2023-10-14 06:00:02,910 ---------------------------------------------------------------------------------------------------- 2023-10-14 06:00:02,912 Model: "SequenceTagger( (embeddings): ByT5Embeddings( (model): T5EncoderModel( (shared): Embedding(384, 1472) (encoder): T5Stack( (embed_tokens): Embedding(384, 1472) (block): ModuleList( (0): T5Block( (layer): ModuleList( (0): T5LayerSelfAttention( (SelfAttention): T5Attention( (q): Linear(in_features=1472, out_features=384, bias=False) (k): Linear(in_features=1472, out_features=384, bias=False) (v): Linear(in_features=1472, out_features=384, bias=False) (o): Linear(in_features=384, out_features=1472, bias=False) (relative_attention_bias): Embedding(32, 6) ) (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) (1): T5LayerFF( (DenseReluDense): T5DenseGatedActDense( (wi_0): Linear(in_features=1472, out_features=3584, bias=False) (wi_1): Linear(in_features=1472, out_features=3584, bias=False) (wo): Linear(in_features=3584, out_features=1472, bias=False) (dropout): Dropout(p=0.1, inplace=False) (act): NewGELUActivation() ) (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) (1-11): 11 x T5Block( (layer): ModuleList( (0): T5LayerSelfAttention( (SelfAttention): T5Attention( (q): Linear(in_features=1472, out_features=384, bias=False) (k): Linear(in_features=1472, out_features=384, bias=False) (v): Linear(in_features=1472, out_features=384, bias=False) (o): Linear(in_features=384, out_features=1472, bias=False) ) (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) (1): T5LayerFF( (DenseReluDense): T5DenseGatedActDense( (wi_0): Linear(in_features=1472, out_features=3584, bias=False) (wi_1): Linear(in_features=1472, out_features=3584, bias=False) (wo): Linear(in_features=3584, out_features=1472, bias=False) (dropout): Dropout(p=0.1, inplace=False) (act): NewGELUActivation() ) (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) ) (final_layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) (locked_dropout): LockedDropout(p=0.5) (linear): Linear(in_features=1472, out_features=13, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-14 06:00:02,912 ---------------------------------------------------------------------------------------------------- 2023-10-14 06:00:02,912 MultiCorpus: 14465 train + 1392 dev + 2432 test sentences - NER_HIPE_2022 Corpus: 14465 train + 1392 dev + 2432 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/letemps/fr/with_doc_seperator 2023-10-14 06:00:02,913 ---------------------------------------------------------------------------------------------------- 2023-10-14 06:00:02,913 Train: 14465 sentences 2023-10-14 06:00:02,913 (train_with_dev=False, train_with_test=False) 2023-10-14 06:00:02,913 ---------------------------------------------------------------------------------------------------- 2023-10-14 06:00:02,913 Training Params: 2023-10-14 06:00:02,913 - learning_rate: "0.00016" 2023-10-14 06:00:02,913 - mini_batch_size: "4" 2023-10-14 06:00:02,913 - max_epochs: "10" 2023-10-14 06:00:02,913 - shuffle: "True" 2023-10-14 06:00:02,913 ---------------------------------------------------------------------------------------------------- 2023-10-14 06:00:02,913 Plugins: 2023-10-14 06:00:02,913 - TensorboardLogger 2023-10-14 06:00:02,913 - LinearScheduler | warmup_fraction: '0.1' 2023-10-14 06:00:02,913 ---------------------------------------------------------------------------------------------------- 2023-10-14 06:00:02,913 Final evaluation on model from best epoch (best-model.pt) 2023-10-14 06:00:02,914 - metric: "('micro avg', 'f1-score')" 2023-10-14 06:00:02,914 ---------------------------------------------------------------------------------------------------- 2023-10-14 06:00:02,914 Computation: 2023-10-14 06:00:02,914 - compute on device: cuda:0 2023-10-14 06:00:02,914 - embedding storage: none 2023-10-14 06:00:02,914 ---------------------------------------------------------------------------------------------------- 2023-10-14 06:00:02,914 Model training base path: "hmbench-letemps/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-3" 2023-10-14 06:00:02,914 ---------------------------------------------------------------------------------------------------- 2023-10-14 06:00:02,914 ---------------------------------------------------------------------------------------------------- 2023-10-14 06:00:02,914 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-14 06:01:53,445 epoch 1 - iter 361/3617 - loss 2.47239461 - time (sec): 110.53 - samples/sec: 341.23 - lr: 0.000016 - momentum: 0.000000 2023-10-14 06:03:37,266 epoch 1 - iter 722/3617 - loss 2.07484525 - time (sec): 214.35 - samples/sec: 347.51 - lr: 0.000032 - momentum: 0.000000 2023-10-14 06:05:25,472 epoch 1 - iter 1083/3617 - loss 1.61456832 - time (sec): 322.56 - samples/sec: 350.12 - lr: 0.000048 - momentum: 0.000000 2023-10-14 06:07:08,611 epoch 1 - iter 1444/3617 - loss 1.27675526 - time (sec): 425.69 - samples/sec: 355.51 - lr: 0.000064 - momentum: 0.000000 2023-10-14 06:08:50,479 epoch 1 - iter 1805/3617 - loss 1.05993658 - time (sec): 527.56 - samples/sec: 358.18 - lr: 0.000080 - momentum: 0.000000 2023-10-14 06:10:29,696 epoch 1 - iter 2166/3617 - loss 0.91231426 - time (sec): 626.78 - samples/sec: 361.53 - lr: 0.000096 - momentum: 0.000000 2023-10-14 06:12:09,104 epoch 1 - iter 2527/3617 - loss 0.80617937 - time (sec): 726.19 - samples/sec: 362.44 - lr: 0.000112 - momentum: 0.000000 2023-10-14 06:13:53,682 epoch 1 - iter 2888/3617 - loss 0.72280386 - time (sec): 830.77 - samples/sec: 362.49 - lr: 0.000128 - momentum: 0.000000 2023-10-14 06:15:37,793 epoch 1 - iter 3249/3617 - loss 0.65137338 - time (sec): 934.88 - samples/sec: 364.42 - lr: 0.000144 - momentum: 0.000000 2023-10-14 06:17:19,931 epoch 1 - iter 3610/3617 - loss 0.59733239 - time (sec): 1037.01 - samples/sec: 365.74 - lr: 0.000160 - momentum: 0.000000 2023-10-14 06:17:21,610 ---------------------------------------------------------------------------------------------------- 2023-10-14 06:17:21,611 EPOCH 1 done: loss 0.5965 - lr: 0.000160 2023-10-14 06:18:02,444 DEV : loss 0.12650439143180847 - f1-score (micro avg) 0.6067 2023-10-14 06:18:02,513 saving best model 2023-10-14 06:18:03,590 ---------------------------------------------------------------------------------------------------- 2023-10-14 06:19:58,085 epoch 2 - iter 361/3617 - loss 0.09250931 - time (sec): 114.49 - samples/sec: 331.01 - lr: 0.000158 - momentum: 0.000000 2023-10-14 06:21:42,643 epoch 2 - iter 722/3617 - loss 0.09407297 - time (sec): 219.05 - samples/sec: 342.51 - lr: 0.000156 - momentum: 0.000000 2023-10-14 06:23:32,345 epoch 2 - iter 1083/3617 - loss 0.09368204 - time (sec): 328.75 - samples/sec: 353.59 - lr: 0.000155 - momentum: 0.000000 2023-10-14 06:25:14,764 epoch 2 - iter 1444/3617 - loss 0.09240527 - time (sec): 431.17 - samples/sec: 359.43 - lr: 0.000153 - momentum: 0.000000 2023-10-14 06:26:54,551 epoch 2 - iter 1805/3617 - loss 0.09322341 - time (sec): 530.96 - samples/sec: 363.88 - lr: 0.000151 - momentum: 0.000000 2023-10-14 06:28:32,658 epoch 2 - iter 2166/3617 - loss 0.09164435 - time (sec): 629.06 - samples/sec: 365.00 - lr: 0.000149 - momentum: 0.000000 2023-10-14 06:30:13,945 epoch 2 - iter 2527/3617 - loss 0.09049817 - time (sec): 730.35 - samples/sec: 365.65 - lr: 0.000148 - momentum: 0.000000 2023-10-14 06:31:53,651 epoch 2 - iter 2888/3617 - loss 0.09055334 - time (sec): 830.06 - samples/sec: 366.60 - lr: 0.000146 - momentum: 0.000000 2023-10-14 06:33:36,995 epoch 2 - iter 3249/3617 - loss 0.09003575 - time (sec): 933.40 - samples/sec: 366.95 - lr: 0.000144 - momentum: 0.000000 2023-10-14 06:35:21,055 epoch 2 - iter 3610/3617 - loss 0.08946027 - time (sec): 1037.46 - samples/sec: 365.54 - lr: 0.000142 - momentum: 0.000000 2023-10-14 06:35:22,984 ---------------------------------------------------------------------------------------------------- 2023-10-14 06:35:22,984 EPOCH 2 done: loss 0.0894 - lr: 0.000142 2023-10-14 06:36:03,645 DEV : loss 0.11830706894397736 - f1-score (micro avg) 0.6257 2023-10-14 06:36:03,706 saving best model 2023-10-14 06:36:09,241 ---------------------------------------------------------------------------------------------------- 2023-10-14 06:37:52,453 epoch 3 - iter 361/3617 - loss 0.06677874 - time (sec): 103.21 - samples/sec: 362.87 - lr: 0.000140 - momentum: 0.000000 2023-10-14 06:39:35,134 epoch 3 - iter 722/3617 - loss 0.06628824 - time (sec): 205.89 - samples/sec: 374.82 - lr: 0.000139 - momentum: 0.000000 2023-10-14 06:41:18,032 epoch 3 - iter 1083/3617 - loss 0.06459732 - time (sec): 308.79 - samples/sec: 372.39 - lr: 0.000137 - momentum: 0.000000 2023-10-14 06:43:02,845 epoch 3 - iter 1444/3617 - loss 0.06301256 - time (sec): 413.60 - samples/sec: 369.24 - lr: 0.000135 - momentum: 0.000000 2023-10-14 06:44:48,454 epoch 3 - iter 1805/3617 - loss 0.06356181 - time (sec): 519.21 - samples/sec: 370.00 - lr: 0.000133 - momentum: 0.000000 2023-10-14 06:46:34,140 epoch 3 - iter 2166/3617 - loss 0.06323142 - time (sec): 624.90 - samples/sec: 368.82 - lr: 0.000132 - momentum: 0.000000 2023-10-14 06:48:23,285 epoch 3 - iter 2527/3617 - loss 0.06443387 - time (sec): 734.04 - samples/sec: 363.27 - lr: 0.000130 - momentum: 0.000000 2023-10-14 06:50:15,844 epoch 3 - iter 2888/3617 - loss 0.06348206 - time (sec): 846.60 - samples/sec: 358.62 - lr: 0.000128 - momentum: 0.000000 2023-10-14 06:52:05,894 epoch 3 - iter 3249/3617 - loss 0.06319253 - time (sec): 956.65 - samples/sec: 355.86 - lr: 0.000126 - momentum: 0.000000 2023-10-14 06:53:56,573 epoch 3 - iter 3610/3617 - loss 0.06352064 - time (sec): 1067.33 - samples/sec: 355.17 - lr: 0.000124 - momentum: 0.000000 2023-10-14 06:53:58,615 ---------------------------------------------------------------------------------------------------- 2023-10-14 06:53:58,615 EPOCH 3 done: loss 0.0634 - lr: 0.000124 2023-10-14 06:54:40,585 DEV : loss 0.18002015352249146 - f1-score (micro avg) 0.6241 2023-10-14 06:54:40,652 ---------------------------------------------------------------------------------------------------- 2023-10-14 06:56:25,424 epoch 4 - iter 361/3617 - loss 0.04589010 - time (sec): 104.77 - samples/sec: 349.69 - lr: 0.000123 - momentum: 0.000000 2023-10-14 06:58:11,340 epoch 4 - iter 722/3617 - loss 0.04524784 - time (sec): 210.69 - samples/sec: 357.08 - lr: 0.000121 - momentum: 0.000000 2023-10-14 07:00:01,015 epoch 4 - iter 1083/3617 - loss 0.04459280 - time (sec): 320.36 - samples/sec: 352.15 - lr: 0.000119 - momentum: 0.000000 2023-10-14 07:01:47,053 epoch 4 - iter 1444/3617 - loss 0.04241941 - time (sec): 426.40 - samples/sec: 351.10 - lr: 0.000117 - momentum: 0.000000 2023-10-14 07:03:29,823 epoch 4 - iter 1805/3617 - loss 0.04232486 - time (sec): 529.17 - samples/sec: 354.82 - lr: 0.000116 - momentum: 0.000000 2023-10-14 07:05:12,160 epoch 4 - iter 2166/3617 - loss 0.04384413 - time (sec): 631.51 - samples/sec: 359.79 - lr: 0.000114 - momentum: 0.000000 2023-10-14 07:06:59,230 epoch 4 - iter 2527/3617 - loss 0.04460396 - time (sec): 738.58 - samples/sec: 360.08 - lr: 0.000112 - momentum: 0.000000 2023-10-14 07:08:41,529 epoch 4 - iter 2888/3617 - loss 0.04544588 - time (sec): 840.88 - samples/sec: 360.44 - lr: 0.000110 - momentum: 0.000000 2023-10-14 07:10:25,972 epoch 4 - iter 3249/3617 - loss 0.04605584 - time (sec): 945.32 - samples/sec: 361.87 - lr: 0.000108 - momentum: 0.000000 2023-10-14 07:12:06,260 epoch 4 - iter 3610/3617 - loss 0.04631529 - time (sec): 1045.61 - samples/sec: 362.78 - lr: 0.000107 - momentum: 0.000000 2023-10-14 07:12:07,916 ---------------------------------------------------------------------------------------------------- 2023-10-14 07:12:07,916 EPOCH 4 done: loss 0.0463 - lr: 0.000107 2023-10-14 07:12:47,955 DEV : loss 0.2199789583683014 - f1-score (micro avg) 0.6424 2023-10-14 07:12:48,014 saving best model 2023-10-14 07:12:49,081 ---------------------------------------------------------------------------------------------------- 2023-10-14 07:14:30,776 epoch 5 - iter 361/3617 - loss 0.02732386 - time (sec): 101.69 - samples/sec: 378.30 - lr: 0.000105 - momentum: 0.000000 2023-10-14 07:16:19,492 epoch 5 - iter 722/3617 - loss 0.02918770 - time (sec): 210.41 - samples/sec: 363.23 - lr: 0.000103 - momentum: 0.000000 2023-10-14 07:18:11,252 epoch 5 - iter 1083/3617 - loss 0.03233304 - time (sec): 322.17 - samples/sec: 352.73 - lr: 0.000101 - momentum: 0.000000 2023-10-14 07:20:01,023 epoch 5 - iter 1444/3617 - loss 0.03284681 - time (sec): 431.94 - samples/sec: 347.83 - lr: 0.000100 - momentum: 0.000000 2023-10-14 07:21:44,113 epoch 5 - iter 1805/3617 - loss 0.03214533 - time (sec): 535.03 - samples/sec: 353.10 - lr: 0.000098 - momentum: 0.000000 2023-10-14 07:23:24,045 epoch 5 - iter 2166/3617 - loss 0.03217787 - time (sec): 634.96 - samples/sec: 356.66 - lr: 0.000096 - momentum: 0.000000 2023-10-14 07:25:04,587 epoch 5 - iter 2527/3617 - loss 0.03195592 - time (sec): 735.50 - samples/sec: 357.47 - lr: 0.000094 - momentum: 0.000000 2023-10-14 07:26:45,009 epoch 5 - iter 2888/3617 - loss 0.03247486 - time (sec): 835.93 - samples/sec: 359.86 - lr: 0.000092 - momentum: 0.000000 2023-10-14 07:28:30,143 epoch 5 - iter 3249/3617 - loss 0.03317830 - time (sec): 941.06 - samples/sec: 361.44 - lr: 0.000091 - momentum: 0.000000 2023-10-14 07:30:10,900 epoch 5 - iter 3610/3617 - loss 0.03266682 - time (sec): 1041.82 - samples/sec: 364.12 - lr: 0.000089 - momentum: 0.000000 2023-10-14 07:30:12,602 ---------------------------------------------------------------------------------------------------- 2023-10-14 07:30:12,603 EPOCH 5 done: loss 0.0327 - lr: 0.000089 2023-10-14 07:30:54,210 DEV : loss 0.24780842661857605 - f1-score (micro avg) 0.6605 2023-10-14 07:30:54,277 saving best model 2023-10-14 07:30:59,067 ---------------------------------------------------------------------------------------------------- 2023-10-14 07:32:41,778 epoch 6 - iter 361/3617 - loss 0.01937433 - time (sec): 102.70 - samples/sec: 382.13 - lr: 0.000087 - momentum: 0.000000 2023-10-14 07:34:22,211 epoch 6 - iter 722/3617 - loss 0.01983213 - time (sec): 203.13 - samples/sec: 379.33 - lr: 0.000085 - momentum: 0.000000 2023-10-14 07:36:00,973 epoch 6 - iter 1083/3617 - loss 0.02234270 - time (sec): 301.89 - samples/sec: 379.02 - lr: 0.000084 - momentum: 0.000000 2023-10-14 07:37:43,093 epoch 6 - iter 1444/3617 - loss 0.02403649 - time (sec): 404.01 - samples/sec: 373.77 - lr: 0.000082 - momentum: 0.000000 2023-10-14 07:39:32,914 epoch 6 - iter 1805/3617 - loss 0.02399559 - time (sec): 513.83 - samples/sec: 365.51 - lr: 0.000080 - momentum: 0.000000 2023-10-14 07:41:13,200 epoch 6 - iter 2166/3617 - loss 0.02304717 - time (sec): 614.12 - samples/sec: 366.92 - lr: 0.000078 - momentum: 0.000000 2023-10-14 07:42:57,050 epoch 6 - iter 2527/3617 - loss 0.02261731 - time (sec): 717.97 - samples/sec: 369.09 - lr: 0.000076 - momentum: 0.000000 2023-10-14 07:44:39,865 epoch 6 - iter 2888/3617 - loss 0.02304199 - time (sec): 820.78 - samples/sec: 370.40 - lr: 0.000075 - momentum: 0.000000 2023-10-14 07:46:20,394 epoch 6 - iter 3249/3617 - loss 0.02338086 - time (sec): 921.31 - samples/sec: 369.43 - lr: 0.000073 - momentum: 0.000000 2023-10-14 07:48:04,126 epoch 6 - iter 3610/3617 - loss 0.02327186 - time (sec): 1025.04 - samples/sec: 369.81 - lr: 0.000071 - momentum: 0.000000 2023-10-14 07:48:06,205 ---------------------------------------------------------------------------------------------------- 2023-10-14 07:48:06,205 EPOCH 6 done: loss 0.0232 - lr: 0.000071 2023-10-14 07:48:44,950 DEV : loss 0.2599741816520691 - f1-score (micro avg) 0.6625 2023-10-14 07:48:45,008 saving best model 2023-10-14 07:48:46,022 ---------------------------------------------------------------------------------------------------- 2023-10-14 07:50:26,907 epoch 7 - iter 361/3617 - loss 0.01112762 - time (sec): 100.88 - samples/sec: 380.64 - lr: 0.000069 - momentum: 0.000000 2023-10-14 07:52:10,004 epoch 7 - iter 722/3617 - loss 0.01292842 - time (sec): 203.98 - samples/sec: 373.47 - lr: 0.000068 - momentum: 0.000000 2023-10-14 07:53:56,320 epoch 7 - iter 1083/3617 - loss 0.01286273 - time (sec): 310.30 - samples/sec: 365.83 - lr: 0.000066 - momentum: 0.000000 2023-10-14 07:55:39,361 epoch 7 - iter 1444/3617 - loss 0.01371638 - time (sec): 413.34 - samples/sec: 370.42 - lr: 0.000064 - momentum: 0.000000 2023-10-14 07:57:19,068 epoch 7 - iter 1805/3617 - loss 0.01339942 - time (sec): 513.04 - samples/sec: 371.39 - lr: 0.000062 - momentum: 0.000000 2023-10-14 07:59:01,846 epoch 7 - iter 2166/3617 - loss 0.01421913 - time (sec): 615.82 - samples/sec: 370.62 - lr: 0.000060 - momentum: 0.000000 2023-10-14 08:00:47,048 epoch 7 - iter 2527/3617 - loss 0.01439729 - time (sec): 721.02 - samples/sec: 370.52 - lr: 0.000059 - momentum: 0.000000 2023-10-14 08:02:27,477 epoch 7 - iter 2888/3617 - loss 0.01432233 - time (sec): 821.45 - samples/sec: 370.63 - lr: 0.000057 - momentum: 0.000000 2023-10-14 08:04:08,239 epoch 7 - iter 3249/3617 - loss 0.01466633 - time (sec): 922.21 - samples/sec: 371.36 - lr: 0.000055 - momentum: 0.000000 2023-10-14 08:05:56,328 epoch 7 - iter 3610/3617 - loss 0.01454880 - time (sec): 1030.30 - samples/sec: 368.20 - lr: 0.000053 - momentum: 0.000000 2023-10-14 08:05:58,204 ---------------------------------------------------------------------------------------------------- 2023-10-14 08:05:58,204 EPOCH 7 done: loss 0.0145 - lr: 0.000053 2023-10-14 08:06:46,371 DEV : loss 0.303059846162796 - f1-score (micro avg) 0.6603 2023-10-14 08:06:46,454 ---------------------------------------------------------------------------------------------------- 2023-10-14 08:08:30,534 epoch 8 - iter 361/3617 - loss 0.00660663 - time (sec): 104.08 - samples/sec: 356.89 - lr: 0.000052 - momentum: 0.000000 2023-10-14 08:10:12,010 epoch 8 - iter 722/3617 - loss 0.00864086 - time (sec): 205.55 - samples/sec: 366.15 - lr: 0.000050 - momentum: 0.000000 2023-10-14 08:11:54,676 epoch 8 - iter 1083/3617 - loss 0.00759540 - time (sec): 308.22 - samples/sec: 372.41 - lr: 0.000048 - momentum: 0.000000 2023-10-14 08:13:35,504 epoch 8 - iter 1444/3617 - loss 0.00839154 - time (sec): 409.05 - samples/sec: 372.79 - lr: 0.000046 - momentum: 0.000000 2023-10-14 08:15:20,638 epoch 8 - iter 1805/3617 - loss 0.00820435 - time (sec): 514.18 - samples/sec: 372.20 - lr: 0.000044 - momentum: 0.000000 2023-10-14 08:17:02,003 epoch 8 - iter 2166/3617 - loss 0.00843406 - time (sec): 615.55 - samples/sec: 371.47 - lr: 0.000043 - momentum: 0.000000 2023-10-14 08:18:41,684 epoch 8 - iter 2527/3617 - loss 0.00913255 - time (sec): 715.23 - samples/sec: 372.84 - lr: 0.000041 - momentum: 0.000000 2023-10-14 08:20:30,352 epoch 8 - iter 2888/3617 - loss 0.00931845 - time (sec): 823.90 - samples/sec: 369.40 - lr: 0.000039 - momentum: 0.000000 2023-10-14 08:22:10,317 epoch 8 - iter 3249/3617 - loss 0.00922798 - time (sec): 923.86 - samples/sec: 369.93 - lr: 0.000037 - momentum: 0.000000 2023-10-14 08:23:58,744 epoch 8 - iter 3610/3617 - loss 0.00921932 - time (sec): 1032.29 - samples/sec: 367.62 - lr: 0.000036 - momentum: 0.000000 2023-10-14 08:24:00,299 ---------------------------------------------------------------------------------------------------- 2023-10-14 08:24:00,299 EPOCH 8 done: loss 0.0092 - lr: 0.000036 2023-10-14 08:24:39,988 DEV : loss 0.3336223363876343 - f1-score (micro avg) 0.6534 2023-10-14 08:24:40,055 ---------------------------------------------------------------------------------------------------- 2023-10-14 08:26:22,702 epoch 9 - iter 361/3617 - loss 0.00528020 - time (sec): 102.65 - samples/sec: 377.08 - lr: 0.000034 - momentum: 0.000000 2023-10-14 08:28:04,835 epoch 9 - iter 722/3617 - loss 0.00615366 - time (sec): 204.78 - samples/sec: 381.57 - lr: 0.000032 - momentum: 0.000000 2023-10-14 08:29:42,817 epoch 9 - iter 1083/3617 - loss 0.00574831 - time (sec): 302.76 - samples/sec: 381.88 - lr: 0.000030 - momentum: 0.000000 2023-10-14 08:31:21,510 epoch 9 - iter 1444/3617 - loss 0.00593530 - time (sec): 401.45 - samples/sec: 383.40 - lr: 0.000028 - momentum: 0.000000 2023-10-14 08:33:07,164 epoch 9 - iter 1805/3617 - loss 0.00576089 - time (sec): 507.11 - samples/sec: 377.05 - lr: 0.000027 - momentum: 0.000000 2023-10-14 08:34:51,885 epoch 9 - iter 2166/3617 - loss 0.00642936 - time (sec): 611.83 - samples/sec: 375.27 - lr: 0.000025 - momentum: 0.000000 2023-10-14 08:36:35,388 epoch 9 - iter 2527/3617 - loss 0.00589172 - time (sec): 715.33 - samples/sec: 373.83 - lr: 0.000023 - momentum: 0.000000 2023-10-14 08:38:16,323 epoch 9 - iter 2888/3617 - loss 0.00596664 - time (sec): 816.27 - samples/sec: 372.98 - lr: 0.000021 - momentum: 0.000000 2023-10-14 08:40:00,323 epoch 9 - iter 3249/3617 - loss 0.00596886 - time (sec): 920.27 - samples/sec: 370.19 - lr: 0.000020 - momentum: 0.000000 2023-10-14 08:41:46,467 epoch 9 - iter 3610/3617 - loss 0.00585104 - time (sec): 1026.41 - samples/sec: 369.47 - lr: 0.000018 - momentum: 0.000000 2023-10-14 08:41:48,372 ---------------------------------------------------------------------------------------------------- 2023-10-14 08:41:48,373 EPOCH 9 done: loss 0.0059 - lr: 0.000018 2023-10-14 08:42:29,867 DEV : loss 0.37816229462623596 - f1-score (micro avg) 0.6517 2023-10-14 08:42:29,939 ---------------------------------------------------------------------------------------------------- 2023-10-14 08:44:16,008 epoch 10 - iter 361/3617 - loss 0.00562974 - time (sec): 106.07 - samples/sec: 355.80 - lr: 0.000016 - momentum: 0.000000 2023-10-14 08:45:57,224 epoch 10 - iter 722/3617 - loss 0.00588847 - time (sec): 207.28 - samples/sec: 356.86 - lr: 0.000014 - momentum: 0.000000 2023-10-14 08:47:37,036 epoch 10 - iter 1083/3617 - loss 0.00580326 - time (sec): 307.09 - samples/sec: 366.26 - lr: 0.000012 - momentum: 0.000000 2023-10-14 08:49:18,323 epoch 10 - iter 1444/3617 - loss 0.00517980 - time (sec): 408.38 - samples/sec: 365.73 - lr: 0.000011 - momentum: 0.000000 2023-10-14 08:50:59,735 epoch 10 - iter 1805/3617 - loss 0.00496664 - time (sec): 509.79 - samples/sec: 368.87 - lr: 0.000009 - momentum: 0.000000 2023-10-14 08:52:39,423 epoch 10 - iter 2166/3617 - loss 0.00448679 - time (sec): 609.48 - samples/sec: 369.82 - lr: 0.000007 - momentum: 0.000000 2023-10-14 08:54:21,432 epoch 10 - iter 2527/3617 - loss 0.00434716 - time (sec): 711.49 - samples/sec: 372.23 - lr: 0.000005 - momentum: 0.000000 2023-10-14 08:56:01,306 epoch 10 - iter 2888/3617 - loss 0.00431111 - time (sec): 811.36 - samples/sec: 372.31 - lr: 0.000004 - momentum: 0.000000 2023-10-14 08:57:43,937 epoch 10 - iter 3249/3617 - loss 0.00426079 - time (sec): 914.00 - samples/sec: 373.34 - lr: 0.000002 - momentum: 0.000000 2023-10-14 08:59:25,667 epoch 10 - iter 3610/3617 - loss 0.00423353 - time (sec): 1015.73 - samples/sec: 373.10 - lr: 0.000000 - momentum: 0.000000 2023-10-14 08:59:27,802 ---------------------------------------------------------------------------------------------------- 2023-10-14 08:59:27,802 EPOCH 10 done: loss 0.0042 - lr: 0.000000 2023-10-14 09:00:08,268 DEV : loss 0.3906834125518799 - f1-score (micro avg) 0.6573 2023-10-14 09:00:09,310 ---------------------------------------------------------------------------------------------------- 2023-10-14 09:00:09,312 Loading model from best epoch ... 2023-10-14 09:00:13,705 SequenceTagger predicts: Dictionary with 13 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org 2023-10-14 09:01:11,888 Results: - F-score (micro) 0.6379 - F-score (macro) 0.4852 - Accuracy 0.4792 By class: precision recall f1-score support loc 0.6534 0.7242 0.6870 591 pers 0.5650 0.7423 0.6416 357 org 0.1702 0.1013 0.1270 79 micro avg 0.5986 0.6826 0.6379 1027 macro avg 0.4629 0.5226 0.4852 1027 weighted avg 0.5855 0.6826 0.6282 1027 2023-10-14 09:01:11,888 ----------------------------------------------------------------------------------------------------