pos-french / training.log
qanastek's picture
Push model
b897cce
2021-03-02 12:06:03,526 ----------------------------------------------------------------------------------------------------
2021-03-02 12:06:03,526 Model: "SequenceTagger(
(embeddings): StackedEmbeddings(
(list_embedding_0): WordEmbeddings('fr')
)
(word_dropout): WordDropout(p=0.05)
(locked_dropout): LockedDropout(p=0.5)
(embedding2nn): Linear(in_features=300, out_features=300, bias=True)
(rnn): LSTM(300, 256, batch_first=True, bidirectional=True)
(linear): Linear(in_features=512, out_features=21, bias=True)
(beta): 1.0
(weights): None
(weight_tensor) None
)"
2021-03-02 12:06:03,526 ----------------------------------------------------------------------------------------------------
2021-03-02 12:06:03,526 Corpus: "Corpus: 14449 train + 1476 dev + 416 test sentences"
2021-03-02 12:06:03,526 ----------------------------------------------------------------------------------------------------
2021-03-02 12:06:03,526 Parameters:
2021-03-02 12:06:03,526 - learning_rate: "0.1"
2021-03-02 12:06:03,526 - mini_batch_size: "64"
2021-03-02 12:06:03,526 - patience: "3"
2021-03-02 12:06:03,526 - anneal_factor: "0.5"
2021-03-02 12:06:03,526 - max_epochs: "150"
2021-03-02 12:06:03,526 - shuffle: "True"
2021-03-02 12:06:03,526 - train_with_dev: "False"
2021-03-02 12:06:03,526 - batch_growth_annealing: "False"
2021-03-02 12:06:03,526 ----------------------------------------------------------------------------------------------------
2021-03-02 12:06:03,526 Model training base path: "resources/taggers/example-pos"
2021-03-02 12:06:03,526 ----------------------------------------------------------------------------------------------------
2021-03-02 12:06:03,526 Device: cuda:0
2021-03-02 12:06:03,526 ----------------------------------------------------------------------------------------------------
2021-03-02 12:06:03,527 Embeddings storage mode: cpu
2021-03-02 12:06:03,529 ----------------------------------------------------------------------------------------------------
2021-03-02 12:06:10,054 epoch 1 - iter 22/226 - loss 65.55180203 - samples/sec: 215.87 - lr: 0.100000
2021-03-02 12:06:16,066 epoch 1 - iter 44/226 - loss 54.34445927 - samples/sec: 234.31 - lr: 0.100000
2021-03-02 12:06:22,324 epoch 1 - iter 66/226 - loss 46.26198378 - samples/sec: 225.12 - lr: 0.100000
2021-03-02 12:06:28,449 epoch 1 - iter 88/226 - loss 40.31244484 - samples/sec: 230.03 - lr: 0.100000
2021-03-02 12:06:34,624 epoch 1 - iter 110/226 - loss 36.06240538 - samples/sec: 228.12 - lr: 0.100000
2021-03-02 12:06:40,651 epoch 1 - iter 132/226 - loss 32.68936389 - samples/sec: 233.73 - lr: 0.100000
2021-03-02 12:06:46,708 epoch 1 - iter 154/226 - loss 30.06182317 - samples/sec: 232.57 - lr: 0.100000
2021-03-02 12:06:55,670 epoch 1 - iter 176/226 - loss 27.93609124 - samples/sec: 157.17 - lr: 0.100000
2021-03-02 12:07:01,627 epoch 1 - iter 198/226 - loss 26.16095592 - samples/sec: 236.41 - lr: 0.100000
2021-03-02 12:07:07,719 epoch 1 - iter 220/226 - loss 24.70696594 - samples/sec: 231.27 - lr: 0.100000
2021-03-02 12:07:09,303 ----------------------------------------------------------------------------------------------------
2021-03-02 12:07:09,303 EPOCH 1 done: loss 24.3529 - lr 0.1000000
2021-03-02 12:07:13,202 DEV : loss 6.874114513397217 - score 0.9124
2021-03-02 12:07:13,278 BAD EPOCHS (no improvement): 0
2021-03-02 12:07:33,520 ----------------------------------------------------------------------------------------------------
2021-03-02 12:07:38,296 epoch 2 - iter 22/226 - loss 11.04434772 - samples/sec: 295.31 - lr: 0.100000
2021-03-02 12:07:43,086 epoch 2 - iter 44/226 - loss 10.87801777 - samples/sec: 294.23 - lr: 0.100000
2021-03-02 12:07:47,960 epoch 2 - iter 66/226 - loss 10.75282822 - samples/sec: 289.14 - lr: 0.100000
2021-03-02 12:07:53,308 epoch 2 - iter 88/226 - loss 10.62418799 - samples/sec: 263.48 - lr: 0.100000
2021-03-02 12:07:57,871 epoch 2 - iter 110/226 - loss 10.39342443 - samples/sec: 308.84 - lr: 0.100000
2021-03-02 12:08:02,453 epoch 2 - iter 132/226 - loss 10.22209210 - samples/sec: 307.59 - lr: 0.100000
2021-03-02 12:08:07,051 epoch 2 - iter 154/226 - loss 10.07944821 - samples/sec: 306.47 - lr: 0.100000
2021-03-02 12:08:12,095 epoch 2 - iter 176/226 - loss 10.03107517 - samples/sec: 279.34 - lr: 0.100000
2021-03-02 12:08:16,838 epoch 2 - iter 198/226 - loss 9.92488862 - samples/sec: 297.16 - lr: 0.100000
2021-03-02 12:08:21,487 epoch 2 - iter 220/226 - loss 9.83373491 - samples/sec: 303.14 - lr: 0.100000
2021-03-02 12:08:22,731 ----------------------------------------------------------------------------------------------------
2021-03-02 12:08:22,732 EPOCH 2 done: loss 9.8215 - lr 0.1000000
2021-03-02 12:08:25,275 DEV : loss 5.277429103851318 - score 0.9289
2021-03-02 12:08:25,351 BAD EPOCHS (no improvement): 0
2021-03-02 12:08:45,552 ----------------------------------------------------------------------------------------------------
2021-03-02 12:08:50,402 epoch 3 - iter 22/226 - loss 8.72399512 - samples/sec: 290.65 - lr: 0.100000
2021-03-02 12:08:54,855 epoch 3 - iter 44/226 - loss 8.58594025 - samples/sec: 316.49 - lr: 0.100000
2021-03-02 12:09:00,055 epoch 3 - iter 66/226 - loss 8.70956379 - samples/sec: 270.94 - lr: 0.100000
2021-03-02 12:09:04,800 epoch 3 - iter 88/226 - loss 8.67046070 - samples/sec: 297.02 - lr: 0.100000
2021-03-02 12:09:09,433 epoch 3 - iter 110/226 - loss 8.59309378 - samples/sec: 304.17 - lr: 0.100000
2021-03-02 12:09:14,270 epoch 3 - iter 132/226 - loss 8.58896746 - samples/sec: 291.33 - lr: 0.100000
2021-03-02 12:09:19,425 epoch 3 - iter 154/226 - loss 8.60759325 - samples/sec: 273.34 - lr: 0.100000
2021-03-02 12:09:23,906 epoch 3 - iter 176/226 - loss 8.57032593 - samples/sec: 314.57 - lr: 0.100000
2021-03-02 12:09:28,543 epoch 3 - iter 198/226 - loss 8.56564222 - samples/sec: 303.89 - lr: 0.100000
2021-03-02 12:09:33,105 epoch 3 - iter 220/226 - loss 8.52381685 - samples/sec: 308.89 - lr: 0.100000
2021-03-02 12:09:34,363 ----------------------------------------------------------------------------------------------------
2021-03-02 12:09:34,363 EPOCH 3 done: loss 8.5228 - lr 0.1000000
2021-03-02 12:09:36,915 DEV : loss 4.803399562835693 - score 0.935
2021-03-02 12:09:36,994 BAD EPOCHS (no improvement): 0
2021-03-02 12:09:57,188 ----------------------------------------------------------------------------------------------------
2021-03-02 12:10:02,369 epoch 4 - iter 22/226 - loss 8.43228316 - samples/sec: 272.16 - lr: 0.100000
2021-03-02 12:10:07,054 epoch 4 - iter 44/226 - loss 8.30056567 - samples/sec: 300.84 - lr: 0.100000
2021-03-02 12:10:11,947 epoch 4 - iter 66/226 - loss 8.28582814 - samples/sec: 287.97 - lr: 0.100000
2021-03-02 12:10:16,621 epoch 4 - iter 88/226 - loss 8.21634536 - samples/sec: 301.46 - lr: 0.100000
2021-03-02 12:10:21,242 epoch 4 - iter 110/226 - loss 8.13665546 - samples/sec: 304.97 - lr: 0.100000
2021-03-02 12:10:26,037 epoch 4 - iter 132/226 - loss 8.14766377 - samples/sec: 293.83 - lr: 0.100000
2021-03-02 12:10:30,512 epoch 4 - iter 154/226 - loss 8.06639609 - samples/sec: 314.92 - lr: 0.100000
2021-03-02 12:10:35,383 epoch 4 - iter 176/226 - loss 8.00978838 - samples/sec: 289.35 - lr: 0.100000
2021-03-02 12:10:39,951 epoch 4 - iter 198/226 - loss 7.97817996 - samples/sec: 308.49 - lr: 0.100000
2021-03-02 12:10:44,667 epoch 4 - iter 220/226 - loss 7.95339045 - samples/sec: 298.76 - lr: 0.100000
2021-03-02 12:10:45,867 ----------------------------------------------------------------------------------------------------
2021-03-02 12:10:45,868 EPOCH 4 done: loss 7.9403 - lr 0.1000000
2021-03-02 12:10:48,422 DEV : loss 4.450038433074951 - score 0.9399
2021-03-02 12:10:48,499 BAD EPOCHS (no improvement): 0
2021-03-02 12:11:08,718 ----------------------------------------------------------------------------------------------------
2021-03-02 12:11:13,497 epoch 5 - iter 22/226 - loss 7.69163119 - samples/sec: 295.16 - lr: 0.100000
2021-03-02 12:11:17,996 epoch 5 - iter 44/226 - loss 7.71907982 - samples/sec: 313.27 - lr: 0.100000
2021-03-02 12:11:23,169 epoch 5 - iter 66/226 - loss 7.74686594 - samples/sec: 272.35 - lr: 0.100000
2021-03-02 12:11:28,137 epoch 5 - iter 88/226 - loss 7.77408807 - samples/sec: 283.65 - lr: 0.100000
2021-03-02 12:11:32,826 epoch 5 - iter 110/226 - loss 7.71197649 - samples/sec: 300.51 - lr: 0.100000
2021-03-02 12:11:37,414 epoch 5 - iter 132/226 - loss 7.65543950 - samples/sec: 307.07 - lr: 0.100000
2021-03-02 12:11:41,946 epoch 5 - iter 154/226 - loss 7.64849317 - samples/sec: 310.89 - lr: 0.100000
2021-03-02 12:11:46,406 epoch 5 - iter 176/226 - loss 7.62513854 - samples/sec: 315.93 - lr: 0.100000
2021-03-02 12:11:51,184 epoch 5 - iter 198/226 - loss 7.62624801 - samples/sec: 294.88 - lr: 0.100000
2021-03-02 12:11:55,960 epoch 5 - iter 220/226 - loss 7.58450952 - samples/sec: 295.01 - lr: 0.100000
2021-03-02 12:11:57,117 ----------------------------------------------------------------------------------------------------
2021-03-02 12:11:57,118 EPOCH 5 done: loss 7.5771 - lr 0.1000000
2021-03-02 12:11:59,662 DEV : loss 4.1144700050354 - score 0.9453
2021-03-02 12:11:59,737 BAD EPOCHS (no improvement): 0
2021-03-02 12:12:19,984 ----------------------------------------------------------------------------------------------------
2021-03-02 12:12:24,848 epoch 6 - iter 22/226 - loss 7.36023253 - samples/sec: 289.85 - lr: 0.100000
2021-03-02 12:12:29,420 epoch 6 - iter 44/226 - loss 7.38352940 - samples/sec: 308.23 - lr: 0.100000
2021-03-02 12:12:33,903 epoch 6 - iter 66/226 - loss 7.23470168 - samples/sec: 314.29 - lr: 0.100000
2021-03-02 12:12:38,494 epoch 6 - iter 88/226 - loss 7.27050326 - samples/sec: 306.89 - lr: 0.100000
2021-03-02 12:12:43,440 epoch 6 - iter 110/226 - loss 7.27176122 - samples/sec: 284.85 - lr: 0.100000
2021-03-02 12:12:48,144 epoch 6 - iter 132/226 - loss 7.22077260 - samples/sec: 299.54 - lr: 0.100000
2021-03-02 12:12:52,884 epoch 6 - iter 154/226 - loss 7.23148984 - samples/sec: 297.26 - lr: 0.100000
2021-03-02 12:12:57,924 epoch 6 - iter 176/226 - loss 7.26751958 - samples/sec: 279.62 - lr: 0.100000
2021-03-02 12:13:02,489 epoch 6 - iter 198/226 - loss 7.26897373 - samples/sec: 308.68 - lr: 0.100000
2021-03-02 12:13:07,150 epoch 6 - iter 220/226 - loss 7.25435415 - samples/sec: 302.36 - lr: 0.100000
2021-03-02 12:13:08,602 ----------------------------------------------------------------------------------------------------
2021-03-02 12:13:08,603 EPOCH 6 done: loss 7.2444 - lr 0.1000000
2021-03-02 12:13:11,149 DEV : loss 3.8972177505493164 - score 0.9465
2021-03-02 12:13:11,224 BAD EPOCHS (no improvement): 0
2021-03-02 12:13:31,440 ----------------------------------------------------------------------------------------------------
2021-03-02 12:13:36,070 epoch 7 - iter 22/226 - loss 7.25298706 - samples/sec: 304.47 - lr: 0.100000
2021-03-02 12:13:40,794 epoch 7 - iter 44/226 - loss 7.00910956 - samples/sec: 298.25 - lr: 0.100000
2021-03-02 12:13:45,358 epoch 7 - iter 66/226 - loss 6.90938247 - samples/sec: 308.72 - lr: 0.100000
2021-03-02 12:13:50,144 epoch 7 - iter 88/226 - loss 6.84979005 - samples/sec: 294.32 - lr: 0.100000
2021-03-02 12:13:54,921 epoch 7 - iter 110/226 - loss 6.88427613 - samples/sec: 294.93 - lr: 0.100000
2021-03-02 12:13:59,520 epoch 7 - iter 132/226 - loss 6.87751160 - samples/sec: 306.35 - lr: 0.100000
2021-03-02 12:14:04,228 epoch 7 - iter 154/226 - loss 6.90348729 - samples/sec: 299.21 - lr: 0.100000
2021-03-02 12:14:08,915 epoch 7 - iter 176/226 - loss 6.89493325 - samples/sec: 300.68 - lr: 0.100000
2021-03-02 12:14:14,132 epoch 7 - iter 198/226 - loss 6.93368601 - samples/sec: 270.07 - lr: 0.100000
2021-03-02 12:14:18,812 epoch 7 - iter 220/226 - loss 6.91712191 - samples/sec: 301.10 - lr: 0.100000
2021-03-02 12:14:20,050 ----------------------------------------------------------------------------------------------------
2021-03-02 12:14:20,050 EPOCH 7 done: loss 6.9287 - lr 0.1000000
2021-03-02 12:14:22,600 DEV : loss 3.6763200759887695 - score 0.9505
2021-03-02 12:14:22,676 BAD EPOCHS (no improvement): 0
2021-03-02 12:14:42,882 ----------------------------------------------------------------------------------------------------
2021-03-02 12:14:47,734 epoch 8 - iter 22/226 - loss 6.82737892 - samples/sec: 290.56 - lr: 0.100000
2021-03-02 12:14:52,838 epoch 8 - iter 44/226 - loss 6.74179873 - samples/sec: 275.99 - lr: 0.100000
2021-03-02 12:14:57,488 epoch 8 - iter 66/226 - loss 6.67415548 - samples/sec: 303.00 - lr: 0.100000
2021-03-02 12:15:02,033 epoch 8 - iter 88/226 - loss 6.70279098 - samples/sec: 309.98 - lr: 0.100000
2021-03-02 12:15:06,932 epoch 8 - iter 110/226 - loss 6.70512675 - samples/sec: 287.59 - lr: 0.100000
2021-03-02 12:15:11,523 epoch 8 - iter 132/226 - loss 6.70713423 - samples/sec: 306.87 - lr: 0.100000
2021-03-02 12:15:16,046 epoch 8 - iter 154/226 - loss 6.71497879 - samples/sec: 311.52 - lr: 0.100000
2021-03-02 12:15:20,727 epoch 8 - iter 176/226 - loss 6.70076570 - samples/sec: 300.92 - lr: 0.100000
2021-03-02 12:15:25,578 epoch 8 - iter 198/226 - loss 6.72709603 - samples/sec: 290.40 - lr: 0.100000
2021-03-02 12:15:30,198 epoch 8 - iter 220/226 - loss 6.73148908 - samples/sec: 304.94 - lr: 0.100000
2021-03-02 12:15:31,424 ----------------------------------------------------------------------------------------------------
2021-03-02 12:15:31,424 EPOCH 8 done: loss 6.7306 - lr 0.1000000
2021-03-02 12:15:33,971 DEV : loss 3.591869354248047 - score 0.951
2021-03-02 12:15:34,047 BAD EPOCHS (no improvement): 0
2021-03-02 12:15:54,255 ----------------------------------------------------------------------------------------------------
2021-03-02 12:15:59,273 epoch 9 - iter 22/226 - loss 6.44558159 - samples/sec: 280.96 - lr: 0.100000
2021-03-02 12:16:04,053 epoch 9 - iter 44/226 - loss 6.67641181 - samples/sec: 294.80 - lr: 0.100000
2021-03-02 12:16:09,034 epoch 9 - iter 66/226 - loss 6.66629868 - samples/sec: 282.84 - lr: 0.100000
2021-03-02 12:16:13,725 epoch 9 - iter 88/226 - loss 6.59650782 - samples/sec: 300.35 - lr: 0.100000
2021-03-02 12:16:18,454 epoch 9 - iter 110/226 - loss 6.58020396 - samples/sec: 297.97 - lr: 0.100000
2021-03-02 12:16:23,194 epoch 9 - iter 132/226 - loss 6.56460140 - samples/sec: 297.24 - lr: 0.100000
2021-03-02 12:16:27,746 epoch 9 - iter 154/226 - loss 6.53671642 - samples/sec: 309.52 - lr: 0.100000
2021-03-02 12:16:32,389 epoch 9 - iter 176/226 - loss 6.54139594 - samples/sec: 303.46 - lr: 0.100000
2021-03-02 12:16:37,185 epoch 9 - iter 198/226 - loss 6.53887560 - samples/sec: 293.73 - lr: 0.100000
2021-03-02 12:16:41,801 epoch 9 - iter 220/226 - loss 6.51803454 - samples/sec: 305.19 - lr: 0.100000
2021-03-02 12:16:43,010 ----------------------------------------------------------------------------------------------------
2021-03-02 12:16:43,010 EPOCH 9 done: loss 6.5296 - lr 0.1000000
2021-03-02 12:16:45,523 DEV : loss 3.529794216156006 - score 0.9514
2021-03-02 12:16:45,602 BAD EPOCHS (no improvement): 0
2021-03-02 12:17:05,853 ----------------------------------------------------------------------------------------------------
2021-03-02 12:17:11,111 epoch 10 - iter 22/226 - loss 6.78708699 - samples/sec: 268.12 - lr: 0.100000
2021-03-02 12:17:15,770 epoch 10 - iter 44/226 - loss 6.65824565 - samples/sec: 302.47 - lr: 0.100000
2021-03-02 12:17:20,346 epoch 10 - iter 66/226 - loss 6.54451286 - samples/sec: 308.00 - lr: 0.100000
2021-03-02 12:17:25,252 epoch 10 - iter 88/226 - loss 6.51053966 - samples/sec: 287.23 - lr: 0.100000
2021-03-02 12:17:30,086 epoch 10 - iter 110/226 - loss 6.49176365 - samples/sec: 291.49 - lr: 0.100000
2021-03-02 12:17:34,744 epoch 10 - iter 132/226 - loss 6.44009249 - samples/sec: 302.48 - lr: 0.100000
2021-03-02 12:17:39,556 epoch 10 - iter 154/226 - loss 6.44174700 - samples/sec: 292.85 - lr: 0.100000
2021-03-02 12:17:44,060 epoch 10 - iter 176/226 - loss 6.44666531 - samples/sec: 312.80 - lr: 0.100000
2021-03-02 12:17:48,636 epoch 10 - iter 198/226 - loss 6.41738678 - samples/sec: 307.88 - lr: 0.100000
2021-03-02 12:17:53,375 epoch 10 - iter 220/226 - loss 6.41472057 - samples/sec: 297.32 - lr: 0.100000
2021-03-02 12:17:54,610 ----------------------------------------------------------------------------------------------------
2021-03-02 12:17:54,610 EPOCH 10 done: loss 6.4051 - lr 0.1000000
2021-03-02 12:17:57,142 DEV : loss 3.361401081085205 - score 0.9543
2021-03-02 12:17:57,218 BAD EPOCHS (no improvement): 0
2021-03-02 12:18:17,320 ----------------------------------------------------------------------------------------------------
2021-03-02 12:18:22,215 epoch 11 - iter 22/226 - loss 6.47508571 - samples/sec: 287.91 - lr: 0.100000
2021-03-02 12:18:26,897 epoch 11 - iter 44/226 - loss 6.34046380 - samples/sec: 300.90 - lr: 0.100000
2021-03-02 12:18:31,553 epoch 11 - iter 66/226 - loss 6.31181885 - samples/sec: 302.58 - lr: 0.100000
2021-03-02 12:18:36,424 epoch 11 - iter 88/226 - loss 6.30399061 - samples/sec: 289.27 - lr: 0.100000
2021-03-02 12:18:41,104 epoch 11 - iter 110/226 - loss 6.25753505 - samples/sec: 301.15 - lr: 0.100000
2021-03-02 12:18:45,644 epoch 11 - iter 132/226 - loss 6.22751645 - samples/sec: 310.43 - lr: 0.100000
2021-03-02 12:18:50,164 epoch 11 - iter 154/226 - loss 6.22338431 - samples/sec: 311.81 - lr: 0.100000
2021-03-02 12:18:55,190 epoch 11 - iter 176/226 - loss 6.21245406 - samples/sec: 280.40 - lr: 0.100000
2021-03-02 12:18:59,921 epoch 11 - iter 198/226 - loss 6.21484717 - samples/sec: 297.86 - lr: 0.100000
2021-03-02 12:19:04,994 epoch 11 - iter 220/226 - loss 6.22279377 - samples/sec: 277.76 - lr: 0.100000
2021-03-02 12:19:06,267 ----------------------------------------------------------------------------------------------------
2021-03-02 12:19:06,267 EPOCH 11 done: loss 6.2259 - lr 0.1000000
2021-03-02 12:19:08,794 DEV : loss 3.307126760482788 - score 0.9546
2021-03-02 12:19:08,870 BAD EPOCHS (no improvement): 0
2021-03-02 12:19:29,093 ----------------------------------------------------------------------------------------------------
2021-03-02 12:19:33,680 epoch 12 - iter 22/226 - loss 5.93327182 - samples/sec: 307.37 - lr: 0.100000
2021-03-02 12:19:38,162 epoch 12 - iter 44/226 - loss 6.07209740 - samples/sec: 314.28 - lr: 0.100000
2021-03-02 12:19:42,724 epoch 12 - iter 66/226 - loss 6.04609233 - samples/sec: 308.81 - lr: 0.100000
2021-03-02 12:19:47,252 epoch 12 - iter 88/226 - loss 6.08296212 - samples/sec: 311.18 - lr: 0.100000
2021-03-02 12:19:52,117 epoch 12 - iter 110/226 - loss 6.07122110 - samples/sec: 289.55 - lr: 0.100000
2021-03-02 12:19:57,158 epoch 12 - iter 132/226 - loss 6.12324406 - samples/sec: 279.47 - lr: 0.100000
2021-03-02 12:20:01,825 epoch 12 - iter 154/226 - loss 6.10276610 - samples/sec: 301.86 - lr: 0.100000
2021-03-02 12:20:06,457 epoch 12 - iter 176/226 - loss 6.07245450 - samples/sec: 304.15 - lr: 0.100000
2021-03-02 12:20:10,979 epoch 12 - iter 198/226 - loss 6.07118061 - samples/sec: 311.58 - lr: 0.100000
2021-03-02 12:20:15,586 epoch 12 - iter 220/226 - loss 6.07307480 - samples/sec: 305.91 - lr: 0.100000
2021-03-02 12:20:16,912 ----------------------------------------------------------------------------------------------------
2021-03-02 12:20:16,912 EPOCH 12 done: loss 6.0713 - lr 0.1000000
2021-03-02 12:20:19,457 DEV : loss 3.209815502166748 - score 0.9563
2021-03-02 12:20:19,532 BAD EPOCHS (no improvement): 0
2021-03-02 12:20:39,724 ----------------------------------------------------------------------------------------------------
2021-03-02 12:20:44,162 epoch 13 - iter 22/226 - loss 5.81342946 - samples/sec: 317.81 - lr: 0.100000
2021-03-02 12:20:48,842 epoch 13 - iter 44/226 - loss 5.78302381 - samples/sec: 301.13 - lr: 0.100000
2021-03-02 12:20:53,599 epoch 13 - iter 66/226 - loss 5.90845238 - samples/sec: 296.23 - lr: 0.100000
2021-03-02 12:20:58,208 epoch 13 - iter 88/226 - loss 5.94706044 - samples/sec: 305.71 - lr: 0.100000
2021-03-02 12:21:02,974 epoch 13 - iter 110/226 - loss 5.97966439 - samples/sec: 295.66 - lr: 0.100000
2021-03-02 12:21:07,653 epoch 13 - iter 132/226 - loss 5.93694615 - samples/sec: 301.17 - lr: 0.100000
2021-03-02 12:21:12,313 epoch 13 - iter 154/226 - loss 5.94886271 - samples/sec: 302.30 - lr: 0.100000
2021-03-02 12:21:17,217 epoch 13 - iter 176/226 - loss 5.94955487 - samples/sec: 287.31 - lr: 0.100000
2021-03-02 12:21:22,505 epoch 13 - iter 198/226 - loss 5.97424826 - samples/sec: 266.45 - lr: 0.100000
2021-03-02 12:21:26,923 epoch 13 - iter 220/226 - loss 5.96865309 - samples/sec: 318.92 - lr: 0.100000
2021-03-02 12:21:28,210 ----------------------------------------------------------------------------------------------------
2021-03-02 12:21:28,211 EPOCH 13 done: loss 5.9666 - lr 0.1000000
2021-03-02 12:21:30,759 DEV : loss 3.2812561988830566 - score 0.9551
2021-03-02 12:21:30,836 BAD EPOCHS (no improvement): 1
2021-03-02 12:21:30,839 ----------------------------------------------------------------------------------------------------
2021-03-02 12:21:35,554 epoch 14 - iter 22/226 - loss 5.81323225 - samples/sec: 298.88 - lr: 0.100000
2021-03-02 12:21:40,252 epoch 14 - iter 44/226 - loss 5.73190269 - samples/sec: 299.97 - lr: 0.100000
2021-03-02 12:21:48,188 epoch 14 - iter 66/226 - loss 5.76177246 - samples/sec: 177.51 - lr: 0.100000
2021-03-02 12:21:52,900 epoch 14 - iter 88/226 - loss 5.79665021 - samples/sec: 298.98 - lr: 0.100000
2021-03-02 12:21:57,848 epoch 14 - iter 110/226 - loss 5.75312194 - samples/sec: 284.71 - lr: 0.100000
2021-03-02 12:22:02,511 epoch 14 - iter 132/226 - loss 5.77030836 - samples/sec: 302.10 - lr: 0.100000
2021-03-02 12:22:07,125 epoch 14 - iter 154/226 - loss 5.83146126 - samples/sec: 305.33 - lr: 0.100000
2021-03-02 12:22:11,674 epoch 14 - iter 176/226 - loss 5.80009570 - samples/sec: 309.72 - lr: 0.100000
2021-03-02 12:22:16,248 epoch 14 - iter 198/226 - loss 5.79420889 - samples/sec: 307.97 - lr: 0.100000
2021-03-02 12:22:21,005 epoch 14 - iter 220/226 - loss 5.82198828 - samples/sec: 296.12 - lr: 0.100000
2021-03-02 12:22:22,181 ----------------------------------------------------------------------------------------------------
2021-03-02 12:22:22,181 EPOCH 14 done: loss 5.8154 - lr 0.1000000
2021-03-02 12:22:24,660 DEV : loss 3.0533204078674316 - score 0.9579
2021-03-02 12:22:24,737 BAD EPOCHS (no improvement): 0
2021-03-02 12:22:44,982 ----------------------------------------------------------------------------------------------------
2021-03-02 12:22:49,441 epoch 15 - iter 22/226 - loss 5.67295259 - samples/sec: 316.16 - lr: 0.100000
2021-03-02 12:22:54,101 epoch 15 - iter 44/226 - loss 5.68083337 - samples/sec: 302.38 - lr: 0.100000
2021-03-02 12:22:58,846 epoch 15 - iter 66/226 - loss 5.68906723 - samples/sec: 296.99 - lr: 0.100000
2021-03-02 12:23:03,586 epoch 15 - iter 88/226 - loss 5.68500325 - samples/sec: 297.27 - lr: 0.100000
2021-03-02 12:23:08,270 epoch 15 - iter 110/226 - loss 5.71941836 - samples/sec: 300.79 - lr: 0.100000
2021-03-02 12:23:13,556 epoch 15 - iter 132/226 - loss 5.73539855 - samples/sec: 266.58 - lr: 0.100000
2021-03-02 12:23:18,121 epoch 15 - iter 154/226 - loss 5.71976479 - samples/sec: 308.60 - lr: 0.100000
2021-03-02 12:23:22,669 epoch 15 - iter 176/226 - loss 5.70429908 - samples/sec: 309.82 - lr: 0.100000
2021-03-02 12:23:27,248 epoch 15 - iter 198/226 - loss 5.69654963 - samples/sec: 307.72 - lr: 0.100000
2021-03-02 12:23:32,011 epoch 15 - iter 220/226 - loss 5.73062568 - samples/sec: 295.79 - lr: 0.100000
2021-03-02 12:23:33,343 ----------------------------------------------------------------------------------------------------
2021-03-02 12:23:33,344 EPOCH 15 done: loss 5.7281 - lr 0.1000000
2021-03-02 12:23:35,885 DEV : loss 3.098984718322754 - score 0.9583
2021-03-02 12:23:35,961 BAD EPOCHS (no improvement): 0
2021-03-02 12:23:56,120 ----------------------------------------------------------------------------------------------------
2021-03-02 12:24:00,657 epoch 16 - iter 22/226 - loss 5.47865434 - samples/sec: 310.73 - lr: 0.100000
2021-03-02 12:24:05,279 epoch 16 - iter 44/226 - loss 5.64975263 - samples/sec: 304.75 - lr: 0.100000
2021-03-02 12:24:10,009 epoch 16 - iter 66/226 - loss 5.75589168 - samples/sec: 297.85 - lr: 0.100000
2021-03-02 12:24:14,660 epoch 16 - iter 88/226 - loss 5.76748121 - samples/sec: 302.86 - lr: 0.100000
2021-03-02 12:24:19,354 epoch 16 - iter 110/226 - loss 5.74031863 - samples/sec: 300.17 - lr: 0.100000
2021-03-02 12:24:24,003 epoch 16 - iter 132/226 - loss 5.70169360 - samples/sec: 303.08 - lr: 0.100000
2021-03-02 12:24:28,940 epoch 16 - iter 154/226 - loss 5.70688458 - samples/sec: 285.46 - lr: 0.100000
2021-03-02 12:24:33,461 epoch 16 - iter 176/226 - loss 5.68941835 - samples/sec: 311.68 - lr: 0.100000
2021-03-02 12:24:38,852 epoch 16 - iter 198/226 - loss 5.70622014 - samples/sec: 261.37 - lr: 0.100000
2021-03-02 12:24:43,407 epoch 16 - iter 220/226 - loss 5.66966486 - samples/sec: 309.35 - lr: 0.100000
2021-03-02 12:24:44,653 ----------------------------------------------------------------------------------------------------
2021-03-02 12:24:44,653 EPOCH 16 done: loss 5.6698 - lr 0.1000000
2021-03-02 12:24:47,200 DEV : loss 3.0057859420776367 - score 0.96
2021-03-02 12:24:47,276 BAD EPOCHS (no improvement): 0
2021-03-02 12:25:07,500 ----------------------------------------------------------------------------------------------------
2021-03-02 12:25:11,868 epoch 17 - iter 22/226 - loss 5.40437154 - samples/sec: 322.74 - lr: 0.100000
2021-03-02 12:25:16,158 epoch 17 - iter 44/226 - loss 5.42225520 - samples/sec: 328.36 - lr: 0.100000
2021-03-02 12:25:20,504 epoch 17 - iter 66/226 - loss 5.51035500 - samples/sec: 324.12 - lr: 0.100000
2021-03-02 12:25:25,388 epoch 17 - iter 88/226 - loss 5.51269228 - samples/sec: 288.40 - lr: 0.100000
2021-03-02 12:25:29,713 epoch 17 - iter 110/226 - loss 5.51827302 - samples/sec: 325.74 - lr: 0.100000
2021-03-02 12:25:34,198 epoch 17 - iter 132/226 - loss 5.52896604 - samples/sec: 314.07 - lr: 0.100000
2021-03-02 12:25:38,930 epoch 17 - iter 154/226 - loss 5.56259544 - samples/sec: 297.64 - lr: 0.100000
2021-03-02 12:25:43,195 epoch 17 - iter 176/226 - loss 5.52899651 - samples/sec: 330.28 - lr: 0.100000
2021-03-02 12:25:47,505 epoch 17 - iter 198/226 - loss 5.52833287 - samples/sec: 326.86 - lr: 0.100000
2021-03-02 12:25:51,837 epoch 17 - iter 220/226 - loss 5.52376431 - samples/sec: 325.23 - lr: 0.100000
2021-03-02 12:25:52,977 ----------------------------------------------------------------------------------------------------
2021-03-02 12:25:52,978 EPOCH 17 done: loss 5.5238 - lr 0.1000000
2021-03-02 12:25:55,534 DEV : loss 2.9501729011535645 - score 0.9609
2021-03-02 12:25:55,610 BAD EPOCHS (no improvement): 0
2021-03-02 12:26:15,851 ----------------------------------------------------------------------------------------------------
2021-03-02 12:26:20,379 epoch 18 - iter 22/226 - loss 5.36729024 - samples/sec: 311.42 - lr: 0.100000
2021-03-02 12:26:25,120 epoch 18 - iter 44/226 - loss 5.56896599 - samples/sec: 297.22 - lr: 0.100000
2021-03-02 12:26:30,019 epoch 18 - iter 66/226 - loss 5.59357333 - samples/sec: 287.64 - lr: 0.100000
2021-03-02 12:26:34,715 epoch 18 - iter 88/226 - loss 5.57033093 - samples/sec: 300.04 - lr: 0.100000
2021-03-02 12:26:39,382 epoch 18 - iter 110/226 - loss 5.53054149 - samples/sec: 301.91 - lr: 0.100000
2021-03-02 12:26:43,943 epoch 18 - iter 132/226 - loss 5.46317616 - samples/sec: 308.89 - lr: 0.100000
2021-03-02 12:26:48,844 epoch 18 - iter 154/226 - loss 5.48394228 - samples/sec: 287.51 - lr: 0.100000
2021-03-02 12:26:53,963 epoch 18 - iter 176/226 - loss 5.45764952 - samples/sec: 275.22 - lr: 0.100000
2021-03-02 12:26:58,640 epoch 18 - iter 198/226 - loss 5.44608234 - samples/sec: 301.21 - lr: 0.100000
2021-03-02 12:27:03,288 epoch 18 - iter 220/226 - loss 5.44462931 - samples/sec: 303.09 - lr: 0.100000
2021-03-02 12:27:04,562 ----------------------------------------------------------------------------------------------------
2021-03-02 12:27:04,563 EPOCH 18 done: loss 5.4394 - lr 0.1000000
2021-03-02 12:27:07,065 DEV : loss 2.8849360942840576 - score 0.9613
2021-03-02 12:27:07,141 BAD EPOCHS (no improvement): 0
2021-03-02 12:27:27,047 ----------------------------------------------------------------------------------------------------
2021-03-02 12:27:32,056 epoch 19 - iter 22/226 - loss 5.62021156 - samples/sec: 281.49 - lr: 0.100000
2021-03-02 12:27:36,806 epoch 19 - iter 44/226 - loss 5.54764419 - samples/sec: 296.70 - lr: 0.100000
2021-03-02 12:27:41,463 epoch 19 - iter 66/226 - loss 5.52793049 - samples/sec: 302.64 - lr: 0.100000
2021-03-02 12:27:45,972 epoch 19 - iter 88/226 - loss 5.39090671 - samples/sec: 312.55 - lr: 0.100000
2021-03-02 12:27:50,572 epoch 19 - iter 110/226 - loss 5.39549201 - samples/sec: 306.32 - lr: 0.100000
2021-03-02 12:27:55,119 epoch 19 - iter 132/226 - loss 5.41734884 - samples/sec: 309.91 - lr: 0.100000
2021-03-02 12:27:59,818 epoch 19 - iter 154/226 - loss 5.43162913 - samples/sec: 299.85 - lr: 0.100000
2021-03-02 12:28:04,737 epoch 19 - iter 176/226 - loss 5.45924142 - samples/sec: 286.45 - lr: 0.100000
2021-03-02 12:28:09,537 epoch 19 - iter 198/226 - loss 5.44415974 - samples/sec: 293.47 - lr: 0.100000
2021-03-02 12:28:14,196 epoch 19 - iter 220/226 - loss 5.43138959 - samples/sec: 302.38 - lr: 0.100000
2021-03-02 12:28:15,498 ----------------------------------------------------------------------------------------------------
2021-03-02 12:28:15,498 EPOCH 19 done: loss 5.4303 - lr 0.1000000
2021-03-02 12:28:18,037 DEV : loss 2.8557286262512207 - score 0.9618
2021-03-02 12:28:18,113 BAD EPOCHS (no improvement): 0
2021-03-02 12:28:38,223 ----------------------------------------------------------------------------------------------------
2021-03-02 12:28:43,638 epoch 20 - iter 22/226 - loss 5.54791353 - samples/sec: 260.26 - lr: 0.100000
2021-03-02 12:28:48,402 epoch 20 - iter 44/226 - loss 5.37923357 - samples/sec: 295.76 - lr: 0.100000
2021-03-02 12:28:52,989 epoch 20 - iter 66/226 - loss 5.29709985 - samples/sec: 307.24 - lr: 0.100000
2021-03-02 12:28:57,804 epoch 20 - iter 88/226 - loss 5.28773033 - samples/sec: 292.72 - lr: 0.100000
2021-03-02 12:29:02,454 epoch 20 - iter 110/226 - loss 5.29019342 - samples/sec: 303.13 - lr: 0.100000
2021-03-02 12:29:07,358 epoch 20 - iter 132/226 - loss 5.32519963 - samples/sec: 287.41 - lr: 0.100000
2021-03-02 12:29:11,987 epoch 20 - iter 154/226 - loss 5.30066517 - samples/sec: 304.59 - lr: 0.100000
2021-03-02 12:29:16,909 epoch 20 - iter 176/226 - loss 5.31746499 - samples/sec: 286.38 - lr: 0.100000
2021-03-02 12:29:21,725 epoch 20 - iter 198/226 - loss 5.32019624 - samples/sec: 292.61 - lr: 0.100000
2021-03-02 12:29:26,564 epoch 20 - iter 220/226 - loss 5.31880456 - samples/sec: 291.27 - lr: 0.100000
2021-03-02 12:29:27,876 ----------------------------------------------------------------------------------------------------
2021-03-02 12:29:27,877 EPOCH 20 done: loss 5.3224 - lr 0.1000000
2021-03-02 12:29:30,422 DEV : loss 2.8642454147338867 - score 0.9611
2021-03-02 12:29:30,498 BAD EPOCHS (no improvement): 1
2021-03-02 12:29:30,500 ----------------------------------------------------------------------------------------------------
2021-03-02 12:29:35,199 epoch 21 - iter 22/226 - loss 5.29684468 - samples/sec: 299.88 - lr: 0.100000
2021-03-02 12:29:40,139 epoch 21 - iter 44/226 - loss 5.22048582 - samples/sec: 285.25 - lr: 0.100000
2021-03-02 12:29:45,356 epoch 21 - iter 66/226 - loss 5.33847497 - samples/sec: 270.07 - lr: 0.100000
2021-03-02 12:29:50,202 epoch 21 - iter 88/226 - loss 5.28791927 - samples/sec: 290.75 - lr: 0.100000
2021-03-02 12:29:55,039 epoch 21 - iter 110/226 - loss 5.27051547 - samples/sec: 291.29 - lr: 0.100000
2021-03-02 12:29:59,942 epoch 21 - iter 132/226 - loss 5.29416718 - samples/sec: 287.32 - lr: 0.100000
2021-03-02 12:30:04,679 epoch 21 - iter 154/226 - loss 5.25552902 - samples/sec: 297.42 - lr: 0.100000
2021-03-02 12:30:09,439 epoch 21 - iter 176/226 - loss 5.27564516 - samples/sec: 296.02 - lr: 0.100000
2021-03-02 12:30:14,177 epoch 21 - iter 198/226 - loss 5.29864389 - samples/sec: 297.38 - lr: 0.100000
2021-03-02 12:30:18,730 epoch 21 - iter 220/226 - loss 5.29099408 - samples/sec: 309.51 - lr: 0.100000
2021-03-02 12:30:20,246 ----------------------------------------------------------------------------------------------------
2021-03-02 12:30:20,246 EPOCH 21 done: loss 5.2912 - lr 0.1000000
2021-03-02 12:30:22,803 DEV : loss 2.8174052238464355 - score 0.9618
2021-03-02 12:30:22,879 BAD EPOCHS (no improvement): 0
2021-03-02 12:30:43,079 ----------------------------------------------------------------------------------------------------
2021-03-02 12:30:47,557 epoch 22 - iter 22/226 - loss 4.91625257 - samples/sec: 315.05 - lr: 0.100000
2021-03-02 12:30:52,245 epoch 22 - iter 44/226 - loss 5.07609846 - samples/sec: 300.60 - lr: 0.100000
2021-03-02 12:30:56,927 epoch 22 - iter 66/226 - loss 5.11949955 - samples/sec: 301.02 - lr: 0.100000
2021-03-02 12:31:01,651 epoch 22 - iter 88/226 - loss 5.17461276 - samples/sec: 298.33 - lr: 0.100000
2021-03-02 12:31:06,458 epoch 22 - iter 110/226 - loss 5.18773076 - samples/sec: 293.16 - lr: 0.100000
2021-03-02 12:31:11,284 epoch 22 - iter 132/226 - loss 5.24821421 - samples/sec: 292.02 - lr: 0.100000
2021-03-02 12:31:15,885 epoch 22 - iter 154/226 - loss 5.28143251 - samples/sec: 306.23 - lr: 0.100000
2021-03-02 12:31:20,590 epoch 22 - iter 176/226 - loss 5.26149465 - samples/sec: 299.51 - lr: 0.100000
2021-03-02 12:31:25,199 epoch 22 - iter 198/226 - loss 5.26389915 - samples/sec: 305.68 - lr: 0.100000
2021-03-02 12:31:30,435 epoch 22 - iter 220/226 - loss 5.24284517 - samples/sec: 269.04 - lr: 0.100000
2021-03-02 12:31:31,790 ----------------------------------------------------------------------------------------------------
2021-03-02 12:31:31,790 EPOCH 22 done: loss 5.2386 - lr 0.1000000
2021-03-02 12:31:34,300 DEV : loss 2.7941231727600098 - score 0.962
2021-03-02 12:31:34,377 BAD EPOCHS (no improvement): 0
2021-03-02 12:31:54,349 ----------------------------------------------------------------------------------------------------
2021-03-02 12:31:59,057 epoch 23 - iter 22/226 - loss 5.12484332 - samples/sec: 299.45 - lr: 0.100000
2021-03-02 12:32:03,690 epoch 23 - iter 44/226 - loss 5.08499758 - samples/sec: 304.09 - lr: 0.100000
2021-03-02 12:32:08,285 epoch 23 - iter 66/226 - loss 5.12916445 - samples/sec: 306.70 - lr: 0.100000
2021-03-02 12:32:13,670 epoch 23 - iter 88/226 - loss 5.22015781 - samples/sec: 261.68 - lr: 0.100000
2021-03-02 12:32:18,430 epoch 23 - iter 110/226 - loss 5.22409021 - samples/sec: 296.08 - lr: 0.100000
2021-03-02 12:32:23,088 epoch 23 - iter 132/226 - loss 5.20612405 - samples/sec: 302.52 - lr: 0.100000
2021-03-02 12:32:27,703 epoch 23 - iter 154/226 - loss 5.16836774 - samples/sec: 305.31 - lr: 0.100000
2021-03-02 12:32:32,306 epoch 23 - iter 176/226 - loss 5.16979364 - samples/sec: 306.12 - lr: 0.100000
2021-03-02 12:32:36,817 epoch 23 - iter 198/226 - loss 5.15159683 - samples/sec: 312.31 - lr: 0.100000
2021-03-02 12:32:41,623 epoch 23 - iter 220/226 - loss 5.18111385 - samples/sec: 293.14 - lr: 0.100000
2021-03-02 12:32:43,142 ----------------------------------------------------------------------------------------------------
2021-03-02 12:32:43,143 EPOCH 23 done: loss 5.1746 - lr 0.1000000
2021-03-02 12:32:45,664 DEV : loss 2.8065927028656006 - score 0.9622
2021-03-02 12:32:45,739 BAD EPOCHS (no improvement): 0
2021-03-02 12:33:05,567 ----------------------------------------------------------------------------------------------------
2021-03-02 12:33:10,366 epoch 24 - iter 22/226 - loss 5.00725503 - samples/sec: 293.75 - lr: 0.100000
2021-03-02 12:33:15,281 epoch 24 - iter 44/226 - loss 5.10604477 - samples/sec: 286.65 - lr: 0.100000
2021-03-02 12:33:20,037 epoch 24 - iter 66/226 - loss 5.12667553 - samples/sec: 296.25 - lr: 0.100000
2021-03-02 12:33:24,625 epoch 24 - iter 88/226 - loss 5.12941140 - samples/sec: 307.08 - lr: 0.100000
2021-03-02 12:33:29,300 epoch 24 - iter 110/226 - loss 5.14817616 - samples/sec: 301.38 - lr: 0.100000
2021-03-02 12:33:34,168 epoch 24 - iter 132/226 - loss 5.14762985 - samples/sec: 289.47 - lr: 0.100000
2021-03-02 12:33:38,901 epoch 24 - iter 154/226 - loss 5.16203197 - samples/sec: 297.77 - lr: 0.100000
2021-03-02 12:33:44,477 epoch 24 - iter 176/226 - loss 5.16515541 - samples/sec: 252.77 - lr: 0.100000
2021-03-02 12:33:49,235 epoch 24 - iter 198/226 - loss 5.16429772 - samples/sec: 296.24 - lr: 0.100000
2021-03-02 12:33:53,860 epoch 24 - iter 220/226 - loss 5.15188599 - samples/sec: 304.76 - lr: 0.100000
2021-03-02 12:33:55,055 ----------------------------------------------------------------------------------------------------
2021-03-02 12:33:55,056 EPOCH 24 done: loss 5.1367 - lr 0.1000000
2021-03-02 12:33:57,618 DEV : loss 2.727389097213745 - score 0.9635
2021-03-02 12:33:57,694 BAD EPOCHS (no improvement): 0
2021-03-02 12:34:17,912 ----------------------------------------------------------------------------------------------------
2021-03-02 12:34:22,726 epoch 25 - iter 22/226 - loss 5.28198654 - samples/sec: 292.82 - lr: 0.100000
2021-03-02 12:34:27,515 epoch 25 - iter 44/226 - loss 5.09249745 - samples/sec: 294.19 - lr: 0.100000
2021-03-02 12:34:32,682 epoch 25 - iter 66/226 - loss 5.01988124 - samples/sec: 272.66 - lr: 0.100000
2021-03-02 12:34:37,463 epoch 25 - iter 88/226 - loss 5.07119706 - samples/sec: 294.67 - lr: 0.100000
2021-03-02 12:34:41,881 epoch 25 - iter 110/226 - loss 5.04652565 - samples/sec: 318.94 - lr: 0.100000
2021-03-02 12:34:46,415 epoch 25 - iter 132/226 - loss 5.05910077 - samples/sec: 310.71 - lr: 0.100000
2021-03-02 12:34:50,911 epoch 25 - iter 154/226 - loss 5.05163763 - samples/sec: 313.36 - lr: 0.100000
2021-03-02 12:34:55,527 epoch 25 - iter 176/226 - loss 5.04420760 - samples/sec: 305.22 - lr: 0.100000
2021-03-02 12:35:00,037 epoch 25 - iter 198/226 - loss 5.03629188 - samples/sec: 312.37 - lr: 0.100000
2021-03-02 12:35:04,729 epoch 25 - iter 220/226 - loss 5.04604290 - samples/sec: 300.27 - lr: 0.100000
2021-03-02 12:35:05,993 ----------------------------------------------------------------------------------------------------
2021-03-02 12:35:05,993 EPOCH 25 done: loss 5.0482 - lr 0.1000000
2021-03-02 12:35:08,545 DEV : loss 2.7719874382019043 - score 0.9622
2021-03-02 12:35:08,621 BAD EPOCHS (no improvement): 1
2021-03-02 12:35:08,623 ----------------------------------------------------------------------------------------------------
2021-03-02 12:35:13,597 epoch 26 - iter 22/226 - loss 5.15162227 - samples/sec: 283.30 - lr: 0.100000
2021-03-02 12:35:18,155 epoch 26 - iter 44/226 - loss 5.03666749 - samples/sec: 309.18 - lr: 0.100000
2021-03-02 12:35:22,736 epoch 26 - iter 66/226 - loss 5.01081347 - samples/sec: 307.58 - lr: 0.100000
2021-03-02 12:35:27,376 epoch 26 - iter 88/226 - loss 4.98357954 - samples/sec: 303.67 - lr: 0.100000
2021-03-02 12:35:32,123 epoch 26 - iter 110/226 - loss 4.99164035 - samples/sec: 296.87 - lr: 0.100000
2021-03-02 12:35:36,882 epoch 26 - iter 132/226 - loss 5.01850202 - samples/sec: 296.08 - lr: 0.100000
2021-03-02 12:35:41,405 epoch 26 - iter 154/226 - loss 5.02703794 - samples/sec: 311.50 - lr: 0.100000
2021-03-02 12:35:45,952 epoch 26 - iter 176/226 - loss 5.01172684 - samples/sec: 309.82 - lr: 0.100000
2021-03-02 12:35:50,544 epoch 26 - iter 198/226 - loss 5.00266386 - samples/sec: 306.83 - lr: 0.100000
2021-03-02 12:35:55,501 epoch 26 - iter 220/226 - loss 5.00865761 - samples/sec: 284.20 - lr: 0.100000
2021-03-02 12:35:56,659 ----------------------------------------------------------------------------------------------------
2021-03-02 12:35:56,659 EPOCH 26 done: loss 5.0126 - lr 0.1000000
2021-03-02 12:35:59,152 DEV : loss 2.764522075653076 - score 0.9626
2021-03-02 12:35:59,228 BAD EPOCHS (no improvement): 2
2021-03-02 12:35:59,230 ----------------------------------------------------------------------------------------------------
2021-03-02 12:36:04,228 epoch 27 - iter 22/226 - loss 4.85356351 - samples/sec: 281.87 - lr: 0.100000
2021-03-02 12:36:08,909 epoch 27 - iter 44/226 - loss 4.84228600 - samples/sec: 300.93 - lr: 0.100000
2021-03-02 12:36:13,528 epoch 27 - iter 66/226 - loss 4.94531875 - samples/sec: 304.98 - lr: 0.100000
2021-03-02 12:36:18,071 epoch 27 - iter 88/226 - loss 4.93810143 - samples/sec: 310.12 - lr: 0.100000
2021-03-02 12:36:22,587 epoch 27 - iter 110/226 - loss 4.95903680 - samples/sec: 311.92 - lr: 0.100000
2021-03-02 12:36:27,365 epoch 27 - iter 132/226 - loss 4.94009423 - samples/sec: 294.95 - lr: 0.100000
2021-03-02 12:36:32,083 epoch 27 - iter 154/226 - loss 4.93020583 - samples/sec: 298.63 - lr: 0.100000
2021-03-02 12:36:36,723 epoch 27 - iter 176/226 - loss 4.92373899 - samples/sec: 303.74 - lr: 0.100000
2021-03-02 12:36:41,265 epoch 27 - iter 198/226 - loss 4.92328239 - samples/sec: 310.29 - lr: 0.100000
2021-03-02 12:36:45,927 epoch 27 - iter 220/226 - loss 4.93293071 - samples/sec: 302.23 - lr: 0.100000
2021-03-02 12:36:47,171 ----------------------------------------------------------------------------------------------------
2021-03-02 12:36:47,171 EPOCH 27 done: loss 4.9323 - lr 0.1000000
2021-03-02 12:36:49,723 DEV : loss 2.750129461288452 - score 0.9635
2021-03-02 12:36:49,799 BAD EPOCHS (no improvement): 3
2021-03-02 12:36:49,801 ----------------------------------------------------------------------------------------------------
2021-03-02 12:36:54,446 epoch 28 - iter 22/226 - loss 4.90738409 - samples/sec: 303.39 - lr: 0.100000
2021-03-02 12:36:59,198 epoch 28 - iter 44/226 - loss 4.93428412 - samples/sec: 296.44 - lr: 0.100000
2021-03-02 12:37:03,681 epoch 28 - iter 66/226 - loss 4.89926485 - samples/sec: 314.26 - lr: 0.100000
2021-03-02 12:37:08,481 epoch 28 - iter 88/226 - loss 4.92100635 - samples/sec: 293.49 - lr: 0.100000
2021-03-02 12:37:13,190 epoch 28 - iter 110/226 - loss 4.92670722 - samples/sec: 299.17 - lr: 0.100000
2021-03-02 12:37:17,702 epoch 28 - iter 132/226 - loss 4.94115182 - samples/sec: 312.21 - lr: 0.100000
2021-03-02 12:37:22,148 epoch 28 - iter 154/226 - loss 4.93711268 - samples/sec: 316.87 - lr: 0.100000
2021-03-02 12:37:27,079 epoch 28 - iter 176/226 - loss 4.91109105 - samples/sec: 285.74 - lr: 0.100000
2021-03-02 12:37:31,758 epoch 28 - iter 198/226 - loss 4.93562892 - samples/sec: 301.09 - lr: 0.100000
2021-03-02 12:37:36,548 epoch 28 - iter 220/226 - loss 4.93658625 - samples/sec: 294.22 - lr: 0.100000
2021-03-02 12:37:37,789 ----------------------------------------------------------------------------------------------------
2021-03-02 12:37:37,790 EPOCH 28 done: loss 4.9341 - lr 0.1000000
2021-03-02 12:37:40,338 DEV : loss 2.7224578857421875 - score 0.9634
2021-03-02 12:37:40,414 BAD EPOCHS (no improvement): 4
2021-03-02 12:37:40,416 ----------------------------------------------------------------------------------------------------
2021-03-02 12:37:45,080 epoch 29 - iter 22/226 - loss 4.74876939 - samples/sec: 302.20 - lr: 0.050000
2021-03-02 12:37:49,590 epoch 29 - iter 44/226 - loss 4.84494490 - samples/sec: 312.45 - lr: 0.050000
2021-03-02 12:37:54,221 epoch 29 - iter 66/226 - loss 4.92065695 - samples/sec: 304.26 - lr: 0.050000
2021-03-02 12:37:58,613 epoch 29 - iter 88/226 - loss 4.89912525 - samples/sec: 320.74 - lr: 0.050000
2021-03-02 12:38:03,225 epoch 29 - iter 110/226 - loss 4.88014929 - samples/sec: 305.53 - lr: 0.050000
2021-03-02 12:38:08,001 epoch 29 - iter 132/226 - loss 4.89052036 - samples/sec: 294.99 - lr: 0.050000
2021-03-02 12:38:12,535 epoch 29 - iter 154/226 - loss 4.82561046 - samples/sec: 310.73 - lr: 0.050000
2021-03-02 12:38:17,731 epoch 29 - iter 176/226 - loss 4.84309944 - samples/sec: 271.14 - lr: 0.050000
2021-03-02 12:38:22,573 epoch 29 - iter 198/226 - loss 4.86181862 - samples/sec: 290.97 - lr: 0.050000
2021-03-02 12:38:27,176 epoch 29 - iter 220/226 - loss 4.83731847 - samples/sec: 306.05 - lr: 0.050000
2021-03-02 12:38:28,609 ----------------------------------------------------------------------------------------------------
2021-03-02 12:38:28,610 EPOCH 29 done: loss 4.8280 - lr 0.0500000
2021-03-02 12:38:31,161 DEV : loss 2.643817901611328 - score 0.965
2021-03-02 12:38:31,243 BAD EPOCHS (no improvement): 0
2021-03-02 12:38:51,437 ----------------------------------------------------------------------------------------------------
2021-03-02 12:38:56,195 epoch 30 - iter 22/226 - loss 4.94236820 - samples/sec: 296.48 - lr: 0.050000
2021-03-02 12:39:01,251 epoch 30 - iter 44/226 - loss 4.87298515 - samples/sec: 278.78 - lr: 0.050000
2021-03-02 12:39:05,847 epoch 30 - iter 66/226 - loss 4.87605703 - samples/sec: 306.66 - lr: 0.050000
2021-03-02 12:39:10,503 epoch 30 - iter 88/226 - loss 4.78286774 - samples/sec: 302.70 - lr: 0.050000
2021-03-02 12:39:15,109 epoch 30 - iter 110/226 - loss 4.70364232 - samples/sec: 305.89 - lr: 0.050000
2021-03-02 12:39:19,951 epoch 30 - iter 132/226 - loss 4.70045568 - samples/sec: 291.05 - lr: 0.050000
2021-03-02 12:39:24,667 epoch 30 - iter 154/226 - loss 4.69287239 - samples/sec: 298.81 - lr: 0.050000
2021-03-02 12:39:29,629 epoch 30 - iter 176/226 - loss 4.75636665 - samples/sec: 283.95 - lr: 0.050000
2021-03-02 12:39:37,328 epoch 30 - iter 198/226 - loss 4.76209223 - samples/sec: 182.94 - lr: 0.050000
2021-03-02 12:39:42,237 epoch 30 - iter 220/226 - loss 4.76975885 - samples/sec: 287.01 - lr: 0.050000
2021-03-02 12:39:43,446 ----------------------------------------------------------------------------------------------------
2021-03-02 12:39:43,447 EPOCH 30 done: loss 4.7663 - lr 0.0500000
2021-03-02 12:39:45,979 DEV : loss 2.677062511444092 - score 0.9645
2021-03-02 12:39:46,055 BAD EPOCHS (no improvement): 1
2021-03-02 12:39:46,057 ----------------------------------------------------------------------------------------------------
2021-03-02 12:39:50,787 epoch 31 - iter 22/226 - loss 4.81702999 - samples/sec: 298.05 - lr: 0.050000
2021-03-02 12:39:55,518 epoch 31 - iter 44/226 - loss 4.76242047 - samples/sec: 297.88 - lr: 0.050000
2021-03-02 12:40:00,325 epoch 31 - iter 66/226 - loss 4.73913543 - samples/sec: 293.18 - lr: 0.050000
2021-03-02 12:40:05,312 epoch 31 - iter 88/226 - loss 4.73114073 - samples/sec: 282.58 - lr: 0.050000
2021-03-02 12:40:09,822 epoch 31 - iter 110/226 - loss 4.73687147 - samples/sec: 312.49 - lr: 0.050000
2021-03-02 12:40:14,582 epoch 31 - iter 132/226 - loss 4.71922659 - samples/sec: 296.02 - lr: 0.050000
2021-03-02 12:40:19,372 epoch 31 - iter 154/226 - loss 4.72078651 - samples/sec: 294.17 - lr: 0.050000
2021-03-02 12:40:24,607 epoch 31 - iter 176/226 - loss 4.75659071 - samples/sec: 269.10 - lr: 0.050000
2021-03-02 12:40:29,279 epoch 31 - iter 198/226 - loss 4.74592781 - samples/sec: 301.61 - lr: 0.050000
2021-03-02 12:40:33,896 epoch 31 - iter 220/226 - loss 4.74337215 - samples/sec: 305.17 - lr: 0.050000
2021-03-02 12:40:35,174 ----------------------------------------------------------------------------------------------------
2021-03-02 12:40:35,174 EPOCH 31 done: loss 4.7402 - lr 0.0500000
2021-03-02 12:40:37,729 DEV : loss 2.6126627922058105 - score 0.9651
2021-03-02 12:40:37,805 BAD EPOCHS (no improvement): 0
2021-03-02 12:40:58,091 ----------------------------------------------------------------------------------------------------
2021-03-02 12:41:02,836 epoch 32 - iter 22/226 - loss 4.79364890 - samples/sec: 297.24 - lr: 0.050000
2021-03-02 12:41:07,472 epoch 32 - iter 44/226 - loss 4.86209493 - samples/sec: 303.98 - lr: 0.050000
2021-03-02 12:41:12,013 epoch 32 - iter 66/226 - loss 4.81317096 - samples/sec: 310.33 - lr: 0.050000
2021-03-02 12:41:16,747 epoch 32 - iter 88/226 - loss 4.80211742 - samples/sec: 297.67 - lr: 0.050000
2021-03-02 12:41:21,479 epoch 32 - iter 110/226 - loss 4.78087717 - samples/sec: 297.77 - lr: 0.050000
2021-03-02 12:41:26,160 epoch 32 - iter 132/226 - loss 4.76927685 - samples/sec: 301.02 - lr: 0.050000
2021-03-02 12:41:31,267 epoch 32 - iter 154/226 - loss 4.76803348 - samples/sec: 275.84 - lr: 0.050000
2021-03-02 12:41:36,006 epoch 32 - iter 176/226 - loss 4.75998304 - samples/sec: 297.30 - lr: 0.050000
2021-03-02 12:41:40,809 epoch 32 - iter 198/226 - loss 4.77127114 - samples/sec: 293.30 - lr: 0.050000
2021-03-02 12:41:45,410 epoch 32 - iter 220/226 - loss 4.75493640 - samples/sec: 306.23 - lr: 0.050000
2021-03-02 12:41:46,555 ----------------------------------------------------------------------------------------------------
2021-03-02 12:41:46,556 EPOCH 32 done: loss 4.7479 - lr 0.0500000
2021-03-02 12:41:49,068 DEV : loss 2.6256086826324463 - score 0.9652
2021-03-02 12:41:49,147 BAD EPOCHS (no improvement): 0
2021-03-02 12:42:09,409 ----------------------------------------------------------------------------------------------------
2021-03-02 12:42:13,722 epoch 33 - iter 22/226 - loss 4.58837919 - samples/sec: 326.93 - lr: 0.050000
2021-03-02 12:42:18,044 epoch 33 - iter 44/226 - loss 4.66200319 - samples/sec: 325.93 - lr: 0.050000
2021-03-02 12:42:22,500 epoch 33 - iter 66/226 - loss 4.65533241 - samples/sec: 316.21 - lr: 0.050000
2021-03-02 12:42:26,786 epoch 33 - iter 88/226 - loss 4.65853417 - samples/sec: 328.70 - lr: 0.050000
2021-03-02 12:42:31,333 epoch 33 - iter 110/226 - loss 4.72646942 - samples/sec: 309.83 - lr: 0.050000
2021-03-02 12:42:35,519 epoch 33 - iter 132/226 - loss 4.72529419 - samples/sec: 336.55 - lr: 0.050000
2021-03-02 12:42:40,135 epoch 33 - iter 154/226 - loss 4.74222747 - samples/sec: 305.21 - lr: 0.050000
2021-03-02 12:42:44,502 epoch 33 - iter 176/226 - loss 4.72501991 - samples/sec: 322.54 - lr: 0.050000
2021-03-02 12:42:48,806 epoch 33 - iter 198/226 - loss 4.69053067 - samples/sec: 327.34 - lr: 0.050000
2021-03-02 12:42:53,793 epoch 33 - iter 220/226 - loss 4.71906058 - samples/sec: 282.43 - lr: 0.050000
2021-03-02 12:42:54,961 ----------------------------------------------------------------------------------------------------
2021-03-02 12:42:54,962 EPOCH 33 done: loss 4.7318 - lr 0.0500000
2021-03-02 12:42:57,506 DEV : loss 2.627054452896118 - score 0.9642
2021-03-02 12:42:57,581 BAD EPOCHS (no improvement): 1
2021-03-02 12:42:57,584 ----------------------------------------------------------------------------------------------------
2021-03-02 12:43:01,988 epoch 34 - iter 22/226 - loss 4.81752751 - samples/sec: 319.87 - lr: 0.050000
2021-03-02 12:43:06,149 epoch 34 - iter 44/226 - loss 4.73167309 - samples/sec: 338.51 - lr: 0.050000
2021-03-02 12:43:10,500 epoch 34 - iter 66/226 - loss 4.67608619 - samples/sec: 323.71 - lr: 0.050000
2021-03-02 12:43:14,893 epoch 34 - iter 88/226 - loss 4.69116685 - samples/sec: 320.59 - lr: 0.050000
2021-03-02 12:43:19,760 epoch 34 - iter 110/226 - loss 4.70973603 - samples/sec: 289.44 - lr: 0.050000
2021-03-02 12:43:24,261 epoch 34 - iter 132/226 - loss 4.69013260 - samples/sec: 312.95 - lr: 0.050000
2021-03-02 12:43:28,593 epoch 34 - iter 154/226 - loss 4.67510394 - samples/sec: 325.17 - lr: 0.050000
2021-03-02 12:43:32,919 epoch 34 - iter 176/226 - loss 4.66639337 - samples/sec: 325.65 - lr: 0.050000
2021-03-02 12:43:37,264 epoch 34 - iter 198/226 - loss 4.68048262 - samples/sec: 324.31 - lr: 0.050000
2021-03-02 12:43:41,535 epoch 34 - iter 220/226 - loss 4.66508796 - samples/sec: 329.88 - lr: 0.050000
2021-03-02 12:43:42,771 ----------------------------------------------------------------------------------------------------
2021-03-02 12:43:42,772 EPOCH 34 done: loss 4.6667 - lr 0.0500000
2021-03-02 12:43:45,307 DEV : loss 2.652263641357422 - score 0.9644
2021-03-02 12:43:45,383 BAD EPOCHS (no improvement): 2
2021-03-02 12:43:45,384 ----------------------------------------------------------------------------------------------------
2021-03-02 12:43:49,779 epoch 35 - iter 22/226 - loss 4.72938051 - samples/sec: 320.68 - lr: 0.050000
2021-03-02 12:43:54,065 epoch 35 - iter 44/226 - loss 4.59454092 - samples/sec: 328.70 - lr: 0.050000
2021-03-02 12:43:58,530 epoch 35 - iter 66/226 - loss 4.62790721 - samples/sec: 315.48 - lr: 0.050000
2021-03-02 12:44:02,881 epoch 35 - iter 88/226 - loss 4.64621079 - samples/sec: 323.75 - lr: 0.050000
2021-03-02 12:44:07,075 epoch 35 - iter 110/226 - loss 4.61482221 - samples/sec: 335.90 - lr: 0.050000
2021-03-02 12:44:11,312 epoch 35 - iter 132/226 - loss 4.61588426 - samples/sec: 332.44 - lr: 0.050000
2021-03-02 12:44:16,243 epoch 35 - iter 154/226 - loss 4.63431394 - samples/sec: 285.68 - lr: 0.050000
2021-03-02 12:44:20,523 epoch 35 - iter 176/226 - loss 4.64901999 - samples/sec: 329.12 - lr: 0.050000
2021-03-02 12:44:25,227 epoch 35 - iter 198/226 - loss 4.65339260 - samples/sec: 299.43 - lr: 0.050000
2021-03-02 12:44:29,568 epoch 35 - iter 220/226 - loss 4.66191226 - samples/sec: 324.48 - lr: 0.050000
2021-03-02 12:44:30,796 ----------------------------------------------------------------------------------------------------
2021-03-02 12:44:30,796 EPOCH 35 done: loss 4.6575 - lr 0.0500000
2021-03-02 12:44:33,349 DEV : loss 2.5764060020446777 - score 0.9653
2021-03-02 12:44:33,432 BAD EPOCHS (no improvement): 0
2021-03-02 12:44:53,721 ----------------------------------------------------------------------------------------------------
2021-03-02 12:44:58,507 epoch 36 - iter 22/226 - loss 4.74134457 - samples/sec: 294.68 - lr: 0.050000
2021-03-02 12:45:03,006 epoch 36 - iter 44/226 - loss 4.64485334 - samples/sec: 313.22 - lr: 0.050000
2021-03-02 12:45:07,498 epoch 36 - iter 66/226 - loss 4.56865860 - samples/sec: 313.76 - lr: 0.050000
2021-03-02 12:45:12,075 epoch 36 - iter 88/226 - loss 4.54897283 - samples/sec: 307.88 - lr: 0.050000
2021-03-02 12:45:16,806 epoch 36 - iter 110/226 - loss 4.57288518 - samples/sec: 297.79 - lr: 0.050000
2021-03-02 12:45:21,551 epoch 36 - iter 132/226 - loss 4.60097296 - samples/sec: 296.95 - lr: 0.050000
2021-03-02 12:45:26,363 epoch 36 - iter 154/226 - loss 4.61330920 - samples/sec: 292.80 - lr: 0.050000
2021-03-02 12:45:31,525 epoch 36 - iter 176/226 - loss 4.62630832 - samples/sec: 272.90 - lr: 0.050000
2021-03-02 12:45:36,219 epoch 36 - iter 198/226 - loss 4.65080644 - samples/sec: 300.08 - lr: 0.050000
2021-03-02 12:45:41,378 epoch 36 - iter 220/226 - loss 4.64752742 - samples/sec: 273.08 - lr: 0.050000
2021-03-02 12:45:42,585 ----------------------------------------------------------------------------------------------------
2021-03-02 12:45:42,586 EPOCH 36 done: loss 4.6433 - lr 0.0500000
2021-03-02 12:45:45,100 DEV : loss 2.5746216773986816 - score 0.9655
2021-03-02 12:45:45,176 BAD EPOCHS (no improvement): 0
2021-03-02 12:46:05,395 ----------------------------------------------------------------------------------------------------
2021-03-02 12:46:10,235 epoch 37 - iter 22/226 - loss 4.66562238 - samples/sec: 291.36 - lr: 0.050000
2021-03-02 12:46:14,705 epoch 37 - iter 44/226 - loss 4.58943291 - samples/sec: 315.30 - lr: 0.050000
2021-03-02 12:46:19,429 epoch 37 - iter 66/226 - loss 4.55338003 - samples/sec: 298.30 - lr: 0.050000
2021-03-02 12:46:23,988 epoch 37 - iter 88/226 - loss 4.52589824 - samples/sec: 309.11 - lr: 0.050000
2021-03-02 12:46:29,150 epoch 37 - iter 110/226 - loss 4.54333296 - samples/sec: 272.96 - lr: 0.050000
2021-03-02 12:46:33,821 epoch 37 - iter 132/226 - loss 4.54862654 - samples/sec: 301.70 - lr: 0.050000
2021-03-02 12:46:38,597 epoch 37 - iter 154/226 - loss 4.55880527 - samples/sec: 295.02 - lr: 0.050000
2021-03-02 12:46:43,447 epoch 37 - iter 176/226 - loss 4.56151819 - samples/sec: 290.49 - lr: 0.050000
2021-03-02 12:46:48,146 epoch 37 - iter 198/226 - loss 4.56235231 - samples/sec: 299.88 - lr: 0.050000
2021-03-02 12:46:52,866 epoch 37 - iter 220/226 - loss 4.57658720 - samples/sec: 298.46 - lr: 0.050000
2021-03-02 12:46:54,080 ----------------------------------------------------------------------------------------------------
2021-03-02 12:46:54,080 EPOCH 37 done: loss 4.5759 - lr 0.0500000
2021-03-02 12:46:56,622 DEV : loss 2.5787086486816406 - score 0.9652
2021-03-02 12:46:56,698 BAD EPOCHS (no improvement): 1
2021-03-02 12:46:56,700 ----------------------------------------------------------------------------------------------------
2021-03-02 12:47:01,340 epoch 38 - iter 22/226 - loss 4.61333420 - samples/sec: 303.68 - lr: 0.050000
2021-03-02 12:47:05,974 epoch 38 - iter 44/226 - loss 4.69679946 - samples/sec: 304.02 - lr: 0.050000
2021-03-02 12:47:10,646 epoch 38 - iter 66/226 - loss 4.69750109 - samples/sec: 301.56 - lr: 0.050000
2021-03-02 12:47:15,171 epoch 38 - iter 88/226 - loss 4.61004476 - samples/sec: 311.32 - lr: 0.050000
2021-03-02 12:47:19,810 epoch 38 - iter 110/226 - loss 4.56752182 - samples/sec: 303.75 - lr: 0.050000
2021-03-02 12:47:24,314 epoch 38 - iter 132/226 - loss 4.54566523 - samples/sec: 312.92 - lr: 0.050000
2021-03-02 12:47:29,131 epoch 38 - iter 154/226 - loss 4.56347666 - samples/sec: 292.57 - lr: 0.050000
2021-03-02 12:47:33,942 epoch 38 - iter 176/226 - loss 4.59016239 - samples/sec: 292.93 - lr: 0.050000
2021-03-02 12:47:39,060 epoch 38 - iter 198/226 - loss 4.60928384 - samples/sec: 275.39 - lr: 0.050000
2021-03-02 12:47:43,843 epoch 38 - iter 220/226 - loss 4.60784330 - samples/sec: 294.60 - lr: 0.050000
2021-03-02 12:47:45,138 ----------------------------------------------------------------------------------------------------
2021-03-02 12:47:45,138 EPOCH 38 done: loss 4.6037 - lr 0.0500000
2021-03-02 12:47:47,670 DEV : loss 2.5750954151153564 - score 0.9655
2021-03-02 12:47:47,746 BAD EPOCHS (no improvement): 2
2021-03-02 12:47:47,747 ----------------------------------------------------------------------------------------------------
2021-03-02 12:47:52,343 epoch 39 - iter 22/226 - loss 4.63389741 - samples/sec: 306.63 - lr: 0.050000
2021-03-02 12:47:57,008 epoch 39 - iter 44/226 - loss 4.59737634 - samples/sec: 302.05 - lr: 0.050000
2021-03-02 12:48:01,858 epoch 39 - iter 66/226 - loss 4.61314438 - samples/sec: 290.55 - lr: 0.050000
2021-03-02 12:48:07,211 epoch 39 - iter 88/226 - loss 4.62947123 - samples/sec: 263.14 - lr: 0.050000
2021-03-02 12:48:11,942 epoch 39 - iter 110/226 - loss 4.61889617 - samples/sec: 297.88 - lr: 0.050000
2021-03-02 12:48:16,530 epoch 39 - iter 132/226 - loss 4.58621254 - samples/sec: 307.10 - lr: 0.050000
2021-03-02 12:48:21,233 epoch 39 - iter 154/226 - loss 4.58238402 - samples/sec: 299.57 - lr: 0.050000
2021-03-02 12:48:25,878 epoch 39 - iter 176/226 - loss 4.55898686 - samples/sec: 303.38 - lr: 0.050000
2021-03-02 12:48:30,345 epoch 39 - iter 198/226 - loss 4.58150591 - samples/sec: 315.40 - lr: 0.050000
2021-03-02 12:48:34,841 epoch 39 - iter 220/226 - loss 4.59080984 - samples/sec: 313.46 - lr: 0.050000
2021-03-02 12:48:36,136 ----------------------------------------------------------------------------------------------------
2021-03-02 12:48:36,136 EPOCH 39 done: loss 4.5999 - lr 0.0500000
2021-03-02 12:48:38,690 DEV : loss 2.546065330505371 - score 0.9663
2021-03-02 12:48:38,766 BAD EPOCHS (no improvement): 0
2021-03-02 12:48:59,020 ----------------------------------------------------------------------------------------------------
2021-03-02 12:49:03,937 epoch 40 - iter 22/226 - loss 4.64358004 - samples/sec: 286.84 - lr: 0.050000
2021-03-02 12:49:08,488 epoch 40 - iter 44/226 - loss 4.63632184 - samples/sec: 309.67 - lr: 0.050000
2021-03-02 12:49:13,189 epoch 40 - iter 66/226 - loss 4.60201601 - samples/sec: 299.78 - lr: 0.050000
2021-03-02 12:49:18,183 epoch 40 - iter 88/226 - loss 4.54942156 - samples/sec: 282.16 - lr: 0.050000
2021-03-02 12:49:23,074 epoch 40 - iter 110/226 - loss 4.54958003 - samples/sec: 288.05 - lr: 0.050000
2021-03-02 12:49:27,748 epoch 40 - iter 132/226 - loss 4.53282000 - samples/sec: 301.45 - lr: 0.050000
2021-03-02 12:49:33,023 epoch 40 - iter 154/226 - loss 4.53505744 - samples/sec: 267.11 - lr: 0.050000
2021-03-02 12:49:37,660 epoch 40 - iter 176/226 - loss 4.53257261 - samples/sec: 303.80 - lr: 0.050000
2021-03-02 12:49:42,135 epoch 40 - iter 198/226 - loss 4.51187128 - samples/sec: 314.88 - lr: 0.050000
2021-03-02 12:49:46,661 epoch 40 - iter 220/226 - loss 4.53239982 - samples/sec: 311.23 - lr: 0.050000
2021-03-02 12:49:47,888 ----------------------------------------------------------------------------------------------------
2021-03-02 12:49:47,889 EPOCH 40 done: loss 4.5349 - lr 0.0500000
2021-03-02 12:49:50,401 DEV : loss 2.5598995685577393 - score 0.9657
2021-03-02 12:49:50,477 BAD EPOCHS (no improvement): 1
2021-03-02 12:49:50,479 ----------------------------------------------------------------------------------------------------
2021-03-02 12:49:55,279 epoch 41 - iter 22/226 - loss 4.53217761 - samples/sec: 293.50 - lr: 0.050000
2021-03-02 12:50:00,038 epoch 41 - iter 44/226 - loss 4.47422222 - samples/sec: 296.02 - lr: 0.050000
2021-03-02 12:50:04,483 epoch 41 - iter 66/226 - loss 4.48689580 - samples/sec: 316.95 - lr: 0.050000
2021-03-02 12:50:09,272 epoch 41 - iter 88/226 - loss 4.49449517 - samples/sec: 294.28 - lr: 0.050000
2021-03-02 12:50:14,042 epoch 41 - iter 110/226 - loss 4.52423211 - samples/sec: 295.39 - lr: 0.050000
2021-03-02 12:50:18,835 epoch 41 - iter 132/226 - loss 4.53089580 - samples/sec: 294.01 - lr: 0.050000
2021-03-02 12:50:23,545 epoch 41 - iter 154/226 - loss 4.55354000 - samples/sec: 299.20 - lr: 0.050000
2021-03-02 12:50:28,149 epoch 41 - iter 176/226 - loss 4.55213997 - samples/sec: 306.10 - lr: 0.050000
2021-03-02 12:50:33,221 epoch 41 - iter 198/226 - loss 4.59473947 - samples/sec: 277.75 - lr: 0.050000
2021-03-02 12:50:37,801 epoch 41 - iter 220/226 - loss 4.57485249 - samples/sec: 307.66 - lr: 0.050000
2021-03-02 12:50:39,097 ----------------------------------------------------------------------------------------------------
2021-03-02 12:50:39,098 EPOCH 41 done: loss 4.5827 - lr 0.0500000
2021-03-02 12:50:41,614 DEV : loss 2.5540454387664795 - score 0.9662
2021-03-02 12:50:41,690 BAD EPOCHS (no improvement): 2
2021-03-02 12:50:41,691 ----------------------------------------------------------------------------------------------------
2021-03-02 12:50:46,776 epoch 42 - iter 22/226 - loss 4.63674366 - samples/sec: 277.12 - lr: 0.050000
2021-03-02 12:50:51,479 epoch 42 - iter 44/226 - loss 4.62284796 - samples/sec: 299.51 - lr: 0.050000
2021-03-02 12:50:56,388 epoch 42 - iter 66/226 - loss 4.58375311 - samples/sec: 287.01 - lr: 0.050000
2021-03-02 12:51:01,212 epoch 42 - iter 88/226 - loss 4.58960593 - samples/sec: 292.01 - lr: 0.050000
2021-03-02 12:51:05,732 epoch 42 - iter 110/226 - loss 4.54630809 - samples/sec: 311.70 - lr: 0.050000
2021-03-02 12:51:10,485 epoch 42 - iter 132/226 - loss 4.54713382 - samples/sec: 296.42 - lr: 0.050000
2021-03-02 12:51:15,189 epoch 42 - iter 154/226 - loss 4.53513395 - samples/sec: 299.48 - lr: 0.050000
2021-03-02 12:51:19,877 epoch 42 - iter 176/226 - loss 4.53678707 - samples/sec: 300.55 - lr: 0.050000
2021-03-02 12:51:24,556 epoch 42 - iter 198/226 - loss 4.51425143 - samples/sec: 301.19 - lr: 0.050000
2021-03-02 12:51:29,168 epoch 42 - iter 220/226 - loss 4.52784019 - samples/sec: 305.53 - lr: 0.050000
2021-03-02 12:51:30,345 ----------------------------------------------------------------------------------------------------
2021-03-02 12:51:30,345 EPOCH 42 done: loss 4.5270 - lr 0.0500000
2021-03-02 12:51:32,903 DEV : loss 2.551039695739746 - score 0.9659
2021-03-02 12:51:32,979 BAD EPOCHS (no improvement): 3
2021-03-02 12:51:32,982 ----------------------------------------------------------------------------------------------------
2021-03-02 12:51:37,547 epoch 43 - iter 22/226 - loss 4.44835871 - samples/sec: 308.76 - lr: 0.050000
2021-03-02 12:51:42,212 epoch 43 - iter 44/226 - loss 4.39226403 - samples/sec: 302.07 - lr: 0.050000
2021-03-02 12:51:46,921 epoch 43 - iter 66/226 - loss 4.46707423 - samples/sec: 299.19 - lr: 0.050000
2021-03-02 12:51:51,659 epoch 43 - iter 88/226 - loss 4.48570982 - samples/sec: 297.32 - lr: 0.050000
2021-03-02 12:51:56,652 epoch 43 - iter 110/226 - loss 4.47885021 - samples/sec: 282.20 - lr: 0.050000
2021-03-02 12:52:01,663 epoch 43 - iter 132/226 - loss 4.49540631 - samples/sec: 281.12 - lr: 0.050000
2021-03-02 12:52:06,337 epoch 43 - iter 154/226 - loss 4.48682522 - samples/sec: 301.47 - lr: 0.050000
2021-03-02 12:52:11,264 epoch 43 - iter 176/226 - loss 4.49818385 - samples/sec: 285.94 - lr: 0.050000
2021-03-02 12:52:15,867 epoch 43 - iter 198/226 - loss 4.51220885 - samples/sec: 306.03 - lr: 0.050000
2021-03-02 12:52:20,629 epoch 43 - iter 220/226 - loss 4.51271212 - samples/sec: 295.83 - lr: 0.050000
2021-03-02 12:52:21,845 ----------------------------------------------------------------------------------------------------
2021-03-02 12:52:21,845 EPOCH 43 done: loss 4.5177 - lr 0.0500000
2021-03-02 12:52:24,391 DEV : loss 2.552358388900757 - score 0.9656
2021-03-02 12:52:24,467 BAD EPOCHS (no improvement): 4
2021-03-02 12:52:24,468 ----------------------------------------------------------------------------------------------------
2021-03-02 12:52:29,169 epoch 44 - iter 22/226 - loss 4.53483922 - samples/sec: 299.82 - lr: 0.025000
2021-03-02 12:52:33,769 epoch 44 - iter 44/226 - loss 4.47704481 - samples/sec: 306.38 - lr: 0.025000
2021-03-02 12:52:38,491 epoch 44 - iter 66/226 - loss 4.46802417 - samples/sec: 298.41 - lr: 0.025000
2021-03-02 12:52:43,210 epoch 44 - iter 88/226 - loss 4.45023485 - samples/sec: 298.69 - lr: 0.025000
2021-03-02 12:52:47,843 epoch 44 - iter 110/226 - loss 4.45438900 - samples/sec: 304.12 - lr: 0.025000
2021-03-02 12:52:52,717 epoch 44 - iter 132/226 - loss 4.45798815 - samples/sec: 289.06 - lr: 0.025000
2021-03-02 12:52:57,833 epoch 44 - iter 154/226 - loss 4.47672932 - samples/sec: 275.40 - lr: 0.025000
2021-03-02 12:53:02,433 epoch 44 - iter 176/226 - loss 4.47722227 - samples/sec: 306.25 - lr: 0.025000
2021-03-02 12:53:07,058 epoch 44 - iter 198/226 - loss 4.49006726 - samples/sec: 304.62 - lr: 0.025000
2021-03-02 12:53:11,867 epoch 44 - iter 220/226 - loss 4.49785168 - samples/sec: 292.93 - lr: 0.025000
2021-03-02 12:53:13,047 ----------------------------------------------------------------------------------------------------
2021-03-02 12:53:13,047 EPOCH 44 done: loss 4.4977 - lr 0.0250000
2021-03-02 12:53:15,552 DEV : loss 2.5111753940582275 - score 0.9661
2021-03-02 12:53:15,628 BAD EPOCHS (no improvement): 1
2021-03-02 12:53:15,630 ----------------------------------------------------------------------------------------------------
2021-03-02 12:53:20,353 epoch 45 - iter 22/226 - loss 4.25703977 - samples/sec: 298.31 - lr: 0.025000
2021-03-02 12:53:25,596 epoch 45 - iter 44/226 - loss 4.35467847 - samples/sec: 268.66 - lr: 0.025000
2021-03-02 12:53:30,073 epoch 45 - iter 66/226 - loss 4.37810516 - samples/sec: 314.70 - lr: 0.025000
2021-03-02 12:53:34,789 epoch 45 - iter 88/226 - loss 4.43256490 - samples/sec: 298.79 - lr: 0.025000
2021-03-02 12:53:39,483 epoch 45 - iter 110/226 - loss 4.39250850 - samples/sec: 300.25 - lr: 0.025000
2021-03-02 12:53:44,182 epoch 45 - iter 132/226 - loss 4.38551346 - samples/sec: 299.88 - lr: 0.025000
2021-03-02 12:53:48,845 epoch 45 - iter 154/226 - loss 4.41887890 - samples/sec: 302.19 - lr: 0.025000
2021-03-02 12:53:53,655 epoch 45 - iter 176/226 - loss 4.44606431 - samples/sec: 292.94 - lr: 0.025000
2021-03-02 12:53:58,461 epoch 45 - iter 198/226 - loss 4.44184223 - samples/sec: 293.21 - lr: 0.025000
2021-03-02 12:54:03,072 epoch 45 - iter 220/226 - loss 4.44377752 - samples/sec: 305.60 - lr: 0.025000
2021-03-02 12:54:04,181 ----------------------------------------------------------------------------------------------------
2021-03-02 12:54:04,182 EPOCH 45 done: loss 4.4325 - lr 0.0250000
2021-03-02 12:54:06,712 DEV : loss 2.512387275695801 - score 0.9665
2021-03-02 12:54:06,787 BAD EPOCHS (no improvement): 0
2021-03-02 12:54:26,820 ----------------------------------------------------------------------------------------------------
2021-03-02 12:54:31,608 epoch 46 - iter 22/226 - loss 4.43599932 - samples/sec: 294.44 - lr: 0.025000
2021-03-02 12:54:36,312 epoch 46 - iter 44/226 - loss 4.39257556 - samples/sec: 299.57 - lr: 0.025000
2021-03-02 12:54:41,588 epoch 46 - iter 66/226 - loss 4.45226271 - samples/sec: 267.03 - lr: 0.025000
2021-03-02 12:54:46,118 epoch 46 - iter 88/226 - loss 4.43657781 - samples/sec: 311.03 - lr: 0.025000
2021-03-02 12:54:53,731 epoch 46 - iter 110/226 - loss 4.40404629 - samples/sec: 185.02 - lr: 0.025000
2021-03-02 12:54:58,440 epoch 46 - iter 132/226 - loss 4.41261988 - samples/sec: 299.19 - lr: 0.025000
2021-03-02 12:55:03,248 epoch 46 - iter 154/226 - loss 4.42326301 - samples/sec: 293.10 - lr: 0.025000
2021-03-02 12:55:07,935 epoch 46 - iter 176/226 - loss 4.43500176 - samples/sec: 300.72 - lr: 0.025000
2021-03-02 12:55:12,687 epoch 46 - iter 198/226 - loss 4.45748625 - samples/sec: 296.72 - lr: 0.025000
2021-03-02 12:55:17,571 epoch 46 - iter 220/226 - loss 4.45198697 - samples/sec: 288.63 - lr: 0.025000
2021-03-02 12:55:19,163 ----------------------------------------------------------------------------------------------------
2021-03-02 12:55:19,164 EPOCH 46 done: loss 4.4627 - lr 0.0250000
2021-03-02 12:55:21,689 DEV : loss 2.527057647705078 - score 0.9659
2021-03-02 12:55:21,764 BAD EPOCHS (no improvement): 1
2021-03-02 12:55:21,767 ----------------------------------------------------------------------------------------------------
2021-03-02 12:55:26,521 epoch 47 - iter 22/226 - loss 4.47424243 - samples/sec: 296.47 - lr: 0.025000
2021-03-02 12:55:31,237 epoch 47 - iter 44/226 - loss 4.37337967 - samples/sec: 298.85 - lr: 0.025000
2021-03-02 12:55:36,155 epoch 47 - iter 66/226 - loss 4.36831172 - samples/sec: 286.53 - lr: 0.025000
2021-03-02 12:55:41,303 epoch 47 - iter 88/226 - loss 4.37126090 - samples/sec: 273.68 - lr: 0.025000
2021-03-02 12:55:45,996 epoch 47 - iter 110/226 - loss 4.38029807 - samples/sec: 300.28 - lr: 0.025000
2021-03-02 12:55:50,854 epoch 47 - iter 132/226 - loss 4.39032895 - samples/sec: 290.02 - lr: 0.025000
2021-03-02 12:55:55,610 epoch 47 - iter 154/226 - loss 4.41571664 - samples/sec: 296.28 - lr: 0.025000
2021-03-02 12:56:00,373 epoch 47 - iter 176/226 - loss 4.42732771 - samples/sec: 295.86 - lr: 0.025000
2021-03-02 12:56:05,082 epoch 47 - iter 198/226 - loss 4.45415665 - samples/sec: 299.27 - lr: 0.025000
2021-03-02 12:56:09,802 epoch 47 - iter 220/226 - loss 4.45104000 - samples/sec: 298.51 - lr: 0.025000
2021-03-02 12:56:11,131 ----------------------------------------------------------------------------------------------------
2021-03-02 12:56:11,131 EPOCH 47 done: loss 4.4437 - lr 0.0250000
2021-03-02 12:56:13,699 DEV : loss 2.523621082305908 - score 0.9663
2021-03-02 12:56:13,774 BAD EPOCHS (no improvement): 2
2021-03-02 12:56:13,776 ----------------------------------------------------------------------------------------------------
2021-03-02 12:56:18,727 epoch 48 - iter 22/226 - loss 4.40344230 - samples/sec: 284.68 - lr: 0.025000
2021-03-02 12:56:24,018 epoch 48 - iter 44/226 - loss 4.43916564 - samples/sec: 266.42 - lr: 0.025000
2021-03-02 12:56:28,753 epoch 48 - iter 66/226 - loss 4.45499849 - samples/sec: 297.70 - lr: 0.025000
2021-03-02 12:56:33,401 epoch 48 - iter 88/226 - loss 4.42408887 - samples/sec: 303.23 - lr: 0.025000
2021-03-02 12:56:38,106 epoch 48 - iter 110/226 - loss 4.41501962 - samples/sec: 299.57 - lr: 0.025000
2021-03-02 12:56:42,892 epoch 48 - iter 132/226 - loss 4.42519227 - samples/sec: 294.41 - lr: 0.025000
2021-03-02 12:56:47,809 epoch 48 - iter 154/226 - loss 4.41243423 - samples/sec: 286.59 - lr: 0.025000
2021-03-02 12:56:52,528 epoch 48 - iter 176/226 - loss 4.42783547 - samples/sec: 298.61 - lr: 0.025000
2021-03-02 12:56:57,290 epoch 48 - iter 198/226 - loss 4.43767508 - samples/sec: 295.89 - lr: 0.025000
2021-03-02 12:57:02,399 epoch 48 - iter 220/226 - loss 4.45665219 - samples/sec: 275.80 - lr: 0.025000
2021-03-02 12:57:03,662 ----------------------------------------------------------------------------------------------------
2021-03-02 12:57:03,663 EPOCH 48 done: loss 4.4523 - lr 0.0250000
2021-03-02 12:57:06,199 DEV : loss 2.5409979820251465 - score 0.9656
2021-03-02 12:57:06,285 BAD EPOCHS (no improvement): 3
2021-03-02 12:57:06,287 ----------------------------------------------------------------------------------------------------
2021-03-02 12:57:11,285 epoch 49 - iter 22/226 - loss 4.43316382 - samples/sec: 281.90 - lr: 0.025000
2021-03-02 12:57:15,993 epoch 49 - iter 44/226 - loss 4.36996498 - samples/sec: 299.31 - lr: 0.025000
2021-03-02 12:57:20,951 epoch 49 - iter 66/226 - loss 4.33475170 - samples/sec: 284.28 - lr: 0.025000
2021-03-02 12:57:25,636 epoch 49 - iter 88/226 - loss 4.35022047 - samples/sec: 300.85 - lr: 0.025000
2021-03-02 12:57:30,419 epoch 49 - iter 110/226 - loss 4.39891044 - samples/sec: 294.77 - lr: 0.025000
2021-03-02 12:57:35,153 epoch 49 - iter 132/226 - loss 4.40406680 - samples/sec: 297.80 - lr: 0.025000
2021-03-02 12:57:39,868 epoch 49 - iter 154/226 - loss 4.40158756 - samples/sec: 298.92 - lr: 0.025000
2021-03-02 12:57:44,661 epoch 49 - iter 176/226 - loss 4.41793779 - samples/sec: 294.06 - lr: 0.025000
2021-03-02 12:57:49,934 epoch 49 - iter 198/226 - loss 4.40179443 - samples/sec: 267.19 - lr: 0.025000
2021-03-02 12:57:54,646 epoch 49 - iter 220/226 - loss 4.41186606 - samples/sec: 299.07 - lr: 0.025000
2021-03-02 12:57:55,913 ----------------------------------------------------------------------------------------------------
2021-03-02 12:57:55,914 EPOCH 49 done: loss 4.4144 - lr 0.0250000
2021-03-02 12:57:58,427 DEV : loss 2.519404172897339 - score 0.9658
2021-03-02 12:57:58,503 BAD EPOCHS (no improvement): 4
2021-03-02 12:57:58,504 ----------------------------------------------------------------------------------------------------
2021-03-02 12:58:03,547 epoch 50 - iter 22/226 - loss 4.52288650 - samples/sec: 279.46 - lr: 0.012500
2021-03-02 12:58:08,208 epoch 50 - iter 44/226 - loss 4.32049709 - samples/sec: 302.30 - lr: 0.012500
2021-03-02 12:58:13,251 epoch 50 - iter 66/226 - loss 4.36778089 - samples/sec: 279.37 - lr: 0.012500
2021-03-02 12:58:17,910 epoch 50 - iter 88/226 - loss 4.33578483 - samples/sec: 302.42 - lr: 0.012500
2021-03-02 12:58:22,819 epoch 50 - iter 110/226 - loss 4.34236643 - samples/sec: 287.03 - lr: 0.012500
2021-03-02 12:58:28,081 epoch 50 - iter 132/226 - loss 4.38018810 - samples/sec: 267.82 - lr: 0.012500
2021-03-02 12:58:32,815 epoch 50 - iter 154/226 - loss 4.37534744 - samples/sec: 297.75 - lr: 0.012500
2021-03-02 12:58:37,652 epoch 50 - iter 176/226 - loss 4.36774680 - samples/sec: 291.47 - lr: 0.012500
2021-03-02 12:58:42,357 epoch 50 - iter 198/226 - loss 4.37220785 - samples/sec: 299.61 - lr: 0.012500
2021-03-02 12:58:47,006 epoch 50 - iter 220/226 - loss 4.36893465 - samples/sec: 303.23 - lr: 0.012500
2021-03-02 12:58:48,283 ----------------------------------------------------------------------------------------------------
2021-03-02 12:58:48,284 EPOCH 50 done: loss 4.3708 - lr 0.0125000
2021-03-02 12:58:50,873 DEV : loss 2.512692928314209 - score 0.9662
2021-03-02 12:58:50,949 BAD EPOCHS (no improvement): 1
2021-03-02 12:58:50,952 ----------------------------------------------------------------------------------------------------
2021-03-02 12:58:55,763 epoch 51 - iter 22/226 - loss 4.36745239 - samples/sec: 292.92 - lr: 0.012500
2021-03-02 12:59:00,444 epoch 51 - iter 44/226 - loss 4.31490788 - samples/sec: 301.00 - lr: 0.012500
2021-03-02 12:59:05,278 epoch 51 - iter 66/226 - loss 4.36543504 - samples/sec: 291.47 - lr: 0.012500
2021-03-02 12:59:09,987 epoch 51 - iter 88/226 - loss 4.38624905 - samples/sec: 299.26 - lr: 0.012500
2021-03-02 12:59:14,888 epoch 51 - iter 110/226 - loss 4.43135549 - samples/sec: 287.50 - lr: 0.012500
2021-03-02 12:59:19,845 epoch 51 - iter 132/226 - loss 4.43995669 - samples/sec: 284.22 - lr: 0.012500
2021-03-02 12:59:24,605 epoch 51 - iter 154/226 - loss 4.41196162 - samples/sec: 296.03 - lr: 0.012500
2021-03-02 12:59:29,770 epoch 51 - iter 176/226 - loss 4.41252837 - samples/sec: 272.79 - lr: 0.012500
2021-03-02 12:59:34,494 epoch 51 - iter 198/226 - loss 4.40284634 - samples/sec: 298.30 - lr: 0.012500
2021-03-02 12:59:39,211 epoch 51 - iter 220/226 - loss 4.39833342 - samples/sec: 298.81 - lr: 0.012500
2021-03-02 12:59:40,558 ----------------------------------------------------------------------------------------------------
2021-03-02 12:59:40,559 EPOCH 51 done: loss 4.4031 - lr 0.0125000
2021-03-02 12:59:43,120 DEV : loss 2.497544288635254 - score 0.9662
2021-03-02 12:59:43,196 BAD EPOCHS (no improvement): 2
2021-03-02 12:59:43,197 ----------------------------------------------------------------------------------------------------
2021-03-02 12:59:47,873 epoch 52 - iter 22/226 - loss 4.35761580 - samples/sec: 301.53 - lr: 0.012500
2021-03-02 12:59:52,560 epoch 52 - iter 44/226 - loss 4.37267067 - samples/sec: 300.75 - lr: 0.012500
2021-03-02 12:59:57,434 epoch 52 - iter 66/226 - loss 4.35270388 - samples/sec: 289.11 - lr: 0.012500
2021-03-02 13:00:02,207 epoch 52 - iter 88/226 - loss 4.36679342 - samples/sec: 295.25 - lr: 0.012500
2021-03-02 13:00:07,018 epoch 52 - iter 110/226 - loss 4.36123488 - samples/sec: 292.93 - lr: 0.012500
2021-03-02 13:00:12,463 epoch 52 - iter 132/226 - loss 4.38560674 - samples/sec: 258.74 - lr: 0.012500
2021-03-02 13:00:17,247 epoch 52 - iter 154/226 - loss 4.38522749 - samples/sec: 294.54 - lr: 0.012500
2021-03-02 13:00:21,956 epoch 52 - iter 176/226 - loss 4.38007321 - samples/sec: 299.30 - lr: 0.012500
2021-03-02 13:00:26,713 epoch 52 - iter 198/226 - loss 4.37926297 - samples/sec: 296.17 - lr: 0.012500
2021-03-02 13:00:31,479 epoch 52 - iter 220/226 - loss 4.37746708 - samples/sec: 295.61 - lr: 0.012500
2021-03-02 13:00:32,731 ----------------------------------------------------------------------------------------------------
2021-03-02 13:00:32,731 EPOCH 52 done: loss 4.3786 - lr 0.0125000
2021-03-02 13:00:35,290 DEV : loss 2.495731830596924 - score 0.9665
2021-03-02 13:00:35,373 BAD EPOCHS (no improvement): 0
2021-03-02 13:00:55,253 ----------------------------------------------------------------------------------------------------
2021-03-02 13:01:00,684 epoch 53 - iter 22/226 - loss 4.41808392 - samples/sec: 259.68 - lr: 0.012500
2021-03-02 13:01:05,336 epoch 53 - iter 44/226 - loss 4.28492777 - samples/sec: 303.04 - lr: 0.012500
2021-03-02 13:01:10,120 epoch 53 - iter 66/226 - loss 4.31709432 - samples/sec: 294.60 - lr: 0.012500
2021-03-02 13:01:15,173 epoch 53 - iter 88/226 - loss 4.36027656 - samples/sec: 278.96 - lr: 0.012500
2021-03-02 13:01:20,003 epoch 53 - iter 110/226 - loss 4.33060572 - samples/sec: 291.75 - lr: 0.012500
2021-03-02 13:01:24,928 epoch 53 - iter 132/226 - loss 4.35057095 - samples/sec: 286.18 - lr: 0.012500
2021-03-02 13:01:29,639 epoch 53 - iter 154/226 - loss 4.35398010 - samples/sec: 299.12 - lr: 0.012500
2021-03-02 13:01:34,334 epoch 53 - iter 176/226 - loss 4.36657007 - samples/sec: 300.12 - lr: 0.012500
2021-03-02 13:01:38,996 epoch 53 - iter 198/226 - loss 4.36325798 - samples/sec: 302.25 - lr: 0.012500
2021-03-02 13:01:43,591 epoch 53 - iter 220/226 - loss 4.35401464 - samples/sec: 306.62 - lr: 0.012500
2021-03-02 13:01:44,857 ----------------------------------------------------------------------------------------------------
2021-03-02 13:01:44,858 EPOCH 53 done: loss 4.3553 - lr 0.0125000
2021-03-02 13:01:47,413 DEV : loss 2.4923317432403564 - score 0.967
2021-03-02 13:01:47,503 BAD EPOCHS (no improvement): 0
2021-03-02 13:02:07,711 ----------------------------------------------------------------------------------------------------
2021-03-02 13:02:12,536 epoch 54 - iter 22/226 - loss 4.29365627 - samples/sec: 292.38 - lr: 0.012500
2021-03-02 13:02:17,140 epoch 54 - iter 44/226 - loss 4.14311670 - samples/sec: 306.18 - lr: 0.012500
2021-03-02 13:02:21,884 epoch 54 - iter 66/226 - loss 4.23902313 - samples/sec: 297.20 - lr: 0.012500
2021-03-02 13:02:26,781 epoch 54 - iter 88/226 - loss 4.25932849 - samples/sec: 287.77 - lr: 0.012500
2021-03-02 13:02:31,978 epoch 54 - iter 110/226 - loss 4.28414596 - samples/sec: 271.20 - lr: 0.012500
2021-03-02 13:02:36,676 epoch 54 - iter 132/226 - loss 4.26848371 - samples/sec: 299.96 - lr: 0.012500
2021-03-02 13:02:41,864 epoch 54 - iter 154/226 - loss 4.30260206 - samples/sec: 271.63 - lr: 0.012500
2021-03-02 13:02:46,495 epoch 54 - iter 176/226 - loss 4.32594041 - samples/sec: 304.25 - lr: 0.012500
2021-03-02 13:02:51,281 epoch 54 - iter 198/226 - loss 4.33806878 - samples/sec: 294.43 - lr: 0.012500
2021-03-02 13:02:55,963 epoch 54 - iter 220/226 - loss 4.34028748 - samples/sec: 300.89 - lr: 0.012500
2021-03-02 13:02:57,212 ----------------------------------------------------------------------------------------------------
2021-03-02 13:02:57,212 EPOCH 54 done: loss 4.3337 - lr 0.0125000
2021-03-02 13:02:59,757 DEV : loss 2.5062971115112305 - score 0.9663
2021-03-02 13:02:59,852 BAD EPOCHS (no improvement): 1
2021-03-02 13:02:59,854 ----------------------------------------------------------------------------------------------------
2021-03-02 13:03:04,935 epoch 55 - iter 22/226 - loss 4.37281468 - samples/sec: 277.41 - lr: 0.012500
2021-03-02 13:03:10,083 epoch 55 - iter 44/226 - loss 4.40878471 - samples/sec: 273.75 - lr: 0.012500
2021-03-02 13:03:14,824 epoch 55 - iter 66/226 - loss 4.37000871 - samples/sec: 297.35 - lr: 0.012500
2021-03-02 13:03:19,569 epoch 55 - iter 88/226 - loss 4.37240147 - samples/sec: 297.08 - lr: 0.012500
2021-03-02 13:03:24,124 epoch 55 - iter 110/226 - loss 4.36359795 - samples/sec: 309.46 - lr: 0.012500
2021-03-02 13:03:28,930 epoch 55 - iter 132/226 - loss 4.37392366 - samples/sec: 293.30 - lr: 0.012500
2021-03-02 13:03:33,663 epoch 55 - iter 154/226 - loss 4.35191769 - samples/sec: 297.68 - lr: 0.012500
2021-03-02 13:03:38,330 epoch 55 - iter 176/226 - loss 4.35314028 - samples/sec: 302.02 - lr: 0.012500
2021-03-02 13:03:43,086 epoch 55 - iter 198/226 - loss 4.37046503 - samples/sec: 296.32 - lr: 0.012500
2021-03-02 13:03:47,734 epoch 55 - iter 220/226 - loss 4.37404605 - samples/sec: 303.12 - lr: 0.012500
2021-03-02 13:03:48,981 ----------------------------------------------------------------------------------------------------
2021-03-02 13:03:48,982 EPOCH 55 done: loss 4.3766 - lr 0.0125000
2021-03-02 13:03:51,485 DEV : loss 2.5064191818237305 - score 0.9667
2021-03-02 13:03:51,561 BAD EPOCHS (no improvement): 2
2021-03-02 13:03:51,563 ----------------------------------------------------------------------------------------------------
2021-03-02 13:03:56,315 epoch 56 - iter 22/226 - loss 4.18596031 - samples/sec: 296.54 - lr: 0.012500
2021-03-02 13:04:01,061 epoch 56 - iter 44/226 - loss 4.32869118 - samples/sec: 296.86 - lr: 0.012500
2021-03-02 13:04:05,661 epoch 56 - iter 66/226 - loss 4.29250342 - samples/sec: 306.24 - lr: 0.012500
2021-03-02 13:04:10,994 epoch 56 - iter 88/226 - loss 4.36594531 - samples/sec: 264.21 - lr: 0.012500
2021-03-02 13:04:15,648 epoch 56 - iter 110/226 - loss 4.38667238 - samples/sec: 302.81 - lr: 0.012500
2021-03-02 13:04:20,870 epoch 56 - iter 132/226 - loss 4.42383655 - samples/sec: 269.87 - lr: 0.012500
2021-03-02 13:04:25,689 epoch 56 - iter 154/226 - loss 4.41748182 - samples/sec: 292.55 - lr: 0.012500
2021-03-02 13:04:30,218 epoch 56 - iter 176/226 - loss 4.37229963 - samples/sec: 311.24 - lr: 0.012500
2021-03-02 13:04:34,884 epoch 56 - iter 198/226 - loss 4.36431905 - samples/sec: 302.08 - lr: 0.012500
2021-03-02 13:04:39,621 epoch 56 - iter 220/226 - loss 4.36178348 - samples/sec: 297.55 - lr: 0.012500
2021-03-02 13:04:40,828 ----------------------------------------------------------------------------------------------------
2021-03-02 13:04:40,829 EPOCH 56 done: loss 4.3661 - lr 0.0125000
2021-03-02 13:04:43,388 DEV : loss 2.4929332733154297 - score 0.9665
2021-03-02 13:04:43,464 BAD EPOCHS (no improvement): 3
2021-03-02 13:04:43,466 ----------------------------------------------------------------------------------------------------
2021-03-02 13:04:48,191 epoch 57 - iter 22/226 - loss 4.44058960 - samples/sec: 298.24 - lr: 0.012500
2021-03-02 13:04:53,401 epoch 57 - iter 44/226 - loss 4.42105395 - samples/sec: 270.48 - lr: 0.012500
2021-03-02 13:04:58,047 epoch 57 - iter 66/226 - loss 4.33351970 - samples/sec: 303.30 - lr: 0.012500
2021-03-02 13:05:02,773 epoch 57 - iter 88/226 - loss 4.32152503 - samples/sec: 298.08 - lr: 0.012500
2021-03-02 13:05:07,907 epoch 57 - iter 110/226 - loss 4.34187689 - samples/sec: 274.48 - lr: 0.012500
2021-03-02 13:05:12,687 epoch 57 - iter 132/226 - loss 4.34218816 - samples/sec: 294.72 - lr: 0.012500
2021-03-02 13:05:17,389 epoch 57 - iter 154/226 - loss 4.35290486 - samples/sec: 299.70 - lr: 0.012500
2021-03-02 13:05:22,031 epoch 57 - iter 176/226 - loss 4.34411273 - samples/sec: 303.59 - lr: 0.012500
2021-03-02 13:05:26,844 epoch 57 - iter 198/226 - loss 4.34143786 - samples/sec: 292.81 - lr: 0.012500
2021-03-02 13:05:31,653 epoch 57 - iter 220/226 - loss 4.34239795 - samples/sec: 293.05 - lr: 0.012500
2021-03-02 13:05:32,911 ----------------------------------------------------------------------------------------------------
2021-03-02 13:05:32,912 EPOCH 57 done: loss 4.3425 - lr 0.0125000
2021-03-02 13:05:35,467 DEV : loss 2.496828079223633 - score 0.9668
2021-03-02 13:05:35,543 BAD EPOCHS (no improvement): 4
2021-03-02 13:05:35,545 ----------------------------------------------------------------------------------------------------
2021-03-02 13:05:40,262 epoch 58 - iter 22/226 - loss 4.26106959 - samples/sec: 298.79 - lr: 0.006250
2021-03-02 13:05:45,003 epoch 58 - iter 44/226 - loss 4.20379769 - samples/sec: 297.30 - lr: 0.006250
2021-03-02 13:05:49,734 epoch 58 - iter 66/226 - loss 4.27125142 - samples/sec: 297.85 - lr: 0.006250
2021-03-02 13:05:54,563 epoch 58 - iter 88/226 - loss 4.29687362 - samples/sec: 291.87 - lr: 0.006250
2021-03-02 13:05:59,782 epoch 58 - iter 110/226 - loss 4.34484733 - samples/sec: 270.00 - lr: 0.006250
2021-03-02 13:06:04,522 epoch 58 - iter 132/226 - loss 4.32826960 - samples/sec: 297.28 - lr: 0.006250
2021-03-02 13:06:09,323 epoch 58 - iter 154/226 - loss 4.34798136 - samples/sec: 293.49 - lr: 0.006250
2021-03-02 13:06:14,075 epoch 58 - iter 176/226 - loss 4.36027329 - samples/sec: 296.57 - lr: 0.006250
2021-03-02 13:06:18,838 epoch 58 - iter 198/226 - loss 4.35729354 - samples/sec: 295.83 - lr: 0.006250
2021-03-02 13:06:23,863 epoch 58 - iter 220/226 - loss 4.36424401 - samples/sec: 280.37 - lr: 0.006250
2021-03-02 13:06:25,116 ----------------------------------------------------------------------------------------------------
2021-03-02 13:06:25,116 EPOCH 58 done: loss 4.3546 - lr 0.0062500
2021-03-02 13:06:27,676 DEV : loss 2.4992542266845703 - score 0.9665
2021-03-02 13:06:27,754 BAD EPOCHS (no improvement): 1
2021-03-02 13:06:27,756 ----------------------------------------------------------------------------------------------------
2021-03-02 13:06:32,717 epoch 59 - iter 22/226 - loss 4.36384961 - samples/sec: 284.10 - lr: 0.006250
2021-03-02 13:06:37,449 epoch 59 - iter 44/226 - loss 4.25188421 - samples/sec: 297.86 - lr: 0.006250
2021-03-02 13:06:42,254 epoch 59 - iter 66/226 - loss 4.34230541 - samples/sec: 293.45 - lr: 0.006250
2021-03-02 13:06:47,116 epoch 59 - iter 88/226 - loss 4.32333521 - samples/sec: 289.92 - lr: 0.006250
2021-03-02 13:06:51,901 epoch 59 - iter 110/226 - loss 4.32133018 - samples/sec: 294.55 - lr: 0.006250
2021-03-02 13:06:56,635 epoch 59 - iter 132/226 - loss 4.27928636 - samples/sec: 297.73 - lr: 0.006250
2021-03-02 13:07:01,807 epoch 59 - iter 154/226 - loss 4.30622992 - samples/sec: 272.41 - lr: 0.006250
2021-03-02 13:07:06,442 epoch 59 - iter 176/226 - loss 4.28481213 - samples/sec: 304.04 - lr: 0.006250
2021-03-02 13:07:11,149 epoch 59 - iter 198/226 - loss 4.29091401 - samples/sec: 299.38 - lr: 0.006250
2021-03-02 13:07:15,955 epoch 59 - iter 220/226 - loss 4.30595973 - samples/sec: 293.18 - lr: 0.006250
2021-03-02 13:07:17,283 ----------------------------------------------------------------------------------------------------
2021-03-02 13:07:17,283 EPOCH 59 done: loss 4.3105 - lr 0.0062500
2021-03-02 13:07:19,780 DEV : loss 2.4926538467407227 - score 0.9666
2021-03-02 13:07:19,856 BAD EPOCHS (no improvement): 2
2021-03-02 13:07:19,858 ----------------------------------------------------------------------------------------------------
2021-03-02 13:07:24,481 epoch 60 - iter 22/226 - loss 4.09951949 - samples/sec: 304.83 - lr: 0.006250
2021-03-02 13:07:29,273 epoch 60 - iter 44/226 - loss 4.09169726 - samples/sec: 294.03 - lr: 0.006250
2021-03-02 13:07:34,016 epoch 60 - iter 66/226 - loss 4.23531460 - samples/sec: 297.08 - lr: 0.006250
2021-03-02 13:07:38,782 epoch 60 - iter 88/226 - loss 4.23644064 - samples/sec: 295.58 - lr: 0.006250
2021-03-02 13:07:43,491 epoch 60 - iter 110/226 - loss 4.26407306 - samples/sec: 299.27 - lr: 0.006250
2021-03-02 13:07:48,337 epoch 60 - iter 132/226 - loss 4.27595105 - samples/sec: 290.84 - lr: 0.006250
2021-03-02 13:07:53,184 epoch 60 - iter 154/226 - loss 4.28220802 - samples/sec: 290.81 - lr: 0.006250
2021-03-02 13:07:57,924 epoch 60 - iter 176/226 - loss 4.29955173 - samples/sec: 297.39 - lr: 0.006250
2021-03-02 13:08:03,109 epoch 60 - iter 198/226 - loss 4.32500822 - samples/sec: 271.79 - lr: 0.006250
2021-03-02 13:08:07,797 epoch 60 - iter 220/226 - loss 4.32919365 - samples/sec: 300.63 - lr: 0.006250
2021-03-02 13:08:09,069 ----------------------------------------------------------------------------------------------------
2021-03-02 13:08:09,070 EPOCH 60 done: loss 4.3229 - lr 0.0062500
2021-03-02 13:08:11,629 DEV : loss 2.4904088973999023 - score 0.9664
2021-03-02 13:08:11,705 BAD EPOCHS (no improvement): 3
2021-03-02 13:08:11,708 ----------------------------------------------------------------------------------------------------
2021-03-02 13:08:16,330 epoch 61 - iter 22/226 - loss 4.20603855 - samples/sec: 304.91 - lr: 0.006250
2021-03-02 13:08:21,083 epoch 61 - iter 44/226 - loss 4.15998923 - samples/sec: 296.47 - lr: 0.006250
2021-03-02 13:08:26,474 epoch 61 - iter 66/226 - loss 4.29516100 - samples/sec: 261.36 - lr: 0.006250
2021-03-02 13:08:31,271 epoch 61 - iter 88/226 - loss 4.30427675 - samples/sec: 293.74 - lr: 0.006250
2021-03-02 13:08:36,318 epoch 61 - iter 110/226 - loss 4.31281058 - samples/sec: 279.14 - lr: 0.006250
2021-03-02 13:08:40,985 epoch 61 - iter 132/226 - loss 4.30920996 - samples/sec: 301.91 - lr: 0.006250
2021-03-02 13:08:45,757 epoch 61 - iter 154/226 - loss 4.30101481 - samples/sec: 295.30 - lr: 0.006250
2021-03-02 13:08:50,426 epoch 61 - iter 176/226 - loss 4.31437570 - samples/sec: 301.78 - lr: 0.006250
2021-03-02 13:08:55,027 epoch 61 - iter 198/226 - loss 4.30838462 - samples/sec: 306.39 - lr: 0.006250
2021-03-02 13:08:59,824 epoch 61 - iter 220/226 - loss 4.31669483 - samples/sec: 293.88 - lr: 0.006250
2021-03-02 13:09:01,052 ----------------------------------------------------------------------------------------------------
2021-03-02 13:09:01,053 EPOCH 61 done: loss 4.3158 - lr 0.0062500
2021-03-02 13:09:06,599 DEV : loss 2.4885008335113525 - score 0.967
2021-03-02 13:09:06,675 BAD EPOCHS (no improvement): 0
2021-03-02 13:09:26,755 ----------------------------------------------------------------------------------------------------
2021-03-02 13:09:31,712 epoch 62 - iter 22/226 - loss 4.13374297 - samples/sec: 284.33 - lr: 0.006250
2021-03-02 13:09:36,658 epoch 62 - iter 44/226 - loss 4.29302059 - samples/sec: 284.80 - lr: 0.006250
2021-03-02 13:09:41,410 epoch 62 - iter 66/226 - loss 4.29262086 - samples/sec: 296.51 - lr: 0.006250
2021-03-02 13:09:46,143 epoch 62 - iter 88/226 - loss 4.31671095 - samples/sec: 297.62 - lr: 0.006250
2021-03-02 13:09:50,713 epoch 62 - iter 110/226 - loss 4.30470178 - samples/sec: 308.28 - lr: 0.006250
2021-03-02 13:09:55,389 epoch 62 - iter 132/226 - loss 4.26259225 - samples/sec: 301.30 - lr: 0.006250
2021-03-02 13:09:59,928 epoch 62 - iter 154/226 - loss 4.27064904 - samples/sec: 310.43 - lr: 0.006250
2021-03-02 13:10:05,178 epoch 62 - iter 176/226 - loss 4.28858240 - samples/sec: 268.34 - lr: 0.006250
2021-03-02 13:10:09,783 epoch 62 - iter 198/226 - loss 4.28246615 - samples/sec: 305.95 - lr: 0.006250
2021-03-02 13:10:14,604 epoch 62 - iter 220/226 - loss 4.31031967 - samples/sec: 292.22 - lr: 0.006250
2021-03-02 13:10:15,884 ----------------------------------------------------------------------------------------------------
2021-03-02 13:10:15,884 EPOCH 62 done: loss 4.3063 - lr 0.0062500
2021-03-02 13:10:18,447 DEV : loss 2.4861385822296143 - score 0.9669
2021-03-02 13:10:18,523 BAD EPOCHS (no improvement): 1
2021-03-02 13:10:18,526 ----------------------------------------------------------------------------------------------------
2021-03-02 13:10:23,162 epoch 63 - iter 22/226 - loss 4.33507828 - samples/sec: 304.00 - lr: 0.006250
2021-03-02 13:10:27,892 epoch 63 - iter 44/226 - loss 4.25919961 - samples/sec: 298.08 - lr: 0.006250
2021-03-02 13:10:32,670 epoch 63 - iter 66/226 - loss 4.25104011 - samples/sec: 295.01 - lr: 0.006250
2021-03-02 13:10:37,377 epoch 63 - iter 88/226 - loss 4.24235984 - samples/sec: 299.46 - lr: 0.006250
2021-03-02 13:10:42,191 epoch 63 - iter 110/226 - loss 4.26371402 - samples/sec: 292.78 - lr: 0.006250
2021-03-02 13:10:46,842 epoch 63 - iter 132/226 - loss 4.27632514 - samples/sec: 303.02 - lr: 0.006250
2021-03-02 13:10:51,653 epoch 63 - iter 154/226 - loss 4.29277644 - samples/sec: 292.88 - lr: 0.006250
2021-03-02 13:10:56,472 epoch 63 - iter 176/226 - loss 4.34497060 - samples/sec: 292.47 - lr: 0.006250
2021-03-02 13:11:01,177 epoch 63 - iter 198/226 - loss 4.32045084 - samples/sec: 299.44 - lr: 0.006250
2021-03-02 13:11:06,575 epoch 63 - iter 220/226 - loss 4.32461584 - samples/sec: 261.03 - lr: 0.006250
2021-03-02 13:11:07,866 ----------------------------------------------------------------------------------------------------
2021-03-02 13:11:07,866 EPOCH 63 done: loss 4.3195 - lr 0.0062500
2021-03-02 13:11:10,377 DEV : loss 2.4990551471710205 - score 0.9663
2021-03-02 13:11:10,464 BAD EPOCHS (no improvement): 2
2021-03-02 13:11:10,466 ----------------------------------------------------------------------------------------------------
2021-03-02 13:11:15,068 epoch 64 - iter 22/226 - loss 4.25240617 - samples/sec: 306.16 - lr: 0.006250
2021-03-02 13:11:19,742 epoch 64 - iter 44/226 - loss 4.37724643 - samples/sec: 301.47 - lr: 0.006250
2021-03-02 13:11:24,371 epoch 64 - iter 66/226 - loss 4.37063458 - samples/sec: 304.39 - lr: 0.006250
2021-03-02 13:11:29,490 epoch 64 - iter 88/226 - loss 4.36685121 - samples/sec: 275.31 - lr: 0.006250
2021-03-02 13:11:34,283 epoch 64 - iter 110/226 - loss 4.37177055 - samples/sec: 294.04 - lr: 0.006250
2021-03-02 13:11:39,662 epoch 64 - iter 132/226 - loss 4.33910349 - samples/sec: 262.01 - lr: 0.006250
2021-03-02 13:11:44,389 epoch 64 - iter 154/226 - loss 4.33639950 - samples/sec: 298.24 - lr: 0.006250
2021-03-02 13:11:49,201 epoch 64 - iter 176/226 - loss 4.32123409 - samples/sec: 292.88 - lr: 0.006250
2021-03-02 13:11:53,964 epoch 64 - iter 198/226 - loss 4.31215822 - samples/sec: 295.88 - lr: 0.006250
2021-03-02 13:11:58,627 epoch 64 - iter 220/226 - loss 4.33287764 - samples/sec: 302.26 - lr: 0.006250
2021-03-02 13:11:59,934 ----------------------------------------------------------------------------------------------------
2021-03-02 13:11:59,935 EPOCH 64 done: loss 4.3377 - lr 0.0062500
2021-03-02 13:12:02,481 DEV : loss 2.4925572872161865 - score 0.9664
2021-03-02 13:12:02,557 BAD EPOCHS (no improvement): 3
2021-03-02 13:12:02,560 ----------------------------------------------------------------------------------------------------
2021-03-02 13:12:07,375 epoch 65 - iter 22/226 - loss 4.36403564 - samples/sec: 292.61 - lr: 0.006250
2021-03-02 13:12:12,516 epoch 65 - iter 44/226 - loss 4.34434488 - samples/sec: 274.07 - lr: 0.006250
2021-03-02 13:12:17,480 epoch 65 - iter 66/226 - loss 4.33470583 - samples/sec: 283.79 - lr: 0.006250
2021-03-02 13:12:22,091 epoch 65 - iter 88/226 - loss 4.29391651 - samples/sec: 305.57 - lr: 0.006250
2021-03-02 13:12:26,794 epoch 65 - iter 110/226 - loss 4.26951983 - samples/sec: 299.62 - lr: 0.006250
2021-03-02 13:12:31,697 epoch 65 - iter 132/226 - loss 4.28317517 - samples/sec: 287.39 - lr: 0.006250
2021-03-02 13:12:36,376 epoch 65 - iter 154/226 - loss 4.28415810 - samples/sec: 301.20 - lr: 0.006250
2021-03-02 13:12:41,148 epoch 65 - iter 176/226 - loss 4.32618828 - samples/sec: 295.39 - lr: 0.006250
2021-03-02 13:12:45,954 epoch 65 - iter 198/226 - loss 4.32527640 - samples/sec: 293.33 - lr: 0.006250
2021-03-02 13:12:50,760 epoch 65 - iter 220/226 - loss 4.29527885 - samples/sec: 293.23 - lr: 0.006250
2021-03-02 13:12:51,999 ----------------------------------------------------------------------------------------------------
2021-03-02 13:12:52,000 EPOCH 65 done: loss 4.3012 - lr 0.0062500
2021-03-02 13:12:54,556 DEV : loss 2.484386920928955 - score 0.9666
2021-03-02 13:12:54,631 BAD EPOCHS (no improvement): 4
2021-03-02 13:12:54,633 ----------------------------------------------------------------------------------------------------
2021-03-02 13:12:59,767 epoch 66 - iter 22/226 - loss 4.18128212 - samples/sec: 274.48 - lr: 0.003125
2021-03-02 13:13:04,666 epoch 66 - iter 44/226 - loss 4.28490644 - samples/sec: 287.66 - lr: 0.003125
2021-03-02 13:13:09,469 epoch 66 - iter 66/226 - loss 4.31464386 - samples/sec: 293.40 - lr: 0.003125
2021-03-02 13:13:14,019 epoch 66 - iter 88/226 - loss 4.30477922 - samples/sec: 309.70 - lr: 0.003125
2021-03-02 13:13:18,565 epoch 66 - iter 110/226 - loss 4.29704779 - samples/sec: 309.93 - lr: 0.003125
2021-03-02 13:13:23,378 epoch 66 - iter 132/226 - loss 4.30234110 - samples/sec: 292.76 - lr: 0.003125
2021-03-02 13:13:28,089 epoch 66 - iter 154/226 - loss 4.31574966 - samples/sec: 299.12 - lr: 0.003125
2021-03-02 13:13:32,845 epoch 66 - iter 176/226 - loss 4.32172206 - samples/sec: 296.29 - lr: 0.003125
2021-03-02 13:13:37,538 epoch 66 - iter 198/226 - loss 4.34454674 - samples/sec: 300.23 - lr: 0.003125
2021-03-02 13:13:42,534 epoch 66 - iter 220/226 - loss 4.33337953 - samples/sec: 282.00 - lr: 0.003125
2021-03-02 13:13:43,733 ----------------------------------------------------------------------------------------------------
2021-03-02 13:13:43,733 EPOCH 66 done: loss 4.3291 - lr 0.0031250
2021-03-02 13:13:46,293 DEV : loss 2.491257667541504 - score 0.9667
2021-03-02 13:13:46,368 BAD EPOCHS (no improvement): 1
2021-03-02 13:13:46,371 ----------------------------------------------------------------------------------------------------
2021-03-02 13:13:51,113 epoch 67 - iter 22/226 - loss 4.27824794 - samples/sec: 297.22 - lr: 0.003125
2021-03-02 13:13:55,993 epoch 67 - iter 44/226 - loss 4.21884942 - samples/sec: 288.85 - lr: 0.003125
2021-03-02 13:14:00,932 epoch 67 - iter 66/226 - loss 4.32129142 - samples/sec: 285.40 - lr: 0.003125
2021-03-02 13:14:05,990 epoch 67 - iter 88/226 - loss 4.36543566 - samples/sec: 278.62 - lr: 0.003125
2021-03-02 13:14:10,516 epoch 67 - iter 110/226 - loss 4.33429869 - samples/sec: 311.47 - lr: 0.003125
2021-03-02 13:14:15,250 epoch 67 - iter 132/226 - loss 4.35361455 - samples/sec: 297.68 - lr: 0.003125
2021-03-02 13:14:20,149 epoch 67 - iter 154/226 - loss 4.33627153 - samples/sec: 287.66 - lr: 0.003125
2021-03-02 13:14:25,379 epoch 67 - iter 176/226 - loss 4.35663834 - samples/sec: 269.40 - lr: 0.003125
2021-03-02 13:14:30,170 epoch 67 - iter 198/226 - loss 4.34670517 - samples/sec: 294.14 - lr: 0.003125
2021-03-02 13:14:34,880 epoch 67 - iter 220/226 - loss 4.33609443 - samples/sec: 299.13 - lr: 0.003125
2021-03-02 13:14:36,155 ----------------------------------------------------------------------------------------------------
2021-03-02 13:14:36,156 EPOCH 67 done: loss 4.3436 - lr 0.0031250
2021-03-02 13:14:38,681 DEV : loss 2.4913344383239746 - score 0.9664
2021-03-02 13:14:38,759 BAD EPOCHS (no improvement): 2
2021-03-02 13:14:38,761 ----------------------------------------------------------------------------------------------------
2021-03-02 13:14:43,927 epoch 68 - iter 22/226 - loss 4.49483841 - samples/sec: 272.75 - lr: 0.003125
2021-03-02 13:14:48,662 epoch 68 - iter 44/226 - loss 4.45968708 - samples/sec: 297.52 - lr: 0.003125
2021-03-02 13:14:53,558 epoch 68 - iter 66/226 - loss 4.42061170 - samples/sec: 287.81 - lr: 0.003125
2021-03-02 13:14:58,425 epoch 68 - iter 88/226 - loss 4.33403922 - samples/sec: 289.52 - lr: 0.003125
2021-03-02 13:15:03,343 epoch 68 - iter 110/226 - loss 4.33430160 - samples/sec: 286.64 - lr: 0.003125
2021-03-02 13:15:08,036 epoch 68 - iter 132/226 - loss 4.33344324 - samples/sec: 300.42 - lr: 0.003125
2021-03-02 13:15:12,714 epoch 68 - iter 154/226 - loss 4.31723025 - samples/sec: 301.32 - lr: 0.003125
2021-03-02 13:15:17,544 epoch 68 - iter 176/226 - loss 4.32971062 - samples/sec: 291.81 - lr: 0.003125
2021-03-02 13:15:22,251 epoch 68 - iter 198/226 - loss 4.31811722 - samples/sec: 299.40 - lr: 0.003125
2021-03-02 13:15:26,781 epoch 68 - iter 220/226 - loss 4.31711131 - samples/sec: 311.15 - lr: 0.003125
2021-03-02 13:15:28,059 ----------------------------------------------------------------------------------------------------
2021-03-02 13:15:28,059 EPOCH 68 done: loss 4.3056 - lr 0.0031250
2021-03-02 13:15:30,596 DEV : loss 2.493551731109619 - score 0.9666
2021-03-02 13:15:30,671 BAD EPOCHS (no improvement): 3
2021-03-02 13:15:30,674 ----------------------------------------------------------------------------------------------------
2021-03-02 13:15:35,890 epoch 69 - iter 22/226 - loss 4.16718330 - samples/sec: 270.08 - lr: 0.003125
2021-03-02 13:15:40,707 epoch 69 - iter 44/226 - loss 4.19844761 - samples/sec: 292.50 - lr: 0.003125
2021-03-02 13:15:45,603 epoch 69 - iter 66/226 - loss 4.14184868 - samples/sec: 287.75 - lr: 0.003125
2021-03-02 13:15:50,219 epoch 69 - iter 88/226 - loss 4.16330922 - samples/sec: 305.24 - lr: 0.003125
2021-03-02 13:15:55,038 epoch 69 - iter 110/226 - loss 4.21263310 - samples/sec: 292.32 - lr: 0.003125
2021-03-02 13:16:00,045 epoch 69 - iter 132/226 - loss 4.26754699 - samples/sec: 281.45 - lr: 0.003125
2021-03-02 13:16:04,800 epoch 69 - iter 154/226 - loss 4.30130609 - samples/sec: 296.37 - lr: 0.003125
2021-03-02 13:16:09,350 epoch 69 - iter 176/226 - loss 4.28134439 - samples/sec: 309.83 - lr: 0.003125
2021-03-02 13:16:14,168 epoch 69 - iter 198/226 - loss 4.26753044 - samples/sec: 292.56 - lr: 0.003125
2021-03-02 13:16:18,728 epoch 69 - iter 220/226 - loss 4.27297871 - samples/sec: 309.15 - lr: 0.003125
2021-03-02 13:16:19,983 ----------------------------------------------------------------------------------------------------
2021-03-02 13:16:19,984 EPOCH 69 done: loss 4.2727 - lr 0.0031250
2021-03-02 13:16:22,543 DEV : loss 2.491436004638672 - score 0.9664
2021-03-02 13:16:22,621 BAD EPOCHS (no improvement): 4
2021-03-02 13:16:22,623 ----------------------------------------------------------------------------------------------------
2021-03-02 13:16:27,366 epoch 70 - iter 22/226 - loss 4.18295419 - samples/sec: 297.16 - lr: 0.001563
2021-03-02 13:16:31,958 epoch 70 - iter 44/226 - loss 4.19861320 - samples/sec: 306.85 - lr: 0.001563
2021-03-02 13:16:36,599 epoch 70 - iter 66/226 - loss 4.23787712 - samples/sec: 303.68 - lr: 0.001563
2021-03-02 13:16:41,647 epoch 70 - iter 88/226 - loss 4.22543062 - samples/sec: 279.13 - lr: 0.001563
2021-03-02 13:16:46,675 epoch 70 - iter 110/226 - loss 4.23100817 - samples/sec: 280.21 - lr: 0.001563
2021-03-02 13:16:51,373 epoch 70 - iter 132/226 - loss 4.22704873 - samples/sec: 299.95 - lr: 0.001563
2021-03-02 13:16:56,478 epoch 70 - iter 154/226 - loss 4.25983016 - samples/sec: 275.96 - lr: 0.001563
2021-03-02 13:17:01,198 epoch 70 - iter 176/226 - loss 4.28039803 - samples/sec: 298.51 - lr: 0.001563
2021-03-02 13:17:05,880 epoch 70 - iter 198/226 - loss 4.28662405 - samples/sec: 300.97 - lr: 0.001563
2021-03-02 13:17:10,512 epoch 70 - iter 220/226 - loss 4.28769312 - samples/sec: 304.27 - lr: 0.001563
2021-03-02 13:17:11,799 ----------------------------------------------------------------------------------------------------
2021-03-02 13:17:11,799 EPOCH 70 done: loss 4.2978 - lr 0.0015625
2021-03-02 13:17:14,366 DEV : loss 2.489553213119507 - score 0.9665
2021-03-02 13:17:14,442 BAD EPOCHS (no improvement): 1
2021-03-02 13:17:14,444 ----------------------------------------------------------------------------------------------------
2021-03-02 13:17:19,442 epoch 71 - iter 22/226 - loss 4.24423717 - samples/sec: 282.05 - lr: 0.001563
2021-03-02 13:17:24,355 epoch 71 - iter 44/226 - loss 4.18954397 - samples/sec: 286.90 - lr: 0.001563
2021-03-02 13:17:29,015 epoch 71 - iter 66/226 - loss 4.22497857 - samples/sec: 302.47 - lr: 0.001563
2021-03-02 13:17:33,707 epoch 71 - iter 88/226 - loss 4.21947516 - samples/sec: 300.36 - lr: 0.001563
2021-03-02 13:17:38,394 epoch 71 - iter 110/226 - loss 4.22763360 - samples/sec: 300.72 - lr: 0.001563
2021-03-02 13:17:43,120 epoch 71 - iter 132/226 - loss 4.22522394 - samples/sec: 298.15 - lr: 0.001563
2021-03-02 13:17:47,972 epoch 71 - iter 154/226 - loss 4.25259125 - samples/sec: 290.44 - lr: 0.001563
2021-03-02 13:17:52,687 epoch 71 - iter 176/226 - loss 4.25220990 - samples/sec: 298.86 - lr: 0.001563
2021-03-02 13:17:57,480 epoch 71 - iter 198/226 - loss 4.26962373 - samples/sec: 293.94 - lr: 0.001563
2021-03-02 13:18:02,761 epoch 71 - iter 220/226 - loss 4.28880819 - samples/sec: 266.82 - lr: 0.001563
2021-03-02 13:18:03,934 ----------------------------------------------------------------------------------------------------
2021-03-02 13:18:03,935 EPOCH 71 done: loss 4.2943 - lr 0.0015625
2021-03-02 13:18:06,462 DEV : loss 2.491175651550293 - score 0.9665
2021-03-02 13:18:06,538 BAD EPOCHS (no improvement): 2
2021-03-02 13:18:06,540 ----------------------------------------------------------------------------------------------------
2021-03-02 13:18:11,456 epoch 72 - iter 22/226 - loss 4.35860097 - samples/sec: 286.64 - lr: 0.001563
2021-03-02 13:18:16,076 epoch 72 - iter 44/226 - loss 4.31160229 - samples/sec: 305.06 - lr: 0.001563
2021-03-02 13:18:20,707 epoch 72 - iter 66/226 - loss 4.28834318 - samples/sec: 304.29 - lr: 0.001563
2021-03-02 13:18:26,005 epoch 72 - iter 88/226 - loss 4.36116729 - samples/sec: 266.01 - lr: 0.001563
2021-03-02 13:18:30,779 epoch 72 - iter 110/226 - loss 4.33143504 - samples/sec: 295.26 - lr: 0.001563
2021-03-02 13:18:35,532 epoch 72 - iter 132/226 - loss 4.36724747 - samples/sec: 296.57 - lr: 0.001563
2021-03-02 13:18:40,305 epoch 72 - iter 154/226 - loss 4.34903653 - samples/sec: 295.34 - lr: 0.001563
2021-03-02 13:18:44,987 epoch 72 - iter 176/226 - loss 4.33688571 - samples/sec: 300.98 - lr: 0.001563
2021-03-02 13:18:49,883 epoch 72 - iter 198/226 - loss 4.33259911 - samples/sec: 287.82 - lr: 0.001563
2021-03-02 13:18:54,527 epoch 72 - iter 220/226 - loss 4.31375696 - samples/sec: 303.51 - lr: 0.001563
2021-03-02 13:18:55,787 ----------------------------------------------------------------------------------------------------
2021-03-02 13:18:55,787 EPOCH 72 done: loss 4.3190 - lr 0.0015625
2021-03-02 13:18:58,315 DEV : loss 2.4857826232910156 - score 0.9667
2021-03-02 13:18:58,391 BAD EPOCHS (no improvement): 3
2021-03-02 13:18:58,393 ----------------------------------------------------------------------------------------------------
2021-03-02 13:19:03,440 epoch 73 - iter 22/226 - loss 4.35148811 - samples/sec: 279.22 - lr: 0.001563
2021-03-02 13:19:08,242 epoch 73 - iter 44/226 - loss 4.42674817 - samples/sec: 293.39 - lr: 0.001563
2021-03-02 13:19:13,025 epoch 73 - iter 66/226 - loss 4.35383603 - samples/sec: 294.58 - lr: 0.001563
2021-03-02 13:19:18,207 epoch 73 - iter 88/226 - loss 4.34294241 - samples/sec: 271.85 - lr: 0.001563
2021-03-02 13:19:23,083 epoch 73 - iter 110/226 - loss 4.34770454 - samples/sec: 288.99 - lr: 0.001563
2021-03-02 13:19:27,513 epoch 73 - iter 132/226 - loss 4.34357672 - samples/sec: 318.12 - lr: 0.001563
2021-03-02 13:19:32,023 epoch 73 - iter 154/226 - loss 4.32390474 - samples/sec: 312.53 - lr: 0.001563
2021-03-02 13:19:36,855 epoch 73 - iter 176/226 - loss 4.31826842 - samples/sec: 291.67 - lr: 0.001563
2021-03-02 13:19:41,842 epoch 73 - iter 198/226 - loss 4.33197009 - samples/sec: 282.61 - lr: 0.001563
2021-03-02 13:19:46,449 epoch 73 - iter 220/226 - loss 4.33008621 - samples/sec: 305.95 - lr: 0.001563
2021-03-02 13:19:47,621 ----------------------------------------------------------------------------------------------------
2021-03-02 13:19:47,622 EPOCH 73 done: loss 4.3249 - lr 0.0015625
2021-03-02 13:19:50,184 DEV : loss 2.4887161254882812 - score 0.9668
2021-03-02 13:19:50,260 BAD EPOCHS (no improvement): 4
2021-03-02 13:19:50,262 ----------------------------------------------------------------------------------------------------
2021-03-02 13:19:54,990 epoch 74 - iter 22/226 - loss 4.32287857 - samples/sec: 298.10 - lr: 0.000781
2021-03-02 13:19:59,770 epoch 74 - iter 44/226 - loss 4.25762545 - samples/sec: 294.82 - lr: 0.000781
2021-03-02 13:20:04,364 epoch 74 - iter 66/226 - loss 4.29113617 - samples/sec: 306.72 - lr: 0.000781
2021-03-02 13:20:09,045 epoch 74 - iter 88/226 - loss 4.29988960 - samples/sec: 301.07 - lr: 0.000781
2021-03-02 13:20:13,694 epoch 74 - iter 110/226 - loss 4.32245228 - samples/sec: 303.06 - lr: 0.000781
2021-03-02 13:20:18,363 epoch 74 - iter 132/226 - loss 4.34273356 - samples/sec: 301.86 - lr: 0.000781
2021-03-02 13:20:22,927 epoch 74 - iter 154/226 - loss 4.36042592 - samples/sec: 308.71 - lr: 0.000781
2021-03-02 13:20:27,706 epoch 74 - iter 176/226 - loss 4.31875739 - samples/sec: 294.80 - lr: 0.000781
2021-03-02 13:20:32,686 epoch 74 - iter 198/226 - loss 4.30989899 - samples/sec: 282.96 - lr: 0.000781
2021-03-02 13:20:37,911 epoch 74 - iter 220/226 - loss 4.32490782 - samples/sec: 269.68 - lr: 0.000781
2021-03-02 13:20:39,250 ----------------------------------------------------------------------------------------------------
2021-03-02 13:20:39,251 EPOCH 74 done: loss 4.3259 - lr 0.0007813
2021-03-02 13:20:41,810 DEV : loss 2.487290859222412 - score 0.9668
2021-03-02 13:20:41,886 BAD EPOCHS (no improvement): 1
2021-03-02 13:20:41,888 ----------------------------------------------------------------------------------------------------
2021-03-02 13:20:47,058 epoch 75 - iter 22/226 - loss 4.50323968 - samples/sec: 272.62 - lr: 0.000781
2021-03-02 13:20:51,687 epoch 75 - iter 44/226 - loss 4.37956695 - samples/sec: 304.50 - lr: 0.000781
2021-03-02 13:20:56,226 epoch 75 - iter 66/226 - loss 4.30477909 - samples/sec: 310.55 - lr: 0.000781
2021-03-02 13:21:00,965 epoch 75 - iter 88/226 - loss 4.27505691 - samples/sec: 297.36 - lr: 0.000781
2021-03-02 13:21:05,539 epoch 75 - iter 110/226 - loss 4.29030255 - samples/sec: 308.10 - lr: 0.000781
2021-03-02 13:21:10,132 epoch 75 - iter 132/226 - loss 4.29887165 - samples/sec: 306.84 - lr: 0.000781
2021-03-02 13:21:15,396 epoch 75 - iter 154/226 - loss 4.30225699 - samples/sec: 267.66 - lr: 0.000781
2021-03-02 13:21:20,026 epoch 75 - iter 176/226 - loss 4.29540141 - samples/sec: 304.35 - lr: 0.000781
2021-03-02 13:21:24,762 epoch 75 - iter 198/226 - loss 4.28508268 - samples/sec: 297.52 - lr: 0.000781
2021-03-02 13:21:29,578 epoch 75 - iter 220/226 - loss 4.30162698 - samples/sec: 292.54 - lr: 0.000781
2021-03-02 13:21:30,905 ----------------------------------------------------------------------------------------------------
2021-03-02 13:21:30,905 EPOCH 75 done: loss 4.3056 - lr 0.0007813
2021-03-02 13:21:33,427 DEV : loss 2.4870996475219727 - score 0.9668
2021-03-02 13:21:33,505 BAD EPOCHS (no improvement): 2
2021-03-02 13:21:33,507 ----------------------------------------------------------------------------------------------------
2021-03-02 13:21:38,207 epoch 76 - iter 22/226 - loss 4.37418475 - samples/sec: 299.77 - lr: 0.000781
2021-03-02 13:21:42,885 epoch 76 - iter 44/226 - loss 4.39546596 - samples/sec: 301.23 - lr: 0.000781
2021-03-02 13:21:47,534 epoch 76 - iter 66/226 - loss 4.41299921 - samples/sec: 303.13 - lr: 0.000781
2021-03-02 13:21:52,248 epoch 76 - iter 88/226 - loss 4.39748039 - samples/sec: 299.00 - lr: 0.000781
2021-03-02 13:21:57,053 epoch 76 - iter 110/226 - loss 4.42552429 - samples/sec: 293.35 - lr: 0.000781
2021-03-02 13:22:01,739 epoch 76 - iter 132/226 - loss 4.40237816 - samples/sec: 300.84 - lr: 0.000781
2021-03-02 13:22:06,584 epoch 76 - iter 154/226 - loss 4.37508377 - samples/sec: 290.91 - lr: 0.000781
2021-03-02 13:22:11,544 epoch 76 - iter 176/226 - loss 4.34854153 - samples/sec: 284.21 - lr: 0.000781
2021-03-02 13:22:16,834 epoch 76 - iter 198/226 - loss 4.35251858 - samples/sec: 266.35 - lr: 0.000781
2021-03-02 13:22:21,523 epoch 76 - iter 220/226 - loss 4.33072009 - samples/sec: 300.58 - lr: 0.000781
2021-03-02 13:22:22,882 ----------------------------------------------------------------------------------------------------
2021-03-02 13:22:22,883 EPOCH 76 done: loss 4.3325 - lr 0.0007813
2021-03-02 13:22:25,416 DEV : loss 2.488741397857666 - score 0.9666
2021-03-02 13:22:25,492 BAD EPOCHS (no improvement): 3
2021-03-02 13:22:25,494 ----------------------------------------------------------------------------------------------------
2021-03-02 13:22:30,472 epoch 77 - iter 22/226 - loss 4.31774791 - samples/sec: 283.06 - lr: 0.000781
2021-03-02 13:22:35,213 epoch 77 - iter 44/226 - loss 4.27422399 - samples/sec: 297.15 - lr: 0.000781
2021-03-02 13:22:40,147 epoch 77 - iter 66/226 - loss 4.31293736 - samples/sec: 285.58 - lr: 0.000781
2021-03-02 13:22:44,815 epoch 77 - iter 88/226 - loss 4.25217537 - samples/sec: 301.88 - lr: 0.000781
2021-03-02 13:22:49,661 epoch 77 - iter 110/226 - loss 4.28297441 - samples/sec: 290.80 - lr: 0.000781
2021-03-02 13:22:54,906 epoch 77 - iter 132/226 - loss 4.28642819 - samples/sec: 268.66 - lr: 0.000781
2021-03-02 13:22:59,608 epoch 77 - iter 154/226 - loss 4.27728049 - samples/sec: 299.80 - lr: 0.000781
2021-03-02 13:23:04,491 epoch 77 - iter 176/226 - loss 4.28222191 - samples/sec: 288.60 - lr: 0.000781
2021-03-02 13:23:09,253 epoch 77 - iter 198/226 - loss 4.28962789 - samples/sec: 296.06 - lr: 0.000781
2021-03-02 13:23:13,941 epoch 77 - iter 220/226 - loss 4.29445159 - samples/sec: 300.63 - lr: 0.000781
2021-03-02 13:23:15,173 ----------------------------------------------------------------------------------------------------
2021-03-02 13:23:15,174 EPOCH 77 done: loss 4.2891 - lr 0.0007813
2021-03-02 13:23:17,732 DEV : loss 2.485323429107666 - score 0.9668
2021-03-02 13:23:17,807 BAD EPOCHS (no improvement): 4
2021-03-02 13:23:17,809 ----------------------------------------------------------------------------------------------------
2021-03-02 13:23:22,488 epoch 78 - iter 22/226 - loss 4.50148922 - samples/sec: 301.23 - lr: 0.000391
2021-03-02 13:23:30,162 epoch 78 - iter 44/226 - loss 4.38568244 - samples/sec: 183.59 - lr: 0.000391
2021-03-02 13:23:35,200 epoch 78 - iter 66/226 - loss 4.38760542 - samples/sec: 279.62 - lr: 0.000391
2021-03-02 13:23:39,790 epoch 78 - iter 88/226 - loss 4.30386083 - samples/sec: 306.93 - lr: 0.000391
2021-03-02 13:23:44,749 epoch 78 - iter 110/226 - loss 4.31430260 - samples/sec: 284.15 - lr: 0.000391
2021-03-02 13:23:49,585 epoch 78 - iter 132/226 - loss 4.32636283 - samples/sec: 291.37 - lr: 0.000391
2021-03-02 13:23:54,297 epoch 78 - iter 154/226 - loss 4.33055210 - samples/sec: 298.99 - lr: 0.000391
2021-03-02 13:23:59,116 epoch 78 - iter 176/226 - loss 4.29225156 - samples/sec: 292.42 - lr: 0.000391
2021-03-02 13:24:04,433 epoch 78 - iter 198/226 - loss 4.31907943 - samples/sec: 265.07 - lr: 0.000391
2021-03-02 13:24:09,118 epoch 78 - iter 220/226 - loss 4.31112100 - samples/sec: 300.93 - lr: 0.000391
2021-03-02 13:24:10,394 ----------------------------------------------------------------------------------------------------
2021-03-02 13:24:10,395 EPOCH 78 done: loss 4.3138 - lr 0.0003906
2021-03-02 13:24:12,955 DEV : loss 2.488582134246826 - score 0.9666
2021-03-02 13:24:13,031 BAD EPOCHS (no improvement): 1
2021-03-02 13:24:13,033 ----------------------------------------------------------------------------------------------------
2021-03-02 13:24:17,776 epoch 79 - iter 22/226 - loss 4.26065703 - samples/sec: 297.15 - lr: 0.000391
2021-03-02 13:24:22,777 epoch 79 - iter 44/226 - loss 4.15566696 - samples/sec: 281.84 - lr: 0.000391
2021-03-02 13:24:27,453 epoch 79 - iter 66/226 - loss 4.21219309 - samples/sec: 301.37 - lr: 0.000391
2021-03-02 13:24:32,117 epoch 79 - iter 88/226 - loss 4.21736598 - samples/sec: 302.15 - lr: 0.000391
2021-03-02 13:24:36,934 epoch 79 - iter 110/226 - loss 4.22537466 - samples/sec: 292.59 - lr: 0.000391
2021-03-02 13:24:41,603 epoch 79 - iter 132/226 - loss 4.23492841 - samples/sec: 301.84 - lr: 0.000391
2021-03-02 13:24:46,915 epoch 79 - iter 154/226 - loss 4.26808377 - samples/sec: 265.26 - lr: 0.000391
2021-03-02 13:24:51,809 epoch 79 - iter 176/226 - loss 4.29945429 - samples/sec: 287.90 - lr: 0.000391
2021-03-02 13:24:56,644 epoch 79 - iter 198/226 - loss 4.30232818 - samples/sec: 291.44 - lr: 0.000391
2021-03-02 13:25:01,517 epoch 79 - iter 220/226 - loss 4.29774692 - samples/sec: 289.11 - lr: 0.000391
2021-03-02 13:25:02,745 ----------------------------------------------------------------------------------------------------
2021-03-02 13:25:02,745 EPOCH 79 done: loss 4.2918 - lr 0.0003906
2021-03-02 13:25:05,301 DEV : loss 2.489116668701172 - score 0.9666
2021-03-02 13:25:05,383 BAD EPOCHS (no improvement): 2
2021-03-02 13:25:05,385 ----------------------------------------------------------------------------------------------------
2021-03-02 13:25:10,579 epoch 80 - iter 22/226 - loss 4.41724111 - samples/sec: 271.35 - lr: 0.000391
2021-03-02 13:25:15,469 epoch 80 - iter 44/226 - loss 4.54239089 - samples/sec: 288.23 - lr: 0.000391
2021-03-02 13:25:20,130 epoch 80 - iter 66/226 - loss 4.38471864 - samples/sec: 302.37 - lr: 0.000391
2021-03-02 13:25:25,006 epoch 80 - iter 88/226 - loss 4.31540930 - samples/sec: 289.12 - lr: 0.000391
2021-03-02 13:25:29,801 epoch 80 - iter 110/226 - loss 4.29716722 - samples/sec: 293.96 - lr: 0.000391
2021-03-02 13:25:34,705 epoch 80 - iter 132/226 - loss 4.27174449 - samples/sec: 287.42 - lr: 0.000391
2021-03-02 13:25:39,399 epoch 80 - iter 154/226 - loss 4.27477719 - samples/sec: 300.26 - lr: 0.000391
2021-03-02 13:25:44,324 epoch 80 - iter 176/226 - loss 4.30132451 - samples/sec: 286.14 - lr: 0.000391
2021-03-02 13:25:48,977 epoch 80 - iter 198/226 - loss 4.28334979 - samples/sec: 302.84 - lr: 0.000391
2021-03-02 13:25:53,678 epoch 80 - iter 220/226 - loss 4.30750553 - samples/sec: 299.78 - lr: 0.000391
2021-03-02 13:25:54,983 ----------------------------------------------------------------------------------------------------
2021-03-02 13:25:54,983 EPOCH 80 done: loss 4.3032 - lr 0.0003906
2021-03-02 13:25:57,493 DEV : loss 2.4872920513153076 - score 0.9667
2021-03-02 13:25:57,580 BAD EPOCHS (no improvement): 3
2021-03-02 13:25:57,583 ----------------------------------------------------------------------------------------------------
2021-03-02 13:26:02,335 epoch 81 - iter 22/226 - loss 4.33592282 - samples/sec: 296.53 - lr: 0.000391
2021-03-02 13:26:07,195 epoch 81 - iter 44/226 - loss 4.37611949 - samples/sec: 289.87 - lr: 0.000391
2021-03-02 13:26:11,906 epoch 81 - iter 66/226 - loss 4.38630977 - samples/sec: 299.07 - lr: 0.000391
2021-03-02 13:26:16,560 epoch 81 - iter 88/226 - loss 4.35786818 - samples/sec: 302.81 - lr: 0.000391
2021-03-02 13:26:21,448 epoch 81 - iter 110/226 - loss 4.35659745 - samples/sec: 288.31 - lr: 0.000391
2021-03-02 13:26:26,182 epoch 81 - iter 132/226 - loss 4.30194785 - samples/sec: 297.77 - lr: 0.000391
2021-03-02 13:26:30,935 epoch 81 - iter 154/226 - loss 4.28811583 - samples/sec: 296.58 - lr: 0.000391
2021-03-02 13:26:35,829 epoch 81 - iter 176/226 - loss 4.27502600 - samples/sec: 288.03 - lr: 0.000391
2021-03-02 13:26:40,740 epoch 81 - iter 198/226 - loss 4.27175879 - samples/sec: 287.03 - lr: 0.000391
2021-03-02 13:26:46,043 epoch 81 - iter 220/226 - loss 4.28588193 - samples/sec: 265.70 - lr: 0.000391
2021-03-02 13:26:47,297 ----------------------------------------------------------------------------------------------------
2021-03-02 13:26:47,298 EPOCH 81 done: loss 4.2931 - lr 0.0003906
2021-03-02 13:26:49,840 DEV : loss 2.4866726398468018 - score 0.9667
2021-03-02 13:26:49,916 BAD EPOCHS (no improvement): 4
2021-03-02 13:26:49,918 ----------------------------------------------------------------------------------------------------
2021-03-02 13:26:54,870 epoch 82 - iter 22/226 - loss 4.20530889 - samples/sec: 284.59 - lr: 0.000195
2021-03-02 13:26:59,809 epoch 82 - iter 44/226 - loss 4.37505933 - samples/sec: 285.24 - lr: 0.000195
2021-03-02 13:27:04,480 epoch 82 - iter 66/226 - loss 4.28321471 - samples/sec: 301.63 - lr: 0.000195
2021-03-02 13:27:09,413 epoch 82 - iter 88/226 - loss 4.24403245 - samples/sec: 285.61 - lr: 0.000195
2021-03-02 13:27:14,415 epoch 82 - iter 110/226 - loss 4.26494537 - samples/sec: 281.66 - lr: 0.000195
2021-03-02 13:27:19,281 epoch 82 - iter 132/226 - loss 4.29876563 - samples/sec: 289.61 - lr: 0.000195
2021-03-02 13:27:23,931 epoch 82 - iter 154/226 - loss 4.28392571 - samples/sec: 303.00 - lr: 0.000195
2021-03-02 13:27:28,723 epoch 82 - iter 176/226 - loss 4.28927607 - samples/sec: 294.10 - lr: 0.000195
2021-03-02 13:27:33,479 epoch 82 - iter 198/226 - loss 4.29717333 - samples/sec: 296.38 - lr: 0.000195
2021-03-02 13:27:38,521 epoch 82 - iter 220/226 - loss 4.30220513 - samples/sec: 279.59 - lr: 0.000195
2021-03-02 13:27:39,768 ----------------------------------------------------------------------------------------------------
2021-03-02 13:27:39,769 EPOCH 82 done: loss 4.3098 - lr 0.0001953
2021-03-02 13:27:42,338 DEV : loss 2.4870944023132324 - score 0.9667
2021-03-02 13:27:42,415 BAD EPOCHS (no improvement): 1
2021-03-02 13:27:42,417 ----------------------------------------------------------------------------------------------------
2021-03-02 13:27:47,304 epoch 83 - iter 22/226 - loss 4.49141608 - samples/sec: 288.39 - lr: 0.000195
2021-03-02 13:27:52,325 epoch 83 - iter 44/226 - loss 4.50347602 - samples/sec: 280.67 - lr: 0.000195
2021-03-02 13:27:57,132 epoch 83 - iter 66/226 - loss 4.45407822 - samples/sec: 293.20 - lr: 0.000195
2021-03-02 13:28:02,388 epoch 83 - iter 88/226 - loss 4.41418025 - samples/sec: 268.07 - lr: 0.000195
2021-03-02 13:28:07,050 epoch 83 - iter 110/226 - loss 4.40449502 - samples/sec: 302.29 - lr: 0.000195
2021-03-02 13:28:11,921 epoch 83 - iter 132/226 - loss 4.39698159 - samples/sec: 289.24 - lr: 0.000195
2021-03-02 13:28:16,770 epoch 83 - iter 154/226 - loss 4.40053011 - samples/sec: 290.61 - lr: 0.000195
2021-03-02 13:28:21,428 epoch 83 - iter 176/226 - loss 4.37908849 - samples/sec: 302.49 - lr: 0.000195
2021-03-02 13:28:26,107 epoch 83 - iter 198/226 - loss 4.35449270 - samples/sec: 301.17 - lr: 0.000195
2021-03-02 13:28:30,931 epoch 83 - iter 220/226 - loss 4.34118999 - samples/sec: 292.10 - lr: 0.000195
2021-03-02 13:28:32,363 ----------------------------------------------------------------------------------------------------
2021-03-02 13:28:32,364 EPOCH 83 done: loss 4.3371 - lr 0.0001953
2021-03-02 13:28:34,941 DEV : loss 2.486717939376831 - score 0.9667
2021-03-02 13:28:35,017 BAD EPOCHS (no improvement): 2
2021-03-02 13:28:35,019 ----------------------------------------------------------------------------------------------------
2021-03-02 13:28:40,050 epoch 84 - iter 22/226 - loss 4.53630028 - samples/sec: 280.21 - lr: 0.000195
2021-03-02 13:28:44,797 epoch 84 - iter 44/226 - loss 4.34582099 - samples/sec: 296.95 - lr: 0.000195
2021-03-02 13:28:49,479 epoch 84 - iter 66/226 - loss 4.36626539 - samples/sec: 301.07 - lr: 0.000195
2021-03-02 13:28:54,220 epoch 84 - iter 88/226 - loss 4.35209656 - samples/sec: 297.27 - lr: 0.000195
2021-03-02 13:28:59,135 epoch 84 - iter 110/226 - loss 4.36782808 - samples/sec: 286.75 - lr: 0.000195
2021-03-02 13:29:03,975 epoch 84 - iter 132/226 - loss 4.33494527 - samples/sec: 291.16 - lr: 0.000195
2021-03-02 13:29:09,271 epoch 84 - iter 154/226 - loss 4.31337755 - samples/sec: 266.05 - lr: 0.000195
2021-03-02 13:29:14,301 epoch 84 - iter 176/226 - loss 4.31934392 - samples/sec: 280.14 - lr: 0.000195
2021-03-02 13:29:19,217 epoch 84 - iter 198/226 - loss 4.33994204 - samples/sec: 286.65 - lr: 0.000195
2021-03-02 13:29:23,932 epoch 84 - iter 220/226 - loss 4.33385026 - samples/sec: 298.83 - lr: 0.000195
2021-03-02 13:29:25,197 ----------------------------------------------------------------------------------------------------
2021-03-02 13:29:25,198 EPOCH 84 done: loss 4.3304 - lr 0.0001953
2021-03-02 13:29:27,730 DEV : loss 2.486844539642334 - score 0.9667
2021-03-02 13:29:27,806 BAD EPOCHS (no improvement): 3
2021-03-02 13:29:27,808 ----------------------------------------------------------------------------------------------------
2021-03-02 13:29:32,491 epoch 85 - iter 22/226 - loss 4.31742632 - samples/sec: 300.87 - lr: 0.000195
2021-03-02 13:29:37,087 epoch 85 - iter 44/226 - loss 4.26423753 - samples/sec: 306.63 - lr: 0.000195
2021-03-02 13:29:41,878 epoch 85 - iter 66/226 - loss 4.27575780 - samples/sec: 294.14 - lr: 0.000195
2021-03-02 13:29:46,830 epoch 85 - iter 88/226 - loss 4.31225107 - samples/sec: 284.65 - lr: 0.000195
2021-03-02 13:29:51,681 epoch 85 - iter 110/226 - loss 4.29813769 - samples/sec: 290.55 - lr: 0.000195
2021-03-02 13:29:56,482 epoch 85 - iter 132/226 - loss 4.36487476 - samples/sec: 293.52 - lr: 0.000195
2021-03-02 13:30:01,667 epoch 85 - iter 154/226 - loss 4.35312672 - samples/sec: 271.81 - lr: 0.000195
2021-03-02 13:30:06,445 epoch 85 - iter 176/226 - loss 4.34220209 - samples/sec: 294.91 - lr: 0.000195
2021-03-02 13:30:11,240 epoch 85 - iter 198/226 - loss 4.33740348 - samples/sec: 293.97 - lr: 0.000195
2021-03-02 13:30:15,958 epoch 85 - iter 220/226 - loss 4.34761800 - samples/sec: 298.66 - lr: 0.000195
2021-03-02 13:30:17,175 ----------------------------------------------------------------------------------------------------
2021-03-02 13:30:17,176 EPOCH 85 done: loss 4.3385 - lr 0.0001953
2021-03-02 13:30:19,699 DEV : loss 2.487819194793701 - score 0.9666
2021-03-02 13:30:19,775 BAD EPOCHS (no improvement): 4
2021-03-02 13:30:19,777 ----------------------------------------------------------------------------------------------------
2021-03-02 13:30:19,777 ----------------------------------------------------------------------------------------------------
2021-03-02 13:30:19,777 learning rate too small - quitting training!
2021-03-02 13:30:19,777 ----------------------------------------------------------------------------------------------------
2021-03-02 13:30:40,083 ----------------------------------------------------------------------------------------------------
2021-03-02 13:30:40,083 Testing using best model ...
2021-03-02 13:30:40,085 loading file resources/taggers/example-pos/best-model.pt
2021-03-02 13:30:49,790 0.9615
2021-03-02 13:30:49,790
Results:
- F-score (micro): 0.9615
- F-score (macro): 0.8973
- Accuracy (incl. no class): 0.9615
By class:
precision recall f1-score support
PRON 0.9689 0.9653 0.9671 548
VERB 0.9545 0.9661 0.9602 825
SCONJ 0.9421 0.8906 0.9157 128
ADP 0.9767 0.9872 0.9819 1483
CCONJ 0.9919 1.0000 0.9959 245
DET 0.9807 0.9939 0.9873 1484
NOUN 0.9496 0.9547 0.9522 1856
ADJ 0.9238 0.9208 0.9223 606
AUX 0.9566 0.9944 0.9751 355
ADV 0.9593 0.9365 0.9478 504
PUNCT 0.9915 0.9815 0.9864 1186
PROPN 0.8969 0.8529 0.8744 510
NUM 0.9457 0.9289 0.9372 225
SYM 0.9706 0.8684 0.9167 38
PART 1.0000 1.0000 1.0000 4
X 0.5556 0.3125 0.4000 16
INTJ 0.4444 0.6667 0.5333 6
accuracy 0.9615 10019
macro avg 0.9064 0.8953 0.8973 10019
weighted avg 0.9612 0.9615 0.9612 10019
2021-03-02 13:30:49,790 ----------------------------------------------------------------------------------------------------