2023-10-17 11:13:06,040 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:13:06,042 Model: "SequenceTagger( (embeddings): TransformerWordEmbeddings( (model): ElectraModel( (embeddings): ElectraEmbeddings( (word_embeddings): Embedding(32001, 768) (position_embeddings): Embedding(512, 768) (token_type_embeddings): Embedding(2, 768) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) (encoder): ElectraEncoder( (layer): ModuleList( (0-11): 12 x ElectraLayer( (attention): ElectraAttention( (self): ElectraSelfAttention( (query): Linear(in_features=768, out_features=768, bias=True) (key): Linear(in_features=768, out_features=768, bias=True) (value): Linear(in_features=768, out_features=768, bias=True) (dropout): Dropout(p=0.1, inplace=False) ) (output): ElectraSelfOutput( (dense): Linear(in_features=768, out_features=768, bias=True) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) (intermediate): ElectraIntermediate( (dense): Linear(in_features=768, out_features=3072, bias=True) (intermediate_act_fn): GELUActivation() ) (output): ElectraOutput( (dense): Linear(in_features=3072, out_features=768, bias=True) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) ) ) ) (locked_dropout): LockedDropout(p=0.5) (linear): Linear(in_features=768, out_features=13, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-17 11:13:06,042 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:13:06,042 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 2023-10-17 11:13:06,042 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:13:06,042 Train: 14465 sentences 2023-10-17 11:13:06,042 (train_with_dev=False, train_with_test=False) 2023-10-17 11:13:06,042 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:13:06,042 Training Params: 2023-10-17 11:13:06,042 - learning_rate: "3e-05" 2023-10-17 11:13:06,042 - mini_batch_size: "4" 2023-10-17 11:13:06,042 - max_epochs: "10" 2023-10-17 11:13:06,042 - shuffle: "True" 2023-10-17 11:13:06,042 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:13:06,042 Plugins: 2023-10-17 11:13:06,042 - TensorboardLogger 2023-10-17 11:13:06,042 - LinearScheduler | warmup_fraction: '0.1' 2023-10-17 11:13:06,042 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:13:06,042 Final evaluation on model from best epoch (best-model.pt) 2023-10-17 11:13:06,042 - metric: "('micro avg', 'f1-score')" 2023-10-17 11:13:06,043 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:13:06,043 Computation: 2023-10-17 11:13:06,043 - compute on device: cuda:0 2023-10-17 11:13:06,043 - embedding storage: none 2023-10-17 11:13:06,043 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:13:06,043 Model training base path: "hmbench-letemps/fr-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2" 2023-10-17 11:13:06,043 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:13:06,043 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:13:06,043 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-17 11:13:29,167 epoch 1 - iter 361/3617 - loss 1.78814431 - time (sec): 23.12 - samples/sec: 1638.45 - lr: 0.000003 - momentum: 0.000000 2023-10-17 11:13:51,939 epoch 1 - iter 722/3617 - loss 1.02838317 - time (sec): 45.89 - samples/sec: 1619.95 - lr: 0.000006 - momentum: 0.000000 2023-10-17 11:14:14,528 epoch 1 - iter 1083/3617 - loss 0.72995508 - time (sec): 68.48 - samples/sec: 1666.05 - lr: 0.000009 - momentum: 0.000000 2023-10-17 11:14:36,884 epoch 1 - iter 1444/3617 - loss 0.58463242 - time (sec): 90.84 - samples/sec: 1681.06 - lr: 0.000012 - momentum: 0.000000 2023-10-17 11:14:59,161 epoch 1 - iter 1805/3617 - loss 0.49786262 - time (sec): 113.12 - samples/sec: 1688.72 - lr: 0.000015 - momentum: 0.000000 2023-10-17 11:15:21,931 epoch 1 - iter 2166/3617 - loss 0.43668822 - time (sec): 135.89 - samples/sec: 1681.37 - lr: 0.000018 - momentum: 0.000000 2023-10-17 11:15:44,174 epoch 1 - iter 2527/3617 - loss 0.38974770 - time (sec): 158.13 - samples/sec: 1685.61 - lr: 0.000021 - momentum: 0.000000 2023-10-17 11:16:06,276 epoch 1 - iter 2888/3617 - loss 0.35471982 - time (sec): 180.23 - samples/sec: 1697.15 - lr: 0.000024 - momentum: 0.000000 2023-10-17 11:16:29,303 epoch 1 - iter 3249/3617 - loss 0.32794931 - time (sec): 203.26 - samples/sec: 1690.14 - lr: 0.000027 - momentum: 0.000000 2023-10-17 11:16:51,321 epoch 1 - iter 3610/3617 - loss 0.30809550 - time (sec): 225.28 - samples/sec: 1683.37 - lr: 0.000030 - momentum: 0.000000 2023-10-17 11:16:51,740 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:16:51,740 EPOCH 1 done: loss 0.3077 - lr: 0.000030 2023-10-17 11:16:57,177 DEV : loss 0.10886511206626892 - f1-score (micro avg) 0.5581 2023-10-17 11:16:57,218 saving best model 2023-10-17 11:16:57,714 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:17:21,078 epoch 2 - iter 361/3617 - loss 0.09908194 - time (sec): 23.36 - samples/sec: 1665.79 - lr: 0.000030 - momentum: 0.000000 2023-10-17 11:17:43,704 epoch 2 - iter 722/3617 - loss 0.09370839 - time (sec): 45.99 - samples/sec: 1678.56 - lr: 0.000029 - momentum: 0.000000 2023-10-17 11:18:04,258 epoch 2 - iter 1083/3617 - loss 0.09669946 - time (sec): 66.54 - samples/sec: 1715.28 - lr: 0.000029 - momentum: 0.000000 2023-10-17 11:18:27,278 epoch 2 - iter 1444/3617 - loss 0.09690415 - time (sec): 89.56 - samples/sec: 1693.33 - lr: 0.000029 - momentum: 0.000000 2023-10-17 11:18:50,785 epoch 2 - iter 1805/3617 - loss 0.09640509 - time (sec): 113.07 - samples/sec: 1667.49 - lr: 0.000028 - momentum: 0.000000 2023-10-17 11:19:13,316 epoch 2 - iter 2166/3617 - loss 0.09736145 - time (sec): 135.60 - samples/sec: 1673.97 - lr: 0.000028 - momentum: 0.000000 2023-10-17 11:19:36,497 epoch 2 - iter 2527/3617 - loss 0.09724732 - time (sec): 158.78 - samples/sec: 1674.03 - lr: 0.000028 - momentum: 0.000000 2023-10-17 11:19:59,357 epoch 2 - iter 2888/3617 - loss 0.09753849 - time (sec): 181.64 - samples/sec: 1671.08 - lr: 0.000027 - momentum: 0.000000 2023-10-17 11:20:21,693 epoch 2 - iter 3249/3617 - loss 0.09835920 - time (sec): 203.98 - samples/sec: 1678.98 - lr: 0.000027 - momentum: 0.000000 2023-10-17 11:20:44,900 epoch 2 - iter 3610/3617 - loss 0.09956665 - time (sec): 227.18 - samples/sec: 1669.58 - lr: 0.000027 - momentum: 0.000000 2023-10-17 11:20:45,322 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:20:45,323 EPOCH 2 done: loss 0.0996 - lr: 0.000027 2023-10-17 11:20:53,342 DEV : loss 0.16037893295288086 - f1-score (micro avg) 0.6557 2023-10-17 11:20:53,398 saving best model 2023-10-17 11:20:54,092 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:21:17,522 epoch 3 - iter 361/3617 - loss 0.07564517 - time (sec): 23.43 - samples/sec: 1569.42 - lr: 0.000026 - momentum: 0.000000 2023-10-17 11:21:40,082 epoch 3 - iter 722/3617 - loss 0.07332119 - time (sec): 45.99 - samples/sec: 1636.38 - lr: 0.000026 - momentum: 0.000000 2023-10-17 11:22:02,267 epoch 3 - iter 1083/3617 - loss 0.07222144 - time (sec): 68.17 - samples/sec: 1661.66 - lr: 0.000026 - momentum: 0.000000 2023-10-17 11:22:25,386 epoch 3 - iter 1444/3617 - loss 0.07619382 - time (sec): 91.29 - samples/sec: 1655.98 - lr: 0.000025 - momentum: 0.000000 2023-10-17 11:22:48,049 epoch 3 - iter 1805/3617 - loss 0.07401567 - time (sec): 113.95 - samples/sec: 1657.30 - lr: 0.000025 - momentum: 0.000000 2023-10-17 11:23:11,429 epoch 3 - iter 2166/3617 - loss 0.07403687 - time (sec): 137.33 - samples/sec: 1661.13 - lr: 0.000025 - momentum: 0.000000 2023-10-17 11:23:34,867 epoch 3 - iter 2527/3617 - loss 0.07637370 - time (sec): 160.77 - samples/sec: 1646.45 - lr: 0.000024 - momentum: 0.000000 2023-10-17 11:23:56,997 epoch 3 - iter 2888/3617 - loss 0.07722497 - time (sec): 182.90 - samples/sec: 1656.67 - lr: 0.000024 - momentum: 0.000000 2023-10-17 11:24:18,777 epoch 3 - iter 3249/3617 - loss 0.07737109 - time (sec): 204.68 - samples/sec: 1671.08 - lr: 0.000024 - momentum: 0.000000 2023-10-17 11:24:40,574 epoch 3 - iter 3610/3617 - loss 0.07738280 - time (sec): 226.48 - samples/sec: 1674.38 - lr: 0.000023 - momentum: 0.000000 2023-10-17 11:24:41,006 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:24:41,007 EPOCH 3 done: loss 0.0773 - lr: 0.000023 2023-10-17 11:24:47,385 DEV : loss 0.19637347757816315 - f1-score (micro avg) 0.641 2023-10-17 11:24:47,429 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:25:07,646 epoch 4 - iter 361/3617 - loss 0.05411431 - time (sec): 20.22 - samples/sec: 1910.72 - lr: 0.000023 - momentum: 0.000000 2023-10-17 11:25:31,466 epoch 4 - iter 722/3617 - loss 0.05682246 - time (sec): 44.03 - samples/sec: 1731.33 - lr: 0.000023 - momentum: 0.000000 2023-10-17 11:25:53,516 epoch 4 - iter 1083/3617 - loss 0.05717665 - time (sec): 66.09 - samples/sec: 1747.56 - lr: 0.000022 - momentum: 0.000000 2023-10-17 11:26:15,688 epoch 4 - iter 1444/3617 - loss 0.05618963 - time (sec): 88.26 - samples/sec: 1736.28 - lr: 0.000022 - momentum: 0.000000 2023-10-17 11:26:38,155 epoch 4 - iter 1805/3617 - loss 0.05712367 - time (sec): 110.72 - samples/sec: 1724.46 - lr: 0.000022 - momentum: 0.000000 2023-10-17 11:27:00,949 epoch 4 - iter 2166/3617 - loss 0.05692867 - time (sec): 133.52 - samples/sec: 1712.86 - lr: 0.000021 - momentum: 0.000000 2023-10-17 11:27:23,467 epoch 4 - iter 2527/3617 - loss 0.05681678 - time (sec): 156.04 - samples/sec: 1709.15 - lr: 0.000021 - momentum: 0.000000 2023-10-17 11:27:46,046 epoch 4 - iter 2888/3617 - loss 0.05638457 - time (sec): 178.62 - samples/sec: 1709.53 - lr: 0.000021 - momentum: 0.000000 2023-10-17 11:28:08,185 epoch 4 - iter 3249/3617 - loss 0.05669655 - time (sec): 200.75 - samples/sec: 1707.91 - lr: 0.000020 - momentum: 0.000000 2023-10-17 11:28:29,984 epoch 4 - iter 3610/3617 - loss 0.05711627 - time (sec): 222.55 - samples/sec: 1704.86 - lr: 0.000020 - momentum: 0.000000 2023-10-17 11:28:30,384 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:28:30,384 EPOCH 4 done: loss 0.0571 - lr: 0.000020 2023-10-17 11:28:37,543 DEV : loss 0.22119659185409546 - f1-score (micro avg) 0.6434 2023-10-17 11:28:37,584 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:28:58,907 epoch 5 - iter 361/3617 - loss 0.02823394 - time (sec): 21.32 - samples/sec: 1779.84 - lr: 0.000020 - momentum: 0.000000 2023-10-17 11:29:20,770 epoch 5 - iter 722/3617 - loss 0.03377243 - time (sec): 43.18 - samples/sec: 1773.25 - lr: 0.000019 - momentum: 0.000000 2023-10-17 11:29:43,045 epoch 5 - iter 1083/3617 - loss 0.03634121 - time (sec): 65.46 - samples/sec: 1746.02 - lr: 0.000019 - momentum: 0.000000 2023-10-17 11:30:05,310 epoch 5 - iter 1444/3617 - loss 0.03734307 - time (sec): 87.72 - samples/sec: 1736.47 - lr: 0.000019 - momentum: 0.000000 2023-10-17 11:30:28,900 epoch 5 - iter 1805/3617 - loss 0.03666498 - time (sec): 111.31 - samples/sec: 1721.54 - lr: 0.000018 - momentum: 0.000000 2023-10-17 11:30:51,224 epoch 5 - iter 2166/3617 - loss 0.03718897 - time (sec): 133.64 - samples/sec: 1727.19 - lr: 0.000018 - momentum: 0.000000 2023-10-17 11:31:13,996 epoch 5 - iter 2527/3617 - loss 0.03914152 - time (sec): 156.41 - samples/sec: 1708.55 - lr: 0.000018 - momentum: 0.000000 2023-10-17 11:31:36,302 epoch 5 - iter 2888/3617 - loss 0.03939405 - time (sec): 178.72 - samples/sec: 1693.34 - lr: 0.000017 - momentum: 0.000000 2023-10-17 11:31:58,970 epoch 5 - iter 3249/3617 - loss 0.03829028 - time (sec): 201.38 - samples/sec: 1694.76 - lr: 0.000017 - momentum: 0.000000 2023-10-17 11:32:22,708 epoch 5 - iter 3610/3617 - loss 0.03943816 - time (sec): 225.12 - samples/sec: 1683.96 - lr: 0.000017 - momentum: 0.000000 2023-10-17 11:32:23,170 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:32:23,170 EPOCH 5 done: loss 0.0394 - lr: 0.000017 2023-10-17 11:32:29,478 DEV : loss 0.29799073934555054 - f1-score (micro avg) 0.6527 2023-10-17 11:32:29,521 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:32:52,384 epoch 6 - iter 361/3617 - loss 0.02573007 - time (sec): 22.86 - samples/sec: 1637.30 - lr: 0.000016 - momentum: 0.000000 2023-10-17 11:33:15,775 epoch 6 - iter 722/3617 - loss 0.02511481 - time (sec): 46.25 - samples/sec: 1615.72 - lr: 0.000016 - momentum: 0.000000 2023-10-17 11:33:38,654 epoch 6 - iter 1083/3617 - loss 0.02682181 - time (sec): 69.13 - samples/sec: 1631.61 - lr: 0.000016 - momentum: 0.000000 2023-10-17 11:34:02,271 epoch 6 - iter 1444/3617 - loss 0.02846690 - time (sec): 92.75 - samples/sec: 1630.95 - lr: 0.000015 - momentum: 0.000000 2023-10-17 11:34:25,914 epoch 6 - iter 1805/3617 - loss 0.02838906 - time (sec): 116.39 - samples/sec: 1631.91 - lr: 0.000015 - momentum: 0.000000 2023-10-17 11:34:49,761 epoch 6 - iter 2166/3617 - loss 0.02877434 - time (sec): 140.24 - samples/sec: 1626.20 - lr: 0.000015 - momentum: 0.000000 2023-10-17 11:35:13,140 epoch 6 - iter 2527/3617 - loss 0.02838507 - time (sec): 163.62 - samples/sec: 1627.41 - lr: 0.000014 - momentum: 0.000000 2023-10-17 11:35:35,438 epoch 6 - iter 2888/3617 - loss 0.02895275 - time (sec): 185.92 - samples/sec: 1635.29 - lr: 0.000014 - momentum: 0.000000 2023-10-17 11:35:58,001 epoch 6 - iter 3249/3617 - loss 0.02881247 - time (sec): 208.48 - samples/sec: 1636.41 - lr: 0.000014 - momentum: 0.000000 2023-10-17 11:36:21,047 epoch 6 - iter 3610/3617 - loss 0.02904294 - time (sec): 231.52 - samples/sec: 1638.64 - lr: 0.000013 - momentum: 0.000000 2023-10-17 11:36:21,475 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:36:21,475 EPOCH 6 done: loss 0.0291 - lr: 0.000013 2023-10-17 11:36:27,876 DEV : loss 0.3385041356086731 - f1-score (micro avg) 0.6408 2023-10-17 11:36:27,920 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:36:50,238 epoch 7 - iter 361/3617 - loss 0.01432701 - time (sec): 22.32 - samples/sec: 1700.15 - lr: 0.000013 - momentum: 0.000000 2023-10-17 11:37:10,807 epoch 7 - iter 722/3617 - loss 0.01555267 - time (sec): 42.88 - samples/sec: 1756.46 - lr: 0.000013 - momentum: 0.000000 2023-10-17 11:37:34,793 epoch 7 - iter 1083/3617 - loss 0.01777501 - time (sec): 66.87 - samples/sec: 1692.09 - lr: 0.000012 - momentum: 0.000000 2023-10-17 11:37:56,843 epoch 7 - iter 1444/3617 - loss 0.02027093 - time (sec): 88.92 - samples/sec: 1699.03 - lr: 0.000012 - momentum: 0.000000 2023-10-17 11:38:19,092 epoch 7 - iter 1805/3617 - loss 0.01980969 - time (sec): 111.17 - samples/sec: 1702.08 - lr: 0.000012 - momentum: 0.000000 2023-10-17 11:38:41,205 epoch 7 - iter 2166/3617 - loss 0.02022525 - time (sec): 133.28 - samples/sec: 1702.67 - lr: 0.000011 - momentum: 0.000000 2023-10-17 11:39:04,065 epoch 7 - iter 2527/3617 - loss 0.02165812 - time (sec): 156.14 - samples/sec: 1700.48 - lr: 0.000011 - momentum: 0.000000 2023-10-17 11:39:27,133 epoch 7 - iter 2888/3617 - loss 0.02180513 - time (sec): 179.21 - samples/sec: 1692.36 - lr: 0.000011 - momentum: 0.000000 2023-10-17 11:39:49,393 epoch 7 - iter 3249/3617 - loss 0.02148273 - time (sec): 201.47 - samples/sec: 1697.80 - lr: 0.000010 - momentum: 0.000000 2023-10-17 11:40:11,734 epoch 7 - iter 3610/3617 - loss 0.02149257 - time (sec): 223.81 - samples/sec: 1694.80 - lr: 0.000010 - momentum: 0.000000 2023-10-17 11:40:12,185 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:40:12,186 EPOCH 7 done: loss 0.0216 - lr: 0.000010 2023-10-17 11:40:18,549 DEV : loss 0.3538167476654053 - f1-score (micro avg) 0.6501 2023-10-17 11:40:18,591 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:40:40,700 epoch 8 - iter 361/3617 - loss 0.01192797 - time (sec): 22.11 - samples/sec: 1705.35 - lr: 0.000010 - momentum: 0.000000 2023-10-17 11:41:03,756 epoch 8 - iter 722/3617 - loss 0.01213621 - time (sec): 45.16 - samples/sec: 1658.34 - lr: 0.000009 - momentum: 0.000000 2023-10-17 11:41:27,927 epoch 8 - iter 1083/3617 - loss 0.01286573 - time (sec): 69.33 - samples/sec: 1624.59 - lr: 0.000009 - momentum: 0.000000 2023-10-17 11:41:51,754 epoch 8 - iter 1444/3617 - loss 0.01313113 - time (sec): 93.16 - samples/sec: 1629.06 - lr: 0.000009 - momentum: 0.000000 2023-10-17 11:42:16,517 epoch 8 - iter 1805/3617 - loss 0.01306628 - time (sec): 117.92 - samples/sec: 1601.35 - lr: 0.000008 - momentum: 0.000000 2023-10-17 11:42:41,872 epoch 8 - iter 2166/3617 - loss 0.01391368 - time (sec): 143.28 - samples/sec: 1578.71 - lr: 0.000008 - momentum: 0.000000 2023-10-17 11:43:06,573 epoch 8 - iter 2527/3617 - loss 0.01391572 - time (sec): 167.98 - samples/sec: 1569.84 - lr: 0.000008 - momentum: 0.000000 2023-10-17 11:43:30,349 epoch 8 - iter 2888/3617 - loss 0.01336401 - time (sec): 191.76 - samples/sec: 1576.08 - lr: 0.000007 - momentum: 0.000000 2023-10-17 11:43:51,479 epoch 8 - iter 3249/3617 - loss 0.01320290 - time (sec): 212.89 - samples/sec: 1603.34 - lr: 0.000007 - momentum: 0.000000 2023-10-17 11:44:14,085 epoch 8 - iter 3610/3617 - loss 0.01301754 - time (sec): 235.49 - samples/sec: 1610.19 - lr: 0.000007 - momentum: 0.000000 2023-10-17 11:44:14,508 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:44:14,508 EPOCH 8 done: loss 0.0130 - lr: 0.000007 2023-10-17 11:44:20,974 DEV : loss 0.3859119117259979 - f1-score (micro avg) 0.6527 2023-10-17 11:44:21,026 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:44:46,418 epoch 9 - iter 361/3617 - loss 0.00859029 - time (sec): 25.39 - samples/sec: 1498.46 - lr: 0.000006 - momentum: 0.000000 2023-10-17 11:45:09,053 epoch 9 - iter 722/3617 - loss 0.00920184 - time (sec): 48.02 - samples/sec: 1550.20 - lr: 0.000006 - momentum: 0.000000 2023-10-17 11:45:31,985 epoch 9 - iter 1083/3617 - loss 0.00936040 - time (sec): 70.96 - samples/sec: 1569.14 - lr: 0.000006 - momentum: 0.000000 2023-10-17 11:45:55,056 epoch 9 - iter 1444/3617 - loss 0.00967607 - time (sec): 94.03 - samples/sec: 1585.73 - lr: 0.000005 - momentum: 0.000000 2023-10-17 11:46:18,172 epoch 9 - iter 1805/3617 - loss 0.00907964 - time (sec): 117.14 - samples/sec: 1605.14 - lr: 0.000005 - momentum: 0.000000 2023-10-17 11:46:41,668 epoch 9 - iter 2166/3617 - loss 0.00893534 - time (sec): 140.64 - samples/sec: 1607.77 - lr: 0.000005 - momentum: 0.000000 2023-10-17 11:47:04,925 epoch 9 - iter 2527/3617 - loss 0.00849365 - time (sec): 163.90 - samples/sec: 1612.69 - lr: 0.000004 - momentum: 0.000000 2023-10-17 11:47:28,072 epoch 9 - iter 2888/3617 - loss 0.00835917 - time (sec): 187.04 - samples/sec: 1618.16 - lr: 0.000004 - momentum: 0.000000 2023-10-17 11:47:51,051 epoch 9 - iter 3249/3617 - loss 0.00834899 - time (sec): 210.02 - samples/sec: 1619.02 - lr: 0.000004 - momentum: 0.000000 2023-10-17 11:48:12,830 epoch 9 - iter 3610/3617 - loss 0.00831963 - time (sec): 231.80 - samples/sec: 1635.73 - lr: 0.000003 - momentum: 0.000000 2023-10-17 11:48:13,250 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:48:13,251 EPOCH 9 done: loss 0.0083 - lr: 0.000003 2023-10-17 11:48:19,769 DEV : loss 0.4002314805984497 - f1-score (micro avg) 0.6641 2023-10-17 11:48:19,813 saving best model 2023-10-17 11:48:20,421 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:48:42,618 epoch 10 - iter 361/3617 - loss 0.00672252 - time (sec): 22.20 - samples/sec: 1651.90 - lr: 0.000003 - momentum: 0.000000 2023-10-17 11:49:05,425 epoch 10 - iter 722/3617 - loss 0.00611528 - time (sec): 45.00 - samples/sec: 1679.01 - lr: 0.000003 - momentum: 0.000000 2023-10-17 11:49:28,674 epoch 10 - iter 1083/3617 - loss 0.00518175 - time (sec): 68.25 - samples/sec: 1637.70 - lr: 0.000002 - momentum: 0.000000 2023-10-17 11:49:51,468 epoch 10 - iter 1444/3617 - loss 0.00501517 - time (sec): 91.04 - samples/sec: 1652.46 - lr: 0.000002 - momentum: 0.000000 2023-10-17 11:50:13,801 epoch 10 - iter 1805/3617 - loss 0.00526015 - time (sec): 113.38 - samples/sec: 1658.65 - lr: 0.000002 - momentum: 0.000000 2023-10-17 11:50:33,920 epoch 10 - iter 2166/3617 - loss 0.00556562 - time (sec): 133.50 - samples/sec: 1695.49 - lr: 0.000001 - momentum: 0.000000 2023-10-17 11:50:57,467 epoch 10 - iter 2527/3617 - loss 0.00515816 - time (sec): 157.04 - samples/sec: 1674.52 - lr: 0.000001 - momentum: 0.000000 2023-10-17 11:51:21,410 epoch 10 - iter 2888/3617 - loss 0.00538365 - time (sec): 180.99 - samples/sec: 1670.23 - lr: 0.000001 - momentum: 0.000000 2023-10-17 11:51:44,574 epoch 10 - iter 3249/3617 - loss 0.00546926 - time (sec): 204.15 - samples/sec: 1673.63 - lr: 0.000000 - momentum: 0.000000 2023-10-17 11:52:07,470 epoch 10 - iter 3610/3617 - loss 0.00532840 - time (sec): 227.05 - samples/sec: 1670.42 - lr: 0.000000 - momentum: 0.000000 2023-10-17 11:52:07,907 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:52:07,908 EPOCH 10 done: loss 0.0053 - lr: 0.000000 2023-10-17 11:52:15,079 DEV : loss 0.4149819612503052 - f1-score (micro avg) 0.6604 2023-10-17 11:52:15,631 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:52:15,632 Loading model from best epoch ... 2023-10-17 11:52:17,415 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 2023-10-17 11:52:25,444 Results: - F-score (micro) 0.6525 - F-score (macro) 0.5043 - Accuracy 0.498 By class: precision recall f1-score support loc 0.6511 0.7800 0.7098 591 pers 0.5768 0.7367 0.6470 357 org 0.1774 0.1392 0.1560 79 micro avg 0.5995 0.7157 0.6525 1027 macro avg 0.4684 0.5520 0.5043 1027 weighted avg 0.5888 0.7157 0.6454 1027 2023-10-17 11:52:25,444 ----------------------------------------------------------------------------------------------------