2023-10-17 11:50:59,878 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:50:59,879 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 11:50:59,879 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:50:59,879 MultiCorpus: 7936 train + 992 dev + 992 test sentences - NER_ICDAR_EUROPEANA Corpus: 7936 train + 992 dev + 992 test sentences - /root/.flair/datasets/ner_icdar_europeana/fr 2023-10-17 11:50:59,879 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:50:59,879 Train: 7936 sentences 2023-10-17 11:50:59,879 (train_with_dev=False, train_with_test=False) 2023-10-17 11:50:59,879 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:50:59,879 Training Params: 2023-10-17 11:50:59,879 - learning_rate: "5e-05" 2023-10-17 11:50:59,879 - mini_batch_size: "4" 2023-10-17 11:50:59,879 - max_epochs: "10" 2023-10-17 11:50:59,879 - shuffle: "True" 2023-10-17 11:50:59,879 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:50:59,879 Plugins: 2023-10-17 11:50:59,879 - TensorboardLogger 2023-10-17 11:50:59,879 - LinearScheduler | warmup_fraction: '0.1' 2023-10-17 11:50:59,879 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:50:59,879 Final evaluation on model from best epoch (best-model.pt) 2023-10-17 11:50:59,879 - metric: "('micro avg', 'f1-score')" 2023-10-17 11:50:59,879 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:50:59,879 Computation: 2023-10-17 11:50:59,880 - compute on device: cuda:0 2023-10-17 11:50:59,880 - embedding storage: none 2023-10-17 11:50:59,880 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:50:59,880 Model training base path: "hmbench-icdar/fr-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2" 2023-10-17 11:50:59,880 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:50:59,880 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:50:59,880 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-17 11:51:08,865 epoch 1 - iter 198/1984 - loss 1.99658672 - time (sec): 8.98 - samples/sec: 1773.62 - lr: 0.000005 - momentum: 0.000000 2023-10-17 11:51:17,977 epoch 1 - iter 396/1984 - loss 1.11484466 - time (sec): 18.10 - samples/sec: 1803.46 - lr: 0.000010 - momentum: 0.000000 2023-10-17 11:51:26,677 epoch 1 - iter 594/1984 - loss 0.82031286 - time (sec): 26.80 - samples/sec: 1822.19 - lr: 0.000015 - momentum: 0.000000 2023-10-17 11:51:35,306 epoch 1 - iter 792/1984 - loss 0.65663170 - time (sec): 35.42 - samples/sec: 1846.14 - lr: 0.000020 - momentum: 0.000000 2023-10-17 11:51:44,319 epoch 1 - iter 990/1984 - loss 0.55179417 - time (sec): 44.44 - samples/sec: 1861.84 - lr: 0.000025 - momentum: 0.000000 2023-10-17 11:51:53,348 epoch 1 - iter 1188/1984 - loss 0.49119338 - time (sec): 53.47 - samples/sec: 1852.30 - lr: 0.000030 - momentum: 0.000000 2023-10-17 11:52:02,343 epoch 1 - iter 1386/1984 - loss 0.44256195 - time (sec): 62.46 - samples/sec: 1858.68 - lr: 0.000035 - momentum: 0.000000 2023-10-17 11:52:11,442 epoch 1 - iter 1584/1984 - loss 0.40740944 - time (sec): 71.56 - samples/sec: 1839.94 - lr: 0.000040 - momentum: 0.000000 2023-10-17 11:52:20,490 epoch 1 - iter 1782/1984 - loss 0.37924328 - time (sec): 80.61 - samples/sec: 1831.13 - lr: 0.000045 - momentum: 0.000000 2023-10-17 11:52:29,468 epoch 1 - iter 1980/1984 - loss 0.35390636 - time (sec): 89.59 - samples/sec: 1827.85 - lr: 0.000050 - momentum: 0.000000 2023-10-17 11:52:29,645 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:52:29,645 EPOCH 1 done: loss 0.3535 - lr: 0.000050 2023-10-17 11:52:32,795 DEV : loss 0.11604664474725723 - f1-score (micro avg) 0.7365 2023-10-17 11:52:32,816 saving best model 2023-10-17 11:52:33,192 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:52:42,328 epoch 2 - iter 198/1984 - loss 0.14192747 - time (sec): 9.13 - samples/sec: 1705.31 - lr: 0.000049 - momentum: 0.000000 2023-10-17 11:52:51,310 epoch 2 - iter 396/1984 - loss 0.12714871 - time (sec): 18.12 - samples/sec: 1747.11 - lr: 0.000049 - momentum: 0.000000 2023-10-17 11:53:00,111 epoch 2 - iter 594/1984 - loss 0.12855381 - time (sec): 26.92 - samples/sec: 1792.02 - lr: 0.000048 - momentum: 0.000000 2023-10-17 11:53:08,839 epoch 2 - iter 792/1984 - loss 0.12867843 - time (sec): 35.65 - samples/sec: 1811.84 - lr: 0.000048 - momentum: 0.000000 2023-10-17 11:53:17,959 epoch 2 - iter 990/1984 - loss 0.12838419 - time (sec): 44.77 - samples/sec: 1825.07 - lr: 0.000047 - momentum: 0.000000 2023-10-17 11:53:26,874 epoch 2 - iter 1188/1984 - loss 0.12740182 - time (sec): 53.68 - samples/sec: 1822.71 - lr: 0.000047 - momentum: 0.000000 2023-10-17 11:53:35,927 epoch 2 - iter 1386/1984 - loss 0.12351792 - time (sec): 62.73 - samples/sec: 1822.16 - lr: 0.000046 - momentum: 0.000000 2023-10-17 11:53:44,983 epoch 2 - iter 1584/1984 - loss 0.12218031 - time (sec): 71.79 - samples/sec: 1830.55 - lr: 0.000046 - momentum: 0.000000 2023-10-17 11:53:54,028 epoch 2 - iter 1782/1984 - loss 0.12157226 - time (sec): 80.83 - samples/sec: 1829.60 - lr: 0.000045 - momentum: 0.000000 2023-10-17 11:54:03,013 epoch 2 - iter 1980/1984 - loss 0.12534668 - time (sec): 89.82 - samples/sec: 1822.84 - lr: 0.000044 - momentum: 0.000000 2023-10-17 11:54:03,190 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:54:03,190 EPOCH 2 done: loss 0.1253 - lr: 0.000044 2023-10-17 11:54:06,991 DEV : loss 0.09573990851640701 - f1-score (micro avg) 0.7572 2023-10-17 11:54:07,012 saving best model 2023-10-17 11:54:07,504 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:54:16,693 epoch 3 - iter 198/1984 - loss 0.09494312 - time (sec): 9.19 - samples/sec: 1831.31 - lr: 0.000044 - momentum: 0.000000 2023-10-17 11:54:25,610 epoch 3 - iter 396/1984 - loss 0.09329297 - time (sec): 18.10 - samples/sec: 1821.18 - lr: 0.000043 - momentum: 0.000000 2023-10-17 11:54:34,690 epoch 3 - iter 594/1984 - loss 0.09450932 - time (sec): 27.18 - samples/sec: 1832.48 - lr: 0.000043 - momentum: 0.000000 2023-10-17 11:54:43,567 epoch 3 - iter 792/1984 - loss 0.09880612 - time (sec): 36.06 - samples/sec: 1796.09 - lr: 0.000042 - momentum: 0.000000 2023-10-17 11:54:52,286 epoch 3 - iter 990/1984 - loss 0.09577827 - time (sec): 44.78 - samples/sec: 1815.05 - lr: 0.000042 - momentum: 0.000000 2023-10-17 11:55:00,816 epoch 3 - iter 1188/1984 - loss 0.09551043 - time (sec): 53.31 - samples/sec: 1829.07 - lr: 0.000041 - momentum: 0.000000 2023-10-17 11:55:09,757 epoch 3 - iter 1386/1984 - loss 0.09500722 - time (sec): 62.25 - samples/sec: 1852.97 - lr: 0.000041 - momentum: 0.000000 2023-10-17 11:55:18,875 epoch 3 - iter 1584/1984 - loss 0.09605364 - time (sec): 71.37 - samples/sec: 1848.06 - lr: 0.000040 - momentum: 0.000000 2023-10-17 11:55:27,832 epoch 3 - iter 1782/1984 - loss 0.09581236 - time (sec): 80.32 - samples/sec: 1834.01 - lr: 0.000039 - momentum: 0.000000 2023-10-17 11:55:36,807 epoch 3 - iter 1980/1984 - loss 0.09610021 - time (sec): 89.30 - samples/sec: 1832.73 - lr: 0.000039 - momentum: 0.000000 2023-10-17 11:55:36,985 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:55:36,986 EPOCH 3 done: loss 0.0960 - lr: 0.000039 2023-10-17 11:55:40,348 DEV : loss 0.12434609979391098 - f1-score (micro avg) 0.7354 2023-10-17 11:55:40,368 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:55:49,368 epoch 4 - iter 198/1984 - loss 0.05981882 - time (sec): 9.00 - samples/sec: 1857.83 - lr: 0.000038 - momentum: 0.000000 2023-10-17 11:55:58,236 epoch 4 - iter 396/1984 - loss 0.06629030 - time (sec): 17.87 - samples/sec: 1808.21 - lr: 0.000038 - momentum: 0.000000 2023-10-17 11:56:07,217 epoch 4 - iter 594/1984 - loss 0.07164876 - time (sec): 26.85 - samples/sec: 1770.52 - lr: 0.000037 - momentum: 0.000000 2023-10-17 11:56:16,630 epoch 4 - iter 792/1984 - loss 0.07351127 - time (sec): 36.26 - samples/sec: 1781.58 - lr: 0.000037 - momentum: 0.000000 2023-10-17 11:56:25,747 epoch 4 - iter 990/1984 - loss 0.07151819 - time (sec): 45.38 - samples/sec: 1786.64 - lr: 0.000036 - momentum: 0.000000 2023-10-17 11:56:34,907 epoch 4 - iter 1188/1984 - loss 0.07095217 - time (sec): 54.54 - samples/sec: 1778.05 - lr: 0.000036 - momentum: 0.000000 2023-10-17 11:56:43,989 epoch 4 - iter 1386/1984 - loss 0.07050035 - time (sec): 63.62 - samples/sec: 1786.16 - lr: 0.000035 - momentum: 0.000000 2023-10-17 11:56:53,161 epoch 4 - iter 1584/1984 - loss 0.07691480 - time (sec): 72.79 - samples/sec: 1792.55 - lr: 0.000034 - momentum: 0.000000 2023-10-17 11:57:02,381 epoch 4 - iter 1782/1984 - loss 0.07542490 - time (sec): 82.01 - samples/sec: 1799.04 - lr: 0.000034 - momentum: 0.000000 2023-10-17 11:57:11,654 epoch 4 - iter 1980/1984 - loss 0.07641851 - time (sec): 91.28 - samples/sec: 1793.27 - lr: 0.000033 - momentum: 0.000000 2023-10-17 11:57:11,837 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:57:11,837 EPOCH 4 done: loss 0.0763 - lr: 0.000033 2023-10-17 11:57:15,244 DEV : loss 0.18466949462890625 - f1-score (micro avg) 0.7408 2023-10-17 11:57:15,266 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:57:24,362 epoch 5 - iter 198/1984 - loss 0.04859297 - time (sec): 9.10 - samples/sec: 1863.76 - lr: 0.000033 - momentum: 0.000000 2023-10-17 11:57:33,455 epoch 5 - iter 396/1984 - loss 0.04635554 - time (sec): 18.19 - samples/sec: 1835.22 - lr: 0.000032 - momentum: 0.000000 2023-10-17 11:57:42,513 epoch 5 - iter 594/1984 - loss 0.04871548 - time (sec): 27.25 - samples/sec: 1821.40 - lr: 0.000032 - momentum: 0.000000 2023-10-17 11:57:51,623 epoch 5 - iter 792/1984 - loss 0.05252835 - time (sec): 36.36 - samples/sec: 1805.94 - lr: 0.000031 - momentum: 0.000000 2023-10-17 11:58:00,844 epoch 5 - iter 990/1984 - loss 0.05388850 - time (sec): 45.58 - samples/sec: 1792.29 - lr: 0.000031 - momentum: 0.000000 2023-10-17 11:58:09,855 epoch 5 - iter 1188/1984 - loss 0.05515150 - time (sec): 54.59 - samples/sec: 1778.02 - lr: 0.000030 - momentum: 0.000000 2023-10-17 11:58:19,340 epoch 5 - iter 1386/1984 - loss 0.05623579 - time (sec): 64.07 - samples/sec: 1779.81 - lr: 0.000029 - momentum: 0.000000 2023-10-17 11:58:28,807 epoch 5 - iter 1584/1984 - loss 0.05445857 - time (sec): 73.54 - samples/sec: 1779.22 - lr: 0.000029 - momentum: 0.000000 2023-10-17 11:58:37,858 epoch 5 - iter 1782/1984 - loss 0.05549020 - time (sec): 82.59 - samples/sec: 1778.92 - lr: 0.000028 - momentum: 0.000000 2023-10-17 11:58:46,900 epoch 5 - iter 1980/1984 - loss 0.05573079 - time (sec): 91.63 - samples/sec: 1786.04 - lr: 0.000028 - momentum: 0.000000 2023-10-17 11:58:47,077 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:58:47,077 EPOCH 5 done: loss 0.0556 - lr: 0.000028 2023-10-17 11:58:50,444 DEV : loss 0.1826779842376709 - f1-score (micro avg) 0.7561 2023-10-17 11:58:50,465 ---------------------------------------------------------------------------------------------------- 2023-10-17 11:58:59,462 epoch 6 - iter 198/1984 - loss 0.03393163 - time (sec): 9.00 - samples/sec: 1795.87 - lr: 0.000027 - momentum: 0.000000 2023-10-17 11:59:08,616 epoch 6 - iter 396/1984 - loss 0.03547336 - time (sec): 18.15 - samples/sec: 1805.34 - lr: 0.000027 - momentum: 0.000000 2023-10-17 11:59:17,653 epoch 6 - iter 594/1984 - loss 0.03602657 - time (sec): 27.19 - samples/sec: 1787.83 - lr: 0.000026 - momentum: 0.000000 2023-10-17 11:59:26,822 epoch 6 - iter 792/1984 - loss 0.03952462 - time (sec): 36.35 - samples/sec: 1804.95 - lr: 0.000026 - momentum: 0.000000 2023-10-17 11:59:36,151 epoch 6 - iter 990/1984 - loss 0.03912622 - time (sec): 45.68 - samples/sec: 1810.09 - lr: 0.000025 - momentum: 0.000000 2023-10-17 11:59:45,232 epoch 6 - iter 1188/1984 - loss 0.03870866 - time (sec): 54.77 - samples/sec: 1813.32 - lr: 0.000024 - momentum: 0.000000 2023-10-17 11:59:53,643 epoch 6 - iter 1386/1984 - loss 0.04046966 - time (sec): 63.18 - samples/sec: 1815.61 - lr: 0.000024 - momentum: 0.000000 2023-10-17 12:00:02,231 epoch 6 - iter 1584/1984 - loss 0.03983182 - time (sec): 71.76 - samples/sec: 1827.05 - lr: 0.000023 - momentum: 0.000000 2023-10-17 12:00:10,739 epoch 6 - iter 1782/1984 - loss 0.03988919 - time (sec): 80.27 - samples/sec: 1834.23 - lr: 0.000023 - momentum: 0.000000 2023-10-17 12:00:19,695 epoch 6 - iter 1980/1984 - loss 0.04066461 - time (sec): 89.23 - samples/sec: 1834.70 - lr: 0.000022 - momentum: 0.000000 2023-10-17 12:00:19,885 ---------------------------------------------------------------------------------------------------- 2023-10-17 12:00:19,885 EPOCH 6 done: loss 0.0406 - lr: 0.000022 2023-10-17 12:00:23,280 DEV : loss 0.1922488510608673 - f1-score (micro avg) 0.7417 2023-10-17 12:00:23,303 ---------------------------------------------------------------------------------------------------- 2023-10-17 12:00:32,599 epoch 7 - iter 198/1984 - loss 0.03507544 - time (sec): 9.29 - samples/sec: 1779.65 - lr: 0.000022 - momentum: 0.000000 2023-10-17 12:00:41,907 epoch 7 - iter 396/1984 - loss 0.03033014 - time (sec): 18.60 - samples/sec: 1789.70 - lr: 0.000021 - momentum: 0.000000 2023-10-17 12:00:51,166 epoch 7 - iter 594/1984 - loss 0.02968150 - time (sec): 27.86 - samples/sec: 1802.14 - lr: 0.000021 - momentum: 0.000000 2023-10-17 12:01:00,212 epoch 7 - iter 792/1984 - loss 0.02789631 - time (sec): 36.91 - samples/sec: 1810.72 - lr: 0.000020 - momentum: 0.000000 2023-10-17 12:01:08,802 epoch 7 - iter 990/1984 - loss 0.02776484 - time (sec): 45.50 - samples/sec: 1838.23 - lr: 0.000019 - momentum: 0.000000 2023-10-17 12:01:17,842 epoch 7 - iter 1188/1984 - loss 0.02661241 - time (sec): 54.54 - samples/sec: 1822.87 - lr: 0.000019 - momentum: 0.000000 2023-10-17 12:01:26,856 epoch 7 - iter 1386/1984 - loss 0.02640200 - time (sec): 63.55 - samples/sec: 1809.99 - lr: 0.000018 - momentum: 0.000000 2023-10-17 12:01:35,967 epoch 7 - iter 1584/1984 - loss 0.02663353 - time (sec): 72.66 - samples/sec: 1806.05 - lr: 0.000018 - momentum: 0.000000 2023-10-17 12:01:44,950 epoch 7 - iter 1782/1984 - loss 0.02669130 - time (sec): 81.64 - samples/sec: 1798.68 - lr: 0.000017 - momentum: 0.000000 2023-10-17 12:01:54,310 epoch 7 - iter 1980/1984 - loss 0.02730554 - time (sec): 91.00 - samples/sec: 1798.73 - lr: 0.000017 - momentum: 0.000000 2023-10-17 12:01:54,485 ---------------------------------------------------------------------------------------------------- 2023-10-17 12:01:54,485 EPOCH 7 done: loss 0.0273 - lr: 0.000017 2023-10-17 12:01:58,292 DEV : loss 0.18583810329437256 - f1-score (micro avg) 0.756 2023-10-17 12:01:58,313 ---------------------------------------------------------------------------------------------------- 2023-10-17 12:02:07,377 epoch 8 - iter 198/1984 - loss 0.01962946 - time (sec): 9.06 - samples/sec: 1832.57 - lr: 0.000016 - momentum: 0.000000 2023-10-17 12:02:16,417 epoch 8 - iter 396/1984 - loss 0.02070604 - time (sec): 18.10 - samples/sec: 1860.49 - lr: 0.000016 - momentum: 0.000000 2023-10-17 12:02:25,338 epoch 8 - iter 594/1984 - loss 0.02069916 - time (sec): 27.02 - samples/sec: 1830.33 - lr: 0.000015 - momentum: 0.000000 2023-10-17 12:02:34,277 epoch 8 - iter 792/1984 - loss 0.02086583 - time (sec): 35.96 - samples/sec: 1822.60 - lr: 0.000014 - momentum: 0.000000 2023-10-17 12:02:43,312 epoch 8 - iter 990/1984 - loss 0.01924450 - time (sec): 45.00 - samples/sec: 1831.26 - lr: 0.000014 - momentum: 0.000000 2023-10-17 12:02:52,365 epoch 8 - iter 1188/1984 - loss 0.01949509 - time (sec): 54.05 - samples/sec: 1819.84 - lr: 0.000013 - momentum: 0.000000 2023-10-17 12:03:01,315 epoch 8 - iter 1386/1984 - loss 0.01960191 - time (sec): 63.00 - samples/sec: 1804.63 - lr: 0.000013 - momentum: 0.000000 2023-10-17 12:03:10,558 epoch 8 - iter 1584/1984 - loss 0.02035400 - time (sec): 72.24 - samples/sec: 1811.71 - lr: 0.000012 - momentum: 0.000000 2023-10-17 12:03:19,519 epoch 8 - iter 1782/1984 - loss 0.01981080 - time (sec): 81.20 - samples/sec: 1809.61 - lr: 0.000012 - momentum: 0.000000 2023-10-17 12:03:28,832 epoch 8 - iter 1980/1984 - loss 0.01959358 - time (sec): 90.52 - samples/sec: 1808.27 - lr: 0.000011 - momentum: 0.000000 2023-10-17 12:03:29,026 ---------------------------------------------------------------------------------------------------- 2023-10-17 12:03:29,027 EPOCH 8 done: loss 0.0196 - lr: 0.000011 2023-10-17 12:03:32,455 DEV : loss 0.23994652926921844 - f1-score (micro avg) 0.7582 2023-10-17 12:03:32,476 saving best model 2023-10-17 12:03:32,872 ---------------------------------------------------------------------------------------------------- 2023-10-17 12:03:42,075 epoch 9 - iter 198/1984 - loss 0.01369371 - time (sec): 9.20 - samples/sec: 1698.60 - lr: 0.000011 - momentum: 0.000000 2023-10-17 12:03:51,142 epoch 9 - iter 396/1984 - loss 0.01542831 - time (sec): 18.27 - samples/sec: 1744.84 - lr: 0.000010 - momentum: 0.000000 2023-10-17 12:04:00,413 epoch 9 - iter 594/1984 - loss 0.01701451 - time (sec): 27.54 - samples/sec: 1752.27 - lr: 0.000009 - momentum: 0.000000 2023-10-17 12:04:09,646 epoch 9 - iter 792/1984 - loss 0.01617942 - time (sec): 36.77 - samples/sec: 1769.71 - lr: 0.000009 - momentum: 0.000000 2023-10-17 12:04:18,868 epoch 9 - iter 990/1984 - loss 0.01513742 - time (sec): 45.99 - samples/sec: 1764.51 - lr: 0.000008 - momentum: 0.000000 2023-10-17 12:04:28,200 epoch 9 - iter 1188/1984 - loss 0.01474000 - time (sec): 55.33 - samples/sec: 1778.49 - lr: 0.000008 - momentum: 0.000000 2023-10-17 12:04:37,255 epoch 9 - iter 1386/1984 - loss 0.01453604 - time (sec): 64.38 - samples/sec: 1788.32 - lr: 0.000007 - momentum: 0.000000 2023-10-17 12:04:46,295 epoch 9 - iter 1584/1984 - loss 0.01380448 - time (sec): 73.42 - samples/sec: 1786.61 - lr: 0.000007 - momentum: 0.000000 2023-10-17 12:04:55,527 epoch 9 - iter 1782/1984 - loss 0.01369184 - time (sec): 82.65 - samples/sec: 1786.61 - lr: 0.000006 - momentum: 0.000000 2023-10-17 12:05:04,642 epoch 9 - iter 1980/1984 - loss 0.01333927 - time (sec): 91.77 - samples/sec: 1784.22 - lr: 0.000006 - momentum: 0.000000 2023-10-17 12:05:04,808 ---------------------------------------------------------------------------------------------------- 2023-10-17 12:05:04,808 EPOCH 9 done: loss 0.0133 - lr: 0.000006 2023-10-17 12:05:08,300 DEV : loss 0.25141069293022156 - f1-score (micro avg) 0.7495 2023-10-17 12:05:08,331 ---------------------------------------------------------------------------------------------------- 2023-10-17 12:05:19,095 epoch 10 - iter 198/1984 - loss 0.00791381 - time (sec): 10.76 - samples/sec: 1555.82 - lr: 0.000005 - momentum: 0.000000 2023-10-17 12:05:28,038 epoch 10 - iter 396/1984 - loss 0.00907342 - time (sec): 19.70 - samples/sec: 1675.12 - lr: 0.000004 - momentum: 0.000000 2023-10-17 12:05:37,094 epoch 10 - iter 594/1984 - loss 0.00841106 - time (sec): 28.76 - samples/sec: 1736.20 - lr: 0.000004 - momentum: 0.000000 2023-10-17 12:05:46,297 epoch 10 - iter 792/1984 - loss 0.00778224 - time (sec): 37.96 - samples/sec: 1732.33 - lr: 0.000003 - momentum: 0.000000 2023-10-17 12:05:55,632 epoch 10 - iter 990/1984 - loss 0.00790979 - time (sec): 47.30 - samples/sec: 1754.39 - lr: 0.000003 - momentum: 0.000000 2023-10-17 12:06:06,086 epoch 10 - iter 1188/1984 - loss 0.00810074 - time (sec): 57.75 - samples/sec: 1728.90 - lr: 0.000002 - momentum: 0.000000 2023-10-17 12:06:16,433 epoch 10 - iter 1386/1984 - loss 0.00812282 - time (sec): 68.10 - samples/sec: 1708.90 - lr: 0.000002 - momentum: 0.000000 2023-10-17 12:06:26,896 epoch 10 - iter 1584/1984 - loss 0.00827390 - time (sec): 78.56 - samples/sec: 1683.18 - lr: 0.000001 - momentum: 0.000000 2023-10-17 12:06:37,153 epoch 10 - iter 1782/1984 - loss 0.00865749 - time (sec): 88.82 - samples/sec: 1664.10 - lr: 0.000001 - momentum: 0.000000 2023-10-17 12:06:47,233 epoch 10 - iter 1980/1984 - loss 0.00828889 - time (sec): 98.90 - samples/sec: 1655.75 - lr: 0.000000 - momentum: 0.000000 2023-10-17 12:06:47,441 ---------------------------------------------------------------------------------------------------- 2023-10-17 12:06:47,441 EPOCH 10 done: loss 0.0083 - lr: 0.000000 2023-10-17 12:06:51,115 DEV : loss 0.2530412971973419 - f1-score (micro avg) 0.7575 2023-10-17 12:06:51,546 ---------------------------------------------------------------------------------------------------- 2023-10-17 12:06:51,547 Loading model from best epoch ... 2023-10-17 12:06:52,995 SequenceTagger predicts: Dictionary with 13 tags: O, S-PER, B-PER, E-PER, I-PER, S-LOC, B-LOC, E-LOC, I-LOC, S-ORG, B-ORG, E-ORG, I-ORG 2023-10-17 12:06:56,574 Results: - F-score (micro) 0.7556 - F-score (macro) 0.6677 - Accuracy 0.6401 By class: precision recall f1-score support LOC 0.8259 0.8183 0.8221 655 PER 0.6522 0.8072 0.7214 223 ORG 0.5000 0.4252 0.4596 127 micro avg 0.7454 0.7662 0.7556 1005 macro avg 0.6594 0.6836 0.6677 1005 weighted avg 0.7462 0.7662 0.7539 1005 2023-10-17 12:06:56,574 ----------------------------------------------------------------------------------------------------