-
Notifications
You must be signed in to change notification settings - Fork 13
/
condition_lstm.log
1246 lines (1245 loc) · 43.4 KB
/
condition_lstm.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
Using cuDNN version 5110 on context None
Mapped name None to device cuda: GeForce GTX 1070 (0000:01:00.0)
Loading data ...
Loading train ...
550152
550149
549364
Loading dev ...
10000
10000
9842
Loading test ...
10000
10000
9824
num_epochs: 20
k: 100
batch_size: 30
display_frequency: 1000
save_frequency: 1000
load previous: False
attention: False
word_by_word: False
Building network ...
unchanged_W.shape: (34283, 300)
oov_in_train_W.shape: (9166, 300)
apply dropout rate 0.2 to decoder
apply dropout mask id 140553116728120 to embedding matrix ...
dropout rate is 0.2
input var is hypo_var
apply dropout mask id 140553116728120 to embedding matrix ...
dropout rate is 0.2
input var is premise_var
Computing updates ...
Compiling functions ...
Training ...
train_df.shape: (549364, 4)
dev_df.shape: (9842, 4)
test_df.shape: (9824, 4)
Starting training...
Seen 30000 samples, time used: 21.720s
current training loss: 0.958078
current training accuracy: 0.554000
saving to ..., time used 21.720s
Seen 60000 samples, time used: 42.674s
current training loss: 0.923707
current training accuracy: 0.581733
saving to ..., time used 20.822s
Seen 90000 samples, time used: 63.609s
current training loss: 0.902004
current training accuracy: 0.596811
saving to ..., time used 20.799s
Seen 120000 samples, time used: 84.638s
current training loss: 0.885108
current training accuracy: 0.607783
saving to ..., time used 20.848s
Seen 150000 samples, time used: 105.824s
current training loss: 0.873300
current training accuracy: 0.616267
saving to ..., time used 21.049s
Seen 180000 samples, time used: 126.915s
current training loss: 0.863084
current training accuracy: 0.623189
saving to ..., time used 20.914s
Seen 210000 samples, time used: 148.019s
current training loss: 0.853904
current training accuracy: 0.628824
saving to ..., time used 20.968s
Seen 240000 samples, time used: 169.109s
current training loss: 0.846369
current training accuracy: 0.633738
saving to ..., time used 20.956s
Seen 270000 samples, time used: 189.900s
current training loss: 0.838683
current training accuracy: 0.638467
saving to ..., time used 20.658s
Seen 300000 samples, time used: 210.762s
current training loss: 0.832061
current training accuracy: 0.642567
saving to ..., time used 20.710s
Seen 330000 samples, time used: 231.617s
current training loss: 0.825482
current training accuracy: 0.646758
saving to ..., time used 20.721s
Seen 360000 samples, time used: 252.512s
current training loss: 0.820523
current training accuracy: 0.650128
saving to ..., time used 20.758s
Seen 390000 samples, time used: 273.429s
current training loss: 0.814548
current training accuracy: 0.653818
saving to ..., time used 20.639s
Seen 420000 samples, time used: 294.459s
current training loss: 0.809283
current training accuracy: 0.657224
saving to ..., time used 20.898s
Seen 450000 samples, time used: 315.497s
current training loss: 0.803960
current training accuracy: 0.660382
saving to ..., time used 20.903s
Seen 480000 samples, time used: 336.654s
current training loss: 0.798672
current training accuracy: 0.663725
saving to ..., time used 20.896s
Seen 510000 samples, time used: 357.499s
current training loss: 0.793693
current training accuracy: 0.666788
saving to ..., time used 20.709s
Seen 540000 samples, time used: 378.518s
current training loss: 0.789148
current training accuracy: 0.669613
saving to ..., time used 20.880s
Epoch 1 of 20 took 386.563s
training loss: 0.787587
training accuracy: 67.06 %
validation loss: 0.658322
validation accuracy: 72.45 %
test loss: 0.654460
test accuracy: 72.61 %
saving to ./snli/condition_model_epoch1.npz
Seen 30000 samples, time used: 20.720s
current training loss: 0.684588
current training accuracy: 0.728933
saving to ..., time used 20.720s
Seen 60000 samples, time used: 41.980s
current training loss: 0.688018
current training accuracy: 0.727950
saving to ..., time used 20.949s
Seen 90000 samples, time used: 63.032s
current training loss: 0.684467
current training accuracy: 0.729811
saving to ..., time used 20.758s
Seen 120000 samples, time used: 83.974s
current training loss: 0.684357
current training accuracy: 0.730100
saving to ..., time used 20.808s
Seen 150000 samples, time used: 105.085s
current training loss: 0.681023
current training accuracy: 0.731947
saving to ..., time used 20.969s
Seen 180000 samples, time used: 126.254s
current training loss: 0.678091
current training accuracy: 0.734194
saving to ..., time used 21.033s
Seen 210000 samples, time used: 147.330s
current training loss: 0.676666
current training accuracy: 0.734357
saving to ..., time used 20.938s
Seen 240000 samples, time used: 168.143s
current training loss: 0.675051
current training accuracy: 0.735138
saving to ..., time used 20.674s
Seen 270000 samples, time used: 189.021s
current training loss: 0.673810
current training accuracy: 0.736000
saving to ..., time used 20.741s
Seen 300000 samples, time used: 209.876s
current training loss: 0.672225
current training accuracy: 0.737017
saving to ..., time used 20.720s
Seen 330000 samples, time used: 230.772s
current training loss: 0.670959
current training accuracy: 0.737448
saving to ..., time used 20.618s
Seen 360000 samples, time used: 251.516s
current training loss: 0.669598
current training accuracy: 0.737864
saving to ..., time used 20.611s
Seen 390000 samples, time used: 272.325s
current training loss: 0.667532
current training accuracy: 0.738836
saving to ..., time used 20.677s
Seen 420000 samples, time used: 293.445s
current training loss: 0.666920
current training accuracy: 0.739395
saving to ..., time used 20.867s
Seen 450000 samples, time used: 314.322s
current training loss: 0.665558
current training accuracy: 0.740096
saving to ..., time used 20.743s
Seen 480000 samples, time used: 335.316s
current training loss: 0.664452
current training accuracy: 0.740692
saving to ..., time used 20.857s
Seen 510000 samples, time used: 356.119s
current training loss: 0.663350
current training accuracy: 0.741204
saving to ..., time used 20.663s
Seen 540000 samples, time used: 376.852s
current training loss: 0.662442
current training accuracy: 0.741854
saving to ..., time used 20.598s
Epoch 2 of 20 took 384.859s
training loss: 0.662066
training accuracy: 74.20 %
validation loss: 0.594461
validation accuracy: 75.51 %
test loss: 0.598361
test accuracy: 75.24 %
saving to ./snli/condition_model_epoch2.npz
Seen 30000 samples, time used: 20.563s
current training loss: 0.618628
current training accuracy: 0.765567
saving to ..., time used 20.563s
Seen 60000 samples, time used: 41.068s
current training loss: 0.619230
current training accuracy: 0.765983
saving to ..., time used 20.243s
Seen 90000 samples, time used: 61.708s
current training loss: 0.621668
current training accuracy: 0.763556
saving to ..., time used 20.348s
Seen 120000 samples, time used: 82.477s
current training loss: 0.622325
current training accuracy: 0.763458
saving to ..., time used 20.634s
Seen 150000 samples, time used: 103.418s
current training loss: 0.622511
current training accuracy: 0.763113
saving to ..., time used 20.663s
Seen 180000 samples, time used: 124.185s
current training loss: 0.620764
current training accuracy: 0.764000
saving to ..., time used 20.634s
Seen 210000 samples, time used: 145.066s
current training loss: 0.620599
current training accuracy: 0.764148
saving to ..., time used 20.730s
Seen 240000 samples, time used: 165.277s
current training loss: 0.620940
current training accuracy: 0.764150
saving to ..., time used 20.079s
Seen 270000 samples, time used: 185.584s
current training loss: 0.620593
current training accuracy: 0.764222
saving to ..., time used 20.151s
Seen 300000 samples, time used: 206.071s
current training loss: 0.619980
current training accuracy: 0.764500
saving to ..., time used 20.358s
Seen 330000 samples, time used: 226.235s
current training loss: 0.619443
current training accuracy: 0.764879
saving to ..., time used 20.032s
Seen 360000 samples, time used: 246.603s
current training loss: 0.618373
current training accuracy: 0.765094
saving to ..., time used 20.237s
Seen 390000 samples, time used: 267.534s
current training loss: 0.618145
current training accuracy: 0.765146
saving to ..., time used 20.800s
Seen 420000 samples, time used: 288.379s
current training loss: 0.617679
current training accuracy: 0.765450
saving to ..., time used 20.709s
Seen 450000 samples, time used: 309.682s
current training loss: 0.617312
current training accuracy: 0.765578
saving to ..., time used 21.174s
Seen 480000 samples, time used: 330.657s
current training loss: 0.616658
current training accuracy: 0.765875
saving to ..., time used 20.820s
Seen 510000 samples, time used: 351.563s
current training loss: 0.616543
current training accuracy: 0.766051
saving to ..., time used 20.777s
Seen 540000 samples, time used: 372.131s
current training loss: 0.616215
current training accuracy: 0.766348
saving to ..., time used 20.433s
Epoch 3 of 20 took 380.056s
training loss: 0.616020
training accuracy: 76.63 %
validation loss: 0.575587
validation accuracy: 76.73 %
test loss: 0.575943
test accuracy: 76.58 %
saving to ./snli/condition_model_epoch3.npz
Seen 30000 samples, time used: 20.794s
current training loss: 0.585951
current training accuracy: 0.779567
saving to ..., time used 20.794s
Seen 60000 samples, time used: 41.774s
current training loss: 0.588213
current training accuracy: 0.779533
saving to ..., time used 20.846s
Seen 90000 samples, time used: 62.819s
current training loss: 0.587461
current training accuracy: 0.780133
saving to ..., time used 20.790s
Seen 120000 samples, time used: 83.291s
current training loss: 0.589118
current training accuracy: 0.779650
saving to ..., time used 20.342s
Seen 150000 samples, time used: 104.634s
current training loss: 0.590210
current training accuracy: 0.779393
saving to ..., time used 21.210s
Seen 180000 samples, time used: 125.763s
current training loss: 0.591271
current training accuracy: 0.779200
saving to ..., time used 20.987s
Seen 210000 samples, time used: 147.189s
current training loss: 0.590988
current training accuracy: 0.779214
saving to ..., time used 21.296s
Seen 240000 samples, time used: 167.692s
current training loss: 0.591482
current training accuracy: 0.779112
saving to ..., time used 20.369s
Seen 270000 samples, time used: 188.217s
current training loss: 0.591617
current training accuracy: 0.779000
saving to ..., time used 20.383s
Seen 300000 samples, time used: 208.741s
current training loss: 0.591320
current training accuracy: 0.779197
saving to ..., time used 20.393s
Seen 330000 samples, time used: 228.981s
current training loss: 0.590882
current training accuracy: 0.779436
saving to ..., time used 20.110s
Seen 360000 samples, time used: 250.588s
current training loss: 0.590751
current training accuracy: 0.779486
saving to ..., time used 21.475s
Seen 390000 samples, time used: 273.482s
current training loss: 0.590438
current training accuracy: 0.779756
saving to ..., time used 22.709s
Seen 420000 samples, time used: 296.118s
current training loss: 0.590388
current training accuracy: 0.779698
saving to ..., time used 22.488s
Seen 450000 samples, time used: 318.948s
current training loss: 0.590438
current training accuracy: 0.779633
saving to ..., time used 22.681s
Seen 480000 samples, time used: 341.866s
current training loss: 0.590528
current training accuracy: 0.779485
saving to ..., time used 22.770s
Seen 510000 samples, time used: 364.728s
current training loss: 0.590519
current training accuracy: 0.779453
saving to ..., time used 22.714s
Seen 540000 samples, time used: 387.624s
current training loss: 0.589969
current training accuracy: 0.779739
saving to ..., time used 22.749s
Epoch 4 of 20 took 396.417s
training loss: 0.589797
training accuracy: 77.98 %
validation loss: 0.549442
validation accuracy: 77.50 %
test loss: 0.558039
test accuracy: 77.91 %
saving to ./snli/condition_model_epoch4.npz
Seen 30000 samples, time used: 19.999s
current training loss: 0.559500
current training accuracy: 0.797000
saving to ..., time used 19.999s
Seen 60000 samples, time used: 39.971s
current training loss: 0.560991
current training accuracy: 0.794800
saving to ..., time used 19.842s
Seen 90000 samples, time used: 60.025s
current training loss: 0.560726
current training accuracy: 0.795222
saving to ..., time used 19.890s
Seen 120000 samples, time used: 80.092s
current training loss: 0.562451
current training accuracy: 0.794050
saving to ..., time used 19.936s
Seen 150000 samples, time used: 100.213s
current training loss: 0.564482
current training accuracy: 0.793273
saving to ..., time used 19.841s
Seen 180000 samples, time used: 120.333s
current training loss: 0.564840
current training accuracy: 0.792783
saving to ..., time used 19.990s
Seen 210000 samples, time used: 140.769s
current training loss: 0.566457
current training accuracy: 0.792133
saving to ..., time used 20.153s
Seen 240000 samples, time used: 160.820s
current training loss: 0.566388
current training accuracy: 0.792167
saving to ..., time used 19.920s
Seen 270000 samples, time used: 180.940s
current training loss: 0.567379
current training accuracy: 0.791633
saving to ..., time used 19.990s
Seen 300000 samples, time used: 201.048s
current training loss: 0.567809
current training accuracy: 0.791243
saving to ..., time used 19.978s
Seen 330000 samples, time used: 221.076s
current training loss: 0.568476
current training accuracy: 0.791024
saving to ..., time used 19.898s
Seen 360000 samples, time used: 241.088s
current training loss: 0.569020
current training accuracy: 0.790767
saving to ..., time used 19.883s
Seen 390000 samples, time used: 261.178s
current training loss: 0.569413
current training accuracy: 0.790321
saving to ..., time used 19.958s
Seen 420000 samples, time used: 281.245s
current training loss: 0.569762
current training accuracy: 0.790195
saving to ..., time used 19.938s
Seen 450000 samples, time used: 301.231s
current training loss: 0.570386
current training accuracy: 0.789949
saving to ..., time used 19.855s
Seen 480000 samples, time used: 321.281s
current training loss: 0.570909
current training accuracy: 0.789733
saving to ..., time used 19.919s
Seen 510000 samples, time used: 341.360s
current training loss: 0.570849
current training accuracy: 0.789684
saving to ..., time used 19.949s
Seen 540000 samples, time used: 361.323s
current training loss: 0.570982
current training accuracy: 0.789613
saving to ..., time used 19.833s
Epoch 5 of 20 took 369.044s
training loss: 0.571151
training accuracy: 78.95 %
validation loss: 0.539496
validation accuracy: 78.31 %
test loss: 0.555925
test accuracy: 77.69 %
saving to ./snli/condition_model_epoch5.npz
Seen 30000 samples, time used: 19.982s
current training loss: 0.558675
current training accuracy: 0.793867
saving to ..., time used 19.982s
Seen 60000 samples, time used: 40.164s
current training loss: 0.556288
current training accuracy: 0.797217
saving to ..., time used 20.012s
Seen 90000 samples, time used: 60.250s
current training loss: 0.554646
current training accuracy: 0.798311
saving to ..., time used 19.956s
Seen 120000 samples, time used: 80.420s
current training loss: 0.555574
current training accuracy: 0.798150
saving to ..., time used 19.882s
Seen 150000 samples, time used: 100.494s
current training loss: 0.556082
current training accuracy: 0.798480
saving to ..., time used 19.944s
Seen 180000 samples, time used: 120.609s
current training loss: 0.556915
current training accuracy: 0.798178
saving to ..., time used 19.986s
Seen 210000 samples, time used: 140.715s
current training loss: 0.557341
current training accuracy: 0.797348
saving to ..., time used 19.976s
Seen 240000 samples, time used: 160.798s
current training loss: 0.557547
current training accuracy: 0.797033
saving to ..., time used 19.953s
Seen 270000 samples, time used: 180.860s
current training loss: 0.557363
current training accuracy: 0.796974
saving to ..., time used 19.931s
Seen 300000 samples, time used: 200.900s
current training loss: 0.557486
current training accuracy: 0.796937
saving to ..., time used 19.910s
Seen 330000 samples, time used: 220.946s
current training loss: 0.557457
current training accuracy: 0.796624
saving to ..., time used 19.914s
Seen 360000 samples, time used: 241.127s
current training loss: 0.557954
current training accuracy: 0.796414
saving to ..., time used 20.051s
Seen 390000 samples, time used: 261.172s
current training loss: 0.558059
current training accuracy: 0.796374
saving to ..., time used 19.915s
Seen 420000 samples, time used: 281.146s
current training loss: 0.558290
current training accuracy: 0.796321
saving to ..., time used 19.843s
Seen 450000 samples, time used: 301.248s
current training loss: 0.558800
current training accuracy: 0.796040
saving to ..., time used 19.972s
Seen 480000 samples, time used: 321.336s
current training loss: 0.559185
current training accuracy: 0.795917
saving to ..., time used 19.957s
Seen 510000 samples, time used: 341.251s
current training loss: 0.559228
current training accuracy: 0.795827
saving to ..., time used 19.785s
Seen 540000 samples, time used: 361.341s
current training loss: 0.559273
current training accuracy: 0.795724
saving to ..., time used 19.960s
Epoch 6 of 20 took 368.996s
training loss: 0.559321
training accuracy: 79.57 %
validation loss: 0.530641
validation accuracy: 78.60 %
test loss: 0.547620
test accuracy: 77.85 %
saving to ./snli/condition_model_epoch6.npz
Seen 30000 samples, time used: 19.866s
current training loss: 0.542376
current training accuracy: 0.805833
saving to ..., time used 19.866s
Seen 60000 samples, time used: 39.931s
current training loss: 0.542239
current training accuracy: 0.804300
saving to ..., time used 19.935s
Seen 90000 samples, time used: 60.231s
current training loss: 0.545090
current training accuracy: 0.803522
saving to ..., time used 20.038s
Seen 120000 samples, time used: 80.306s
current training loss: 0.543134
current training accuracy: 0.802992
saving to ..., time used 19.945s
Seen 150000 samples, time used: 100.576s
current training loss: 0.543938
current training accuracy: 0.803093
saving to ..., time used 19.991s
Seen 180000 samples, time used: 120.569s
current training loss: 0.544114
current training accuracy: 0.803061
saving to ..., time used 19.862s
Seen 210000 samples, time used: 140.624s
current training loss: 0.544456
current training accuracy: 0.802790
saving to ..., time used 19.924s
Seen 240000 samples, time used: 160.600s
current training loss: 0.544219
current training accuracy: 0.802958
saving to ..., time used 19.846s
Seen 270000 samples, time used: 180.661s
current training loss: 0.544502
current training accuracy: 0.802885
saving to ..., time used 19.931s
Seen 300000 samples, time used: 200.656s
current training loss: 0.544761
current training accuracy: 0.802833
saving to ..., time used 19.865s
Seen 330000 samples, time used: 220.816s
current training loss: 0.545016
current training accuracy: 0.802852
saving to ..., time used 20.030s
Seen 360000 samples, time used: 240.877s
current training loss: 0.545514
current training accuracy: 0.802472
saving to ..., time used 19.931s
Seen 390000 samples, time used: 261.021s
current training loss: 0.546039
current training accuracy: 0.802269
saving to ..., time used 20.013s
Seen 420000 samples, time used: 281.155s
current training loss: 0.546203
current training accuracy: 0.802167
saving to ..., time used 20.004s
Seen 450000 samples, time used: 301.252s
current training loss: 0.546451
current training accuracy: 0.801924
saving to ..., time used 19.967s
Seen 480000 samples, time used: 321.306s
current training loss: 0.547163
current training accuracy: 0.801552
saving to ..., time used 19.924s
Seen 510000 samples, time used: 341.433s
current training loss: 0.547584
current training accuracy: 0.801349
saving to ..., time used 19.998s
Seen 540000 samples, time used: 361.523s
current training loss: 0.547888
current training accuracy: 0.801128
saving to ..., time used 19.960s
Epoch 7 of 20 took 369.233s
training loss: 0.548083
training accuracy: 80.10 %
validation loss: 0.529902
validation accuracy: 78.73 %
test loss: 0.543665
test accuracy: 78.62 %
saving to ./snli/condition_model_epoch7.npz
Seen 30000 samples, time used: 19.999s
current training loss: 0.531973
current training accuracy: 0.812500
saving to ..., time used 19.999s
Seen 60000 samples, time used: 40.338s
current training loss: 0.531957
current training accuracy: 0.811650
saving to ..., time used 20.203s
Seen 90000 samples, time used: 60.747s
current training loss: 0.533370
current training accuracy: 0.811178
saving to ..., time used 20.280s
Seen 120000 samples, time used: 80.997s
current training loss: 0.534487
current training accuracy: 0.810383
saving to ..., time used 20.049s
Seen 150000 samples, time used: 101.219s
current training loss: 0.534645
current training accuracy: 0.809640
saving to ..., time used 20.091s
Seen 180000 samples, time used: 121.302s
current training loss: 0.536042
current training accuracy: 0.808778
saving to ..., time used 19.952s
Seen 210000 samples, time used: 141.284s
current training loss: 0.536904
current training accuracy: 0.808138
saving to ..., time used 19.852s
Seen 240000 samples, time used: 161.428s
current training loss: 0.535950
current training accuracy: 0.808267
saving to ..., time used 20.013s
Seen 270000 samples, time used: 181.532s
current training loss: 0.536776
current training accuracy: 0.807993
saving to ..., time used 19.974s
Seen 300000 samples, time used: 201.536s
current training loss: 0.536622
current training accuracy: 0.807890
saving to ..., time used 19.873s
Seen 330000 samples, time used: 221.637s
current training loss: 0.537182
current training accuracy: 0.807627
saving to ..., time used 19.970s
Seen 360000 samples, time used: 241.691s
current training loss: 0.537643
current training accuracy: 0.807564
saving to ..., time used 19.924s
Seen 390000 samples, time used: 261.741s
current training loss: 0.538300
current training accuracy: 0.807121
saving to ..., time used 19.919s
Seen 420000 samples, time used: 281.926s
current training loss: 0.538878
current training accuracy: 0.806748
saving to ..., time used 20.055s
Seen 450000 samples, time used: 302.188s
current training loss: 0.539599
current training accuracy: 0.806560
saving to ..., time used 20.131s
Seen 480000 samples, time used: 322.243s
current training loss: 0.539538
current training accuracy: 0.806554
saving to ..., time used 19.924s
Seen 510000 samples, time used: 342.280s
current training loss: 0.539548
current training accuracy: 0.806531
saving to ..., time used 19.906s
Seen 540000 samples, time used: 362.383s
current training loss: 0.540330
current training accuracy: 0.806093
saving to ..., time used 19.973s
Epoch 8 of 20 took 370.163s
training loss: 0.540504
training accuracy: 80.59 %
validation loss: 0.525441
validation accuracy: 79.31 %
test loss: 0.535627
test accuracy: 78.85 %
saving to ./snli/condition_model_epoch8.npz
Seen 30000 samples, time used: 19.947s
current training loss: 0.522114
current training accuracy: 0.815767
saving to ..., time used 19.947s
Seen 60000 samples, time used: 39.962s
current training loss: 0.522633
current training accuracy: 0.816167
saving to ..., time used 19.885s
Seen 90000 samples, time used: 59.998s
current training loss: 0.525041
current training accuracy: 0.815211
saving to ..., time used 19.898s
Seen 120000 samples, time used: 80.145s
current training loss: 0.526405
current training accuracy: 0.814592
saving to ..., time used 20.017s
Seen 150000 samples, time used: 100.454s
current training loss: 0.527311
current training accuracy: 0.813660
saving to ..., time used 20.029s
Seen 180000 samples, time used: 120.625s
current training loss: 0.527273
current training accuracy: 0.813539
saving to ..., time used 20.040s
Seen 210000 samples, time used: 140.697s
current training loss: 0.527880
current training accuracy: 0.813119
saving to ..., time used 19.942s
Seen 240000 samples, time used: 160.775s
current training loss: 0.528952
current training accuracy: 0.812579
saving to ..., time used 19.948s
Seen 270000 samples, time used: 180.844s
current training loss: 0.529369
current training accuracy: 0.812433
saving to ..., time used 19.938s
Seen 300000 samples, time used: 200.911s
current training loss: 0.530531
current training accuracy: 0.811797
saving to ..., time used 19.937s
Seen 330000 samples, time used: 220.996s
current training loss: 0.531205
current training accuracy: 0.811424
saving to ..., time used 19.955s
Seen 360000 samples, time used: 241.001s
current training loss: 0.532277
current training accuracy: 0.810950
saving to ..., time used 19.874s
Seen 390000 samples, time used: 261.086s
current training loss: 0.532838
current training accuracy: 0.810656
saving to ..., time used 19.955s
Seen 420000 samples, time used: 281.190s
current training loss: 0.532797
current training accuracy: 0.810419
saving to ..., time used 19.974s
Seen 450000 samples, time used: 301.211s
current training loss: 0.532686
current training accuracy: 0.810376
saving to ..., time used 19.891s
Seen 480000 samples, time used: 321.348s
current training loss: 0.533125
current training accuracy: 0.809935
saving to ..., time used 20.007s
Seen 510000 samples, time used: 341.446s
current training loss: 0.533349
current training accuracy: 0.809616
saving to ..., time used 19.968s
Seen 540000 samples, time used: 361.607s
current training loss: 0.533557
current training accuracy: 0.809311
saving to ..., time used 20.030s
Epoch 9 of 20 took 369.263s
training loss: 0.533874
training accuracy: 80.91 %
validation loss: 0.519919
validation accuracy: 79.21 %
test loss: 0.527634
test accuracy: 79.39 %
saving to ./snli/condition_model_epoch9.npz
Seen 30000 samples, time used: 19.952s
current training loss: 0.517376
current training accuracy: 0.820333
saving to ..., time used 19.952s
Seen 60000 samples, time used: 40.148s
current training loss: 0.518832
current training accuracy: 0.817583
saving to ..., time used 20.065s
Seen 90000 samples, time used: 60.328s
current training loss: 0.521582
current training accuracy: 0.816622
saving to ..., time used 19.970s
Seen 120000 samples, time used: 80.443s
current training loss: 0.522696
current training accuracy: 0.815600
saving to ..., time used 19.984s
Seen 150000 samples, time used: 100.677s
current training loss: 0.524173
current training accuracy: 0.815193
saving to ..., time used 20.095s
Seen 180000 samples, time used: 120.764s
current training loss: 0.523170
current training accuracy: 0.815422
saving to ..., time used 19.956s
Seen 210000 samples, time used: 140.819s
current training loss: 0.524001
current training accuracy: 0.814890
saving to ..., time used 19.925s
Seen 240000 samples, time used: 160.933s
current training loss: 0.524759
current training accuracy: 0.814829
saving to ..., time used 19.984s
Seen 270000 samples, time used: 181.004s
current training loss: 0.525777
current training accuracy: 0.814319
saving to ..., time used 19.940s
Seen 300000 samples, time used: 200.985s
current training loss: 0.526574
current training accuracy: 0.814047
saving to ..., time used 19.850s
Seen 330000 samples, time used: 221.139s
current training loss: 0.527200
current training accuracy: 0.813727
saving to ..., time used 20.023s
Seen 360000 samples, time used: 241.225s
current training loss: 0.528082
current training accuracy: 0.813089
saving to ..., time used 19.956s
Seen 390000 samples, time used: 261.296s
current training loss: 0.528221
current training accuracy: 0.812818
saving to ..., time used 19.942s
Seen 420000 samples, time used: 281.264s
current training loss: 0.528214
current training accuracy: 0.812648
saving to ..., time used 19.838s
Seen 450000 samples, time used: 301.315s
current training loss: 0.528070
current training accuracy: 0.812787
saving to ..., time used 19.921s
Seen 480000 samples, time used: 321.328s
current training loss: 0.528378
current training accuracy: 0.812560
saving to ..., time used 19.884s
Seen 510000 samples, time used: 341.277s
current training loss: 0.528757
current training accuracy: 0.812371
saving to ..., time used 19.818s
Seen 540000 samples, time used: 361.410s
current training loss: 0.529339
current training accuracy: 0.812159
saving to ..., time used 20.003s
Epoch 10 of 20 took 369.113s
training loss: 0.529326
training accuracy: 81.22 %
validation loss: 0.522066
validation accuracy: 79.23 %
test loss: 0.530091
test accuracy: 79.71 %
saving to ./snli/condition_model_epoch10.npz
Seen 30000 samples, time used: 19.940s
current training loss: 0.515903
current training accuracy: 0.818667
saving to ..., time used 19.940s
Seen 60000 samples, time used: 39.920s
current training loss: 0.515054
current training accuracy: 0.820067
saving to ..., time used 19.849s
Seen 90000 samples, time used: 60.104s
current training loss: 0.515618
current training accuracy: 0.820322
saving to ..., time used 20.015s
Seen 120000 samples, time used: 80.397s
current training loss: 0.514026
current training accuracy: 0.820617
saving to ..., time used 20.051s
Seen 150000 samples, time used: 100.670s
current training loss: 0.515065
current training accuracy: 0.820073
saving to ..., time used 20.018s
Seen 180000 samples, time used: 120.868s
current training loss: 0.515825
current training accuracy: 0.820006
saving to ..., time used 19.956s
Seen 210000 samples, time used: 140.950s
current training loss: 0.516853
current training accuracy: 0.819338
saving to ..., time used 19.882s
Seen 240000 samples, time used: 161.131s
current training loss: 0.517910
current training accuracy: 0.818925
saving to ..., time used 20.052s
Seen 270000 samples, time used: 181.383s
current training loss: 0.518621
current training accuracy: 0.818500
saving to ..., time used 20.060s
Seen 300000 samples, time used: 201.678s
current training loss: 0.519698
current training accuracy: 0.817827
saving to ..., time used 20.129s
Seen 330000 samples, time used: 221.835s
current training loss: 0.520858
current training accuracy: 0.817100
saving to ..., time used 19.986s
Seen 360000 samples, time used: 241.869s
current training loss: 0.522278
current training accuracy: 0.816461
saving to ..., time used 19.904s
Seen 390000 samples, time used: 261.926s
current training loss: 0.522455
current training accuracy: 0.816374
saving to ..., time used 19.774s
Seen 420000 samples, time used: 281.969s
current training loss: 0.522494
current training accuracy: 0.816417
saving to ..., time used 19.914s
Seen 450000 samples, time used: 302.091s
current training loss: 0.523220
current training accuracy: 0.815993
saving to ..., time used 19.993s
Seen 480000 samples, time used: 322.282s
current training loss: 0.523723
current training accuracy: 0.815719
saving to ..., time used 20.061s
Seen 510000 samples, time used: 342.755s
current training loss: 0.524478
current training accuracy: 0.815435
saving to ..., time used 20.343s
Seen 540000 samples, time used: 363.425s
current training loss: 0.524957
current training accuracy: 0.815069
saving to ..., time used 20.540s
Epoch 11 of 20 took 371.211s
training loss: 0.525169
training accuracy: 81.49 %
validation loss: 0.509843
validation accuracy: 80.00 %
test loss: 0.525920
test accuracy: 79.34 %
saving to ./snli/condition_model_epoch11.npz
Seen 30000 samples, time used: 21.164s
current training loss: 0.506509
current training accuracy: 0.825433
saving to ..., time used 21.164s
Seen 60000 samples, time used: 42.167s
current training loss: 0.512147
current training accuracy: 0.821967
saving to ..., time used 20.747s
Seen 90000 samples, time used: 62.990s
current training loss: 0.513311
current training accuracy: 0.821233
saving to ..., time used 20.691s
Seen 120000 samples, time used: 83.111s
current training loss: 0.512911
current training accuracy: 0.820750
saving to ..., time used 19.938s
Seen 150000 samples, time used: 103.175s
current training loss: 0.512475
current training accuracy: 0.821067
saving to ..., time used 19.933s
Seen 180000 samples, time used: 123.278s
current training loss: 0.513972
current training accuracy: 0.820472
saving to ..., time used 19.973s
Seen 210000 samples, time used: 143.370s
current training loss: 0.515197
current training accuracy: 0.819690
saving to ..., time used 19.962s
Seen 240000 samples, time used: 163.465s
current training loss: 0.515231
current training accuracy: 0.819737
saving to ..., time used 19.965s
Seen 270000 samples, time used: 183.536s
current training loss: 0.515987
current training accuracy: 0.819293
saving to ..., time used 19.941s
Seen 300000 samples, time used: 203.681s
current training loss: 0.516565
current training accuracy: 0.818783
saving to ..., time used 20.017s
Seen 330000 samples, time used: 223.836s
current training loss: 0.517473
current training accuracy: 0.818324
saving to ..., time used 20.026s
Seen 360000 samples, time used: 243.959s
current training loss: 0.518122
current training accuracy: 0.817814
saving to ..., time used 19.993s
Seen 390000 samples, time used: 264.002s
current training loss: 0.518677
current training accuracy: 0.817492
saving to ..., time used 19.914s
Seen 420000 samples, time used: 284.063s
current training loss: 0.519158
current training accuracy: 0.816995
saving to ..., time used 19.930s
Seen 450000 samples, time used: 304.177s
current training loss: 0.519482
current training accuracy: 0.816838
saving to ..., time used 19.984s
Seen 480000 samples, time used: 324.376s
current training loss: 0.519519
current training accuracy: 0.816777
saving to ..., time used 20.070s
Seen 510000 samples, time used: 344.484s
current training loss: 0.519548
current training accuracy: 0.816671
saving to ..., time used 19.978s
Seen 540000 samples, time used: 364.617s
current training loss: 0.519943
current training accuracy: 0.816467
saving to ..., time used 20.003s
Epoch 12 of 20 took 372.297s
training loss: 0.520113
training accuracy: 81.63 %
validation loss: 0.515100
validation accuracy: 79.79 %
test loss: 0.522821
test accuracy: 79.94 %