Iteration 500, loss = 0.0935015
I0805 14:52:46.586527 1392 solver.cpp:229] Train net output #0: loss = 0.0935016 (* 1 = 0.0935016 loss)
I0805 14:52:46.587527 1392 solver.cpp:486] Iteration 500, lr = 0.0367119
I0805 14:52:47.281566 1392 solver.cpp:214] Iteration 600, loss = 0.0859815
I0805 14:52:47.281566 1392 solver.cpp:229] Train net output #0: loss = 0.0859815 (* 1 = 0.0859815 loss)
I0805 14:52:47.282567 1392 solver.cpp:486] Iteration 600, lr = 0.0361213
I0805 14:52:47.979606 1392 solver.cpp:214] Iteration 700, loss = 0.0612219
I0805 14:52:47.979606 1392 solver.cpp:229] Train net output #0: loss = 0.0612219 (* 1 = 0.0612219 loss)
I0805 14:52:47.980607 1392 solver.cpp:486] Iteration 700, lr = 0.0355505
I0805 14:52:48.675647 1392 solver.cpp:214] Iteration 800, loss = 0.18401
I0805 14:52:48.676646 1392 solver.cpp:229] Train net output #0: loss = 0.18401 (* 1 = 0.18401 loss)
I0805 14:52:48.677646 1392 solver.cpp:486] Iteration 800, lr = 0.0349984
I0805 14:52:49.369686 1392 solver.cpp:214] Iteration 900, loss = 0.213497
I0805 14:52:49.369686 1392 solver.cpp:229] Train net output #0: loss = 0.213497 (* 1 = 0.213497 loss)
I0805 14:52:49.370687 1392 solver.cpp:486] Iteration 900, lr = 0.034464
I0805 14:52:50.060725 1392 solver.cpp:294] Iteration 1000, Testing net (#0)
I0805 14:52:50.451748 1392 solver.cpp:343] Test net output #0: accuracy = 0.9893
I0805 14:52:50.452749 1392 solver.cpp:343] Test net output #1: loss = 0.0339307 (* 1 = 0.0339307 loss)
I0805 14:52:50.456748 1392 solver.cpp:214] Iteration 1000, loss = 0.0163326
I0805 14:52:50.457748 1392 solver.cpp:229] Train net output #0: loss = 0.0163326 (* 1 = 0.0163326 loss)
I0805 14:52:50.457748 1392 solver.cpp:486] Iteration 1000, lr = 0.0339466
I0805 14:52:51.157788 1392 solver.cpp:214] Iteration 1100, loss = 0.00104439
I0805 14:52:51.158788 1392 solver.cpp:229] Train net output #0: loss = 0.00104446 (* 1 = 0.00104446 loss)
I0805 14:52:51.159788 1392 solver.cpp:486] Iteration 1100, lr = 0.0334454
I0805 14:52:51.853828 1392 solver.cpp:214] Iteration 1200, loss = 0.0300438
I0805 14:52:51.854828 1392 solver.cpp:229] Train net output #0: loss = 0.0300439 (* 1 = 0.0300439 loss)
I0805 14:52:51.854828 1392 solver.cpp:486] Iteration 1200, lr = 0.0329595
I0805 14:52:52.543867 1392 solver.cpp:214] Iteration 1300, loss = 0.00561851
I0805 14:52:52.544867 1392 solver.cpp:229] Train net output #0: loss = 0.00561857 (* 1 = 0.00561857 loss)
I0805 14:52:52.544867 1392 solver.cpp:486] Iteration 1300, lr = 0.0324883
I0805 14:52:53.241907 1392 solver.cpp:214] Iteration 1400, loss = 0.00394553
I0805 14:52:53.241907 1392 solver.cpp:229] Train net output #0: loss = 0.00394557 (* 1 = 0.00394557 loss)
I0805 14:52:53.242908 1392 solver.cpp:486] Iteration 1400, lr = 0.032031
I0805 14:52:53.938947 1392 solver.cpp:214] Iteration 1500, loss = 0.0390139
I0805 14:52:53.939947 1392 solver.cpp:229] Train net output #0: loss = 0.039014 (* 1 = 0.039014 loss)
I0805 14:52:53.939947 1392 solver.cpp:486] Iteration 1500, lr = 0.0315872
I0805 14:52:54.632987 1392 solver.cpp:214] Iteration 1600, loss = 0.045935
I0805 14:52:54.633987 1392 solver.cpp:229] Train net output #0: loss = 0.0459351 (* 1 = 0.0459351 loss)
I0805 14:52:54.633987 1392 solver.cpp:486] Iteration 1600, lr = 0.0311561
I0805 14:52:55.326027 1392 solver.cpp:214] Iteration 1700, loss = 0.0930766
I0805 14:52:55.326027 1392 solver.cpp:229] Train net output #0: loss = 0.0930766 (* 1 = 0.0930766 loss)
I0805 14:52:55.327028 1392 solver.cpp:486] Iteration 1700, lr = 0.0307373
I0805 14:52:56.022066 1392 solver.cpp:214] Iteration 1800, loss = 0.00682616
I0805 14:52:56.022066 1392 solver.cpp:229] Train net output #0: loss = 0.00682622 (* 1 = 0.00682622 loss)
I0805 14:52:56.023066 1392 solver.cpp:486] Iteration 1800, lr = 0.0303302
I0805 14:52:56.719106 1392 solver.cpp:214] Iteration 1900, loss = 0.122121
I0805 14:52:56.719106 1392 solver.cpp:229] Train net output #0: loss = 0.122121 (* 1 = 0.122121 loss)
I0805 14:52:56.720106 1392 solver.cpp:486] Iteration 1900, lr = 0.0299343
I0805 14:52:57.405145 1392 solver.cpp:294] Iteration 2000, Testing net (#0)
I0805 14:52:57.808168 1392 solver.cpp:343] Test net output #0: accuracy = 0.9898
I0805 14:52:57.809170 1392 solver.cpp:343] Test net output #1: loss = 0.0301438 (* 1 = 0.0301438 loss)
I0805 14:52:57.813169 1392 solver.cpp:214] Iteration 2000, loss = 0.0324452
I0805 14:52:57.814169 1392 solver.cpp:229] Train net output #0: loss = 0.0324453 (* 1 = 0.0324453 loss)
I0805 14:52:57.814169 1392 solver.cpp:486] Iteration 2000, lr = 0.0295491
I0805 14:52:58.507208 1392 solver.cpp:214] Iteration 2100, loss = 0.0160458
I0805 14:52:58.507208 1392 solver.cpp:229] Train net output #0: loss = 0.016046 (* 1 = 0.016046 loss)
I0805 14:52:58.508209 1392 solver.cpp:486] Iteration 2100, lr = 0.0291743
I0805 14:52:59.200248 1392 solver.cpp:214] Iteration 2200, loss = 0.00501125
I0805 14:52:59.201248 1392 solver.cpp:229] Train net output #0: loss = 0.00501135 (* 1 = 0.00501135 loss)
I0805 14:52:59.201248 1392 solver.cpp:486] Iteration 2200, lr = 0.0288094
I0805 14:52:59.896288 1392 solver.cpp:214] Iteration 2300, loss = 0.0237116
I0805 14:52:59.896288 1392 solver.cpp:229] Train net output #0: loss = 0.0237118 (* 1 = 0.0237118 loss)
I0805 14:52:59.897289 1392 solver.cpp:486] Iteration 2300, lr = 0.0284539
I0805 14:53:00.594328 1392 solver.cpp:214] Iteration 2400, loss = 0.0360984
I0805 14:53:00.595329 1392 solver.cpp:229] Train net output #0: loss = 0.0360985 (* 1 = 0.0360985 loss)
I0805 14:53:00.596328 1392 solver.cpp:486] Iteration 2400, lr = 0.0281076
I0805 14:53:01.289368 1392 solver.cpp:214] Iteration 2500, loss = 0.0620604
I0805 14:53:01.289368 1392 solver.cpp:229] Train net output #0: loss = 0.0620605 (* 1 = 0.0620605 loss)
I0805 14:53:01.290369 1392 solver.cpp:486] Iteration 2500, lr = 0.0277701
I0805 14:53:01.985407 1392 solver.cpp:214] Iteration 2600, loss = 0.036869
I0805 14:53:01.985407 1392 solver.cpp:229] Train net output #0: loss = 0.0368692 (* 1 = 0.0368692 loss)
I0805 14:53:01.986407 1392 solver.cpp:486] Iteration 2600, lr = 0.0274411
I0805 14:53:02.676447 1392 solver.cpp:214] Iteration 2700, loss = 0.106417
I0805 14:53:02.676447 1392 solver.cpp:229] Train net output #0: loss = 0.106417 (* 1 = 0.106417 loss)
I0805 14:53:02.677448 1392 solver.cpp:486] Iteration 2700, lr = 0.0271201
I0805 14:53:03.368487 1392 solver.cpp:214] Iteration 2800, loss = 0.000549857
I0805 14:53:03.369488 1392 solver.cpp:229] Train net output #0: loss = 0.000549995 (* 1 = 0.000549995 loss)
I0805 14:53:03.369488 1392 solver.cpp:486] Iteration 2800, lr = 0.026807
I0805 14:53:04.065526 1392 solver.cpp:214] Iteration 2900, loss = 0.0262977
I0805 14:53:04.065526 1392 solver.cpp:229] Train net output #0: loss = 0.0262979 (* 1 = 0.0262979 loss)
I0805 14:53:04.066526 1392 solver.cpp:486] Iteration 2900, lr = 0.0265014
I0805 14:53:04.755566 1392 solver.cpp:294] Iteration 3000, Testing net (#0)
I0805 14:53:05.154589 1392 solver.cpp:343] Test net output #0: accuracy = 0.9933
I0805 14:53:05.155589 1392 solver.cpp:343] Test net output #1: loss = 0.020941 (* 1 = 0.020941 loss)
I0805 14:53:05.159590 1392 solver.cpp:214] Iteration 3000, loss = 0.00563839
I0805 14:53:05.160589 1392 solver.cpp:229] Train net output #0: loss = 0.00563854 (* 1 = 0.00563854 loss)
I0805 14:53:05.160589 1392 solver.cpp:486] Iteration 3000, lr = 0.0262031
I0805 14:53:05.861629 1392 solver.cpp:214] Iteration 3100, loss = 0.00378355
I0805 14:53:05.862629 1392 solver.cpp:229] Train net output #0: loss = 0.0037837 (* 1 = 0.0037837 loss)
I0805 14:53:05.863629 1392 solver.cpp:486] Iteration 3100, lr = 0.0259117
I0805 14:53:06.556669 1392 solver.cpp:214] Iteration 3200, loss = 0.00495449
I0805 14:53:06.557669 1392 solver.cpp:229] Train net output #0: loss = 0.00495464 (* 1 = 0.00495464 loss)
I0805 14:53:06.557669 1392 solver.cpp:486] Iteration 3200, lr = 0.0256272
I0805 14:53:07.251709 1392 solver.cpp:214] Iteration 3300, loss = 0.0154928
I0805 14:53:07.251709 1392 solver.cpp:229] Train net output #0: loss = 0.0154929 (* 1 = 0.0154929 loss)
I0805 14:53:07.252709 1392 solver.cpp:486] Iteration 3300, lr = 0.0253491
I0805 14:53:07.943748 1392 solver.cpp:214] Iteration 3400, loss = 0.0010491
I0805 14:53:07.944748 1392 solver.cpp:229] Train net output #0: loss = 0.00104925 (* 1 = 0.00104925 loss)
I0805 14:53:07.945749 1392 solver.cpp:486] Iteration 3400, lr = 0.0250774
I0805 14:53:08.641788 1392 solver.cpp:214] Iteration 3500, loss = 0.00815444
I0805 14:53:08.642788 1392 solver.cpp:229] Train net output #0: loss = 0.00815459 (* 1 = 0.00815459 loss)
I0805 14:53:08.642788 1392 solver.cpp:486] Iteration 3500, lr = 0.0248117
I0805 14:53:09.335829 1392 solver.cpp:214] Iteration 3600, loss = 0.00881401
I0805 14:53:09.335829 1392 solver.cpp:229] Train net output #0: loss = 0.00881419 (* 1 = 0.00881419 loss)
I0805 14:53:09.336828 1392 solver.cpp:486] Iteration 3600, lr = 0.0245519
I0805 14:53:10.028867 1392 solver.cpp:214] Iteration 3700, loss = 0.0718557
I0805 14:53:10.028867 1392 solver.cpp:229] Train net output #0: loss = 0.0718559 (* 1 = 0.0718559 loss)
I0805 14:53:10.029868 1392 solver.cpp:486] Iteration 3700, lr = 0.0242977
I0805 14:53:10.723907 1392 solver.cpp:214] Iteration 3800, loss = 0.0243846
I0805 14:53:10.723907 1392 solver.cpp:229] Train net output #0: loss = 0.0243848 (* 1 = 0.0243848 loss)
I0805 14:53:10.724907 1392 solver.cpp:486] Iteration 3800, lr = 0.0240491
I0805 14:53:11.417948 1392 solver.cpp:214] Iteration 3900, loss = 0.010712
I0805 14:53:11.418947 1392 solver.cpp:229] Train net output #0: loss = 0.0107122 (* 1 = 0.0107122 loss)
I0805 14:53:11.418947 1392 solver.cpp:486] Iteration 3900, lr = 0.0238058
I0805 14:53:12.104986 1392 solver.cpp:294] Iteration 4000, Testing net (#0)
I0805 14:53:12.506009 1392 solver.cpp:343] Test net output #0: accuracy = 0.9941
I0805 14:53:12.507009 1392 solver.cpp:343] Test net output #1: loss = 0.0183649 (* 1 = 0.0183649 loss)
I0805 14:53:12.512011 1392 solver.cpp:214] Iteration 4000, loss = 0.0114788
I0805 14:53:12.513010 1392 solver.cpp:229] Train net output #0: loss = 0.0114789 (* 1 = 0.0114789 loss)
I0805 14:53:12.513010 1392 solver.cpp:486] Iteration 4000, lr = 0.0235676
I0805 14:53:13.210049 1392 solver.cpp:214] Iteration 4100, loss = 0.000130875
I0805 14:53:13.211050 1392 solver.cpp:229] Train net output #0: loss = 0.000131057 (* 1 = 0.000131057 loss)
I0805 14:53:13.212050 1392 solver.cpp:486] Iteration 4100, lr = 0.0233344
I0805 14:53:13.904089 1392 solver.cpp:214] Iteration 4200, loss = 0.00127553
I0805 14:53:13.905089 1392 solver.cpp:229] Train net output #0: loss = 0.00127569 (* 1 = 0.00127569 loss)
I0805 14:53:13.905089 1392 solver.cpp:486] Iteration 4200, lr = 0.023106
I0805 14:53:14.598129 1392 solver.cpp:214] Iteration 4300, loss = 0.00505828
I0805 14:53:14.598129 1392 solver.cpp:229] Train net output #0: loss = 0.00505844 (* 1 = 0.00505844 loss)
I0805 14:53:14.599129 1392 solver.cpp:486] Iteration 4300, lr = 0.0228822
I0805 14:53:15.295169 1392 solver.cpp:214] Iteration 4400, loss = 0.000817688
I0805 14:53:15.295169 1392 solver.cpp:229] Train net output #0: loss = 0.000817851 (* 1 = 0.000817851 loss)
I0805 14:53:15.296169 1392 solver.cpp:486] Iteration 4400, lr = 0.022663
I0805 14:53:15.997210 1392 solver.cpp:214] Iteration 4500, loss = 0.00614459
I0805 14:53:15.998209 1392 solver.cpp:229] Train net output #0: loss = 0.00614476 (* 1 = 0.00614476 loss)
I0805 14:53:15.998209 1392 solver.cpp:486] Iteration 4500, lr = 0.0224482
I0805 14:53:16.708250 1392 solver.cpp:214] Iteration 4600, loss = 0.0246302
I0805 14:53:16.709250 1392 solver.cpp:229] Train net output #0: loss = 0.0246304 (* 1 = 0.0246304 loss)
I0805 14:53:16.710250 1392 solver.cpp:486] Iteration 4600, lr = 0.0222377
I0805 14:53:17.421290 1392 solver.cpp:214] Iteration 4700, loss = 0.0138153
I0805 14:53:17.421290 1392 solver.cpp:229] Train net output #0: loss = 0.0138154 (* 1 = 0.0138154 loss)
I0805 14:53:17.422291 1392 solver.cpp:486] Iteration 4700, lr = 0.0220312
I0805 14:53:18.138331 1392 solver.cpp:214] Iteration 4800, loss = 0.00829904
I0805 14:53:18.138331 1392 solver.cpp:229] Train net output #0: loss = 0.00829922 (* 1 = 0.00829922 loss)
I0805 14:53:18.139331 1392 solver.cpp:486] Iteration 4800, lr = 0.0218288
I0805 14:53:18.856372 1392 solver.cpp:214] Iteration 4900, loss = 0.000418008
I0805 14:53:18.856372 1392 solver.cpp:229] Train net output #0: loss = 0.000418191 (* 1 = 0.000418191 loss)
I0805 14:53:18.857372 1392 solver.cpp:486] Iteration 4900, lr = 0.0216302
I0805 14:53:19.573413 1392 solver.cpp:361] Snapshotting to lenet_iter_5000.caffemodel
I0805 14:53:19.585414 1392 solver.cpp:369] Snapshotting solver state to lenet_iter_5000.solverstate
I0805 14:53:19.593415 1392 solver.cpp:294] Iteration 5000, Testing net (#0)
I0805 14:53:20.009438 1392 solver.cpp:343] Test net output #0: accuracy = 0.9937
I0805 14:53:20.010438 1392 solver.cpp:343] Test net output #1: loss = 0.0210218 (* 1 = 0.0210218 loss)
I0805 14:53:20.014439 1392 solver.cpp:214] Iteration 5000, loss = 0.0439826
I0805 14:53:20.015439 1392 solver.cpp:229] Train net output #0: loss = 0.0439828 (* 1 = 0.0439828 loss)
I0805 14:53:20.016439 1392 solver.cpp:486] Iteration 5000, lr = 0.0214355
I0805 14:53:20.739480 1392 solver.cpp:214] Iteration 5100, loss = 0.0212865
I0805 14:53:20.739480 1392 solver.cpp:229] Train net output #0: loss = 0.0212867 (* 1 = 0.0212867 loss)
I0805 14:53:20.740480 1392 solver.cpp:486] Iteration 5100, lr = 0.0212444
I0805 14:53:21.462522 1392 solver.cpp:214] Iteration 5200, loss = 0.00098768
I0805 14:53:21.463521 1392 solver.cpp:229] Train net output #0: loss = 0.000987855 (* 1 = 0.000987855 loss)
I0805 14:53:21.464522 1392 solver.cpp:486] Iteration 5200, lr = 0.0210568
I0805 14:53:22.186563 1392 solver.cpp:214] Iteration 5300, loss = 0.000718615
I0805 14:53:22.187563 1392 solver.cpp:229] Train net output #0: loss = 0.000718786 (* 1 = 0.000718786 loss)
I0805 14:53:22.187563 1392 solver.cpp:486] Iteration 5300, lr = 0.0208727
I0805 14:53:22.906605 1392 solver.cpp:214] Iteration 5400, loss = 0.00133868
I0805 14:53:22.907604 1392 solver.cpp:229] Train net output #0: loss = 0.00133886 (* 1 = 0.00133886 loss)
I0805 14:53:22.908604 1392 solver.cpp:486] Iteration 5400, lr = 0.020692
I0805 14:53:23.627645 1392 solver.cpp:214] Iteration 5500, loss = 0.000897586
I0805 14:53:23.628645 1392 solver.cpp:229] Train net output #0: loss = 0.000897764 (* 1 = 0.000897764 loss)
I0805 14:53:23.629645 1392 solver.cpp:486] Iteration 5500, lr = 0.0205146
I0805 14:53:24.359688 1392 solver.cpp:214] Iteration 5600, loss = 9.13558e-005
I0805 14:53:24.360687 1392 solver.cpp:229] Train net output #0: loss = 9.15377e-005 (* 1 = 9.15377e-005 loss)
I0805 14:53:24.360687 1392 solver.cpp:486] Iteration 5600, lr = 0.0203403
I0805 14:53:25.095729 1392 solver.cpp:214] Iteration 5700, loss = 0.0016139
I0805 14:53:25.096729 1392 solver.cpp:229] Train net output #0: loss = 0.00161407 (* 1 = 0.00161407 loss)
I0805 14:53:25.097729 1392 solver.cpp:486] Iteration 5700, lr = 0.0201692
I0805 14:53:25.830771 1392 solver.cpp:214] Iteration 5800, loss = 0.00897269
I0805 14:53:25.830771 1392 solver.cpp:229] Train net output #0: loss = 0.00897286 (* 1 = 0.00897286 loss)
I0805 14:53:25.831771 1392 solver.cpp:486] Iteration 5800, lr = 0.020001
I0805 14:53:26.567813 1392 solver.cpp:214] Iteration 5900, loss = 0.00267578
I0805 14:53:26.567813 1392 solver.cpp:229] Train net output #0: loss = 0.00267595 (* 1 = 0.00267595 loss)
I0805 14:53:26.568814 1392 solver.cpp:486] Iteration 5900, lr = 0.0198358
I0805 14:53:27.298856 1392 solver.cpp:294] Iteration 6000, Testing net (#0)
I0805 14:53:27.719879 1392 solver.cpp:343] Test net output #0: accuracy = 0.9944
I0805 14:53:27.720880 1392 solver.cpp:343] Test net output #1: loss = 0.0177348 (* 1 = 0.0177348 loss)
I0805 14:53:27.725880 1392 solver.cpp:214] Iteration 6000, loss = 0.00269533
I0805 14:53:27.725880 1392 solver.cpp:229] Train net output #0: loss = 0.0026955 (* 1 = 0.0026955 loss)
I0805 14:53:27.726881 1392 solver.cpp:486] Iteration 6000, lr = 0.0196734
I0805 14:53:28.466922 1392 solver.cpp:214] Iteration 6100, loss = 0.001992
I0805 14:53:28.467922 1392 solver.cpp:229] Train net output #0: loss = 0.00199218 (* 1 = 0.00199218 loss)
I0805 14:53:28.468922 1392 solver.cpp:486] Iteration 6100, lr = 0.0195138
I0805 14:53:29.202965 1392 solver.cpp:214] Iteration 6200, loss = 0.0042955
I0805 14:53:29.203964 1392 solver.cpp:229] Train net output #0: loss = 0.00429569 (* 1 = 0.00429569 loss)
I0805 14:53:29.204964 1392 solver.cpp:486] Iteration 6200, lr = 0.0193569
I0805 14:53:29.944007 1392 solver.cpp:214] Iteration 6300, loss = 0.00114569
I0805 14:53:29.944007 1392 solver.cpp:229] Train net output #0: loss = 0.0011459 (* 1 = 0.0011459 loss)
I0805 14:53:29.945008 1392 solver.cpp:486] Iteration 6300, lr = 0.0192027
I0805 14:53:30.684049 1392 solver.cpp:214] Iteration 6400, loss = 0.000806044
I0805 14:53:30.685050 1392 solver.cpp:229] Train net output #0: loss = 0.000806251 (* 1 = 0.000806251 loss)
I0805 14:53:30.685050 1392 solver.cpp:486] Iteration 6400, lr = 0.019051
I0805 14:53:31.425091 1392 solver.cpp:214] Iteration 6500, loss = 0.00453469
I0805 14:53:31.425091 1392 solver.cpp:229] Train net output #0: loss = 0.00453489 (* 1 = 0.00453489 loss)
I0805 14:53:31.426091 1392 solver.cpp:486] Iteration 6500, lr = 0.0189019
I0805 14:53:32.166134 1392 solver.cpp:214] Iteration 6600, loss = 0.0772493
I0805 14:53:32.167135 1392 solver.cpp:229] Train net output #0: loss = 0.0772495 (* 1 = 0.0772495 loss)
I0805 14:53:32.168134 1392 solver.cpp:486] Iteration 6600, lr = 0.0187552
I0805 14:53:32.902176 1392 solver.cpp:214] Iteration 6700, loss = 0.0040816
I0805 14:53:32.903177 1392 solver.cpp:229] Train net output #0: loss = 0.0040818 (* 1 = 0.0040818 loss)
I0805 14:53:32.903177 1392 solver.cpp:486] Iteration 6700, lr = 0.0186108
I0805 14:53:33.641218 1392 solver.cpp:214] Iteration 6800, loss = 0.00129661