2023-09-04 12:41:05,500 ---------------------------------------------------------------------------------------------------- 2023-09-04 12:41:05,501 Model: "SequenceTagger( (embeddings): TransformerWordEmbeddings( (model): BertModel( (embeddings): BertEmbeddings( (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): BertEncoder( (layer): ModuleList( (0-11): 12 x BertLayer( (attention): BertAttention( (self): BertSelfAttention( (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): BertSelfOutput( (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): BertIntermediate( (dense): Linear(in_features=768, out_features=3072, bias=True) (intermediate_act_fn): GELUActivation() ) (output): BertOutput( (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) ) ) ) ) (pooler): BertPooler( (dense): Linear(in_features=768, out_features=768, bias=True) (activation): Tanh() ) ) ) (locked_dropout): LockedDropout(p=0.5) (linear): Linear(in_features=768, out_features=21, bias=True) (loss_function): CrossEntropyLoss() )" 2023-09-04 12:41:05,501 ---------------------------------------------------------------------------------------------------- 2023-09-04 12:41:05,501 MultiCorpus: 5901 train + 1287 dev + 1505 test sentences - NER_HIPE_2022 Corpus: 5901 train + 1287 dev + 1505 test sentences - /app/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/fr/with_doc_seperator 2023-09-04 12:41:05,501 ---------------------------------------------------------------------------------------------------- 2023-09-04 12:41:05,501 Train: 5901 sentences 2023-09-04 12:41:05,501 (train_with_dev=False, train_with_test=False) 2023-09-04 12:41:05,501 ---------------------------------------------------------------------------------------------------- 2023-09-04 12:41:05,501 Training Params: 2023-09-04 12:41:05,501 - learning_rate: "5e-05" 2023-09-04 12:41:05,501 - mini_batch_size: "4" 2023-09-04 12:41:05,502 - max_epochs: "10" 2023-09-04 12:41:05,502 - shuffle: "True" 2023-09-04 12:41:05,502 ---------------------------------------------------------------------------------------------------- 2023-09-04 12:41:05,502 Plugins: 2023-09-04 12:41:05,502 - LinearScheduler | warmup_fraction: '0.1' 2023-09-04 12:41:05,502 ---------------------------------------------------------------------------------------------------- 2023-09-04 12:41:05,502 Final evaluation on model from best epoch (best-model.pt) 2023-09-04 12:41:05,502 - metric: "('micro avg', 'f1-score')" 2023-09-04 12:41:05,502 ---------------------------------------------------------------------------------------------------- 2023-09-04 12:41:05,502 Computation: 2023-09-04 12:41:05,502 - compute on device: cuda:0 2023-09-04 12:41:05,502 - embedding storage: none 2023-09-04 12:41:05,502 ---------------------------------------------------------------------------------------------------- 2023-09-04 12:41:05,502 Model training base path: "hmbench-hipe2020/fr-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2" 2023-09-04 12:41:05,502 ---------------------------------------------------------------------------------------------------- 2023-09-04 12:41:05,502 ---------------------------------------------------------------------------------------------------- 2023-09-04 12:41:21,950 epoch 1 - iter 147/1476 - loss 2.22313355 - time (sec): 16.45 - samples/sec: 1076.86 - lr: 0.000005 - momentum: 0.000000 2023-09-04 12:41:38,671 epoch 1 - iter 294/1476 - loss 1.37374230 - time (sec): 33.17 - samples/sec: 1083.45 - lr: 0.000010 - momentum: 0.000000 2023-09-04 12:41:54,462 epoch 1 - iter 441/1476 - loss 1.05288085 - time (sec): 48.96 - samples/sec: 1061.77 - lr: 0.000015 - momentum: 0.000000 2023-09-04 12:42:10,552 epoch 1 - iter 588/1476 - loss 0.86208252 - time (sec): 65.05 - samples/sec: 1056.52 - lr: 0.000020 - momentum: 0.000000 2023-09-04 12:42:26,859 epoch 1 - iter 735/1476 - loss 0.74659437 - time (sec): 81.36 - samples/sec: 1051.97 - lr: 0.000025 - momentum: 0.000000 2023-09-04 12:42:42,698 epoch 1 - iter 882/1476 - loss 0.66213367 - time (sec): 97.19 - samples/sec: 1052.44 - lr: 0.000030 - momentum: 0.000000 2023-09-04 12:42:58,644 epoch 1 - iter 1029/1476 - loss 0.59907497 - time (sec): 113.14 - samples/sec: 1048.26 - lr: 0.000035 - momentum: 0.000000 2023-09-04 12:43:13,763 epoch 1 - iter 1176/1476 - loss 0.55542267 - time (sec): 128.26 - samples/sec: 1039.70 - lr: 0.000040 - momentum: 0.000000 2023-09-04 12:43:29,571 epoch 1 - iter 1323/1476 - loss 0.51603021 - time (sec): 144.07 - samples/sec: 1038.88 - lr: 0.000045 - momentum: 0.000000 2023-09-04 12:43:45,182 epoch 1 - iter 1470/1476 - loss 0.48435086 - time (sec): 159.68 - samples/sec: 1038.74 - lr: 0.000050 - momentum: 0.000000 2023-09-04 12:43:45,717 ---------------------------------------------------------------------------------------------------- 2023-09-04 12:43:45,717 EPOCH 1 done: loss 0.4835 - lr: 0.000050 2023-09-04 12:44:00,022 DEV : loss 0.12973235547542572 - f1-score (micro avg) 0.72 2023-09-04 12:44:00,050 saving best model 2023-09-04 12:44:00,523 ---------------------------------------------------------------------------------------------------- 2023-09-04 12:44:15,272 epoch 2 - iter 147/1476 - loss 0.13043275 - time (sec): 14.75 - samples/sec: 1032.13 - lr: 0.000049 - momentum: 0.000000 2023-09-04 12:44:30,831 epoch 2 - iter 294/1476 - loss 0.13817560 - time (sec): 30.31 - samples/sec: 1036.87 - lr: 0.000049 - momentum: 0.000000 2023-09-04 12:44:46,416 epoch 2 - iter 441/1476 - loss 0.13982802 - time (sec): 45.89 - samples/sec: 1040.96 - lr: 0.000048 - momentum: 0.000000 2023-09-04 12:45:02,296 epoch 2 - iter 588/1476 - loss 0.13827499 - time (sec): 61.77 - samples/sec: 1032.24 - lr: 0.000048 - momentum: 0.000000 2023-09-04 12:45:17,688 epoch 2 - iter 735/1476 - loss 0.14321867 - time (sec): 77.16 - samples/sec: 1023.62 - lr: 0.000047 - momentum: 0.000000 2023-09-04 12:45:34,057 epoch 2 - iter 882/1476 - loss 0.14413832 - time (sec): 93.53 - samples/sec: 1027.10 - lr: 0.000047 - momentum: 0.000000 2023-09-04 12:45:51,209 epoch 2 - iter 1029/1476 - loss 0.13943760 - time (sec): 110.69 - samples/sec: 1034.12 - lr: 0.000046 - momentum: 0.000000 2023-09-04 12:46:06,983 epoch 2 - iter 1176/1476 - loss 0.13462370 - time (sec): 126.46 - samples/sec: 1035.36 - lr: 0.000046 - momentum: 0.000000 2023-09-04 12:46:23,330 epoch 2 - iter 1323/1476 - loss 0.13833582 - time (sec): 142.81 - samples/sec: 1036.67 - lr: 0.000045 - momentum: 0.000000 2023-09-04 12:46:40,221 epoch 2 - iter 1470/1476 - loss 0.13957066 - time (sec): 159.70 - samples/sec: 1037.02 - lr: 0.000044 - momentum: 0.000000 2023-09-04 12:46:40,879 ---------------------------------------------------------------------------------------------------- 2023-09-04 12:46:40,879 EPOCH 2 done: loss 0.1397 - lr: 0.000044 2023-09-04 12:46:58,622 DEV : loss 0.13705389201641083 - f1-score (micro avg) 0.7321 2023-09-04 12:46:58,651 saving best model 2023-09-04 12:46:59,986 ---------------------------------------------------------------------------------------------------- 2023-09-04 12:47:15,499 epoch 3 - iter 147/1476 - loss 0.07281328 - time (sec): 15.51 - samples/sec: 1000.37 - lr: 0.000044 - momentum: 0.000000 2023-09-04 12:47:31,881 epoch 3 - iter 294/1476 - loss 0.08498610 - time (sec): 31.89 - samples/sec: 1014.38 - lr: 0.000043 - momentum: 0.000000 2023-09-04 12:47:47,727 epoch 3 - iter 441/1476 - loss 0.09120280 - time (sec): 47.74 - samples/sec: 1018.43 - lr: 0.000043 - momentum: 0.000000 2023-09-04 12:48:02,893 epoch 3 - iter 588/1476 - loss 0.08916578 - time (sec): 62.91 - samples/sec: 1019.25 - lr: 0.000042 - momentum: 0.000000 2023-09-04 12:48:19,680 epoch 3 - iter 735/1476 - loss 0.09238527 - time (sec): 79.69 - samples/sec: 1023.49 - lr: 0.000042 - momentum: 0.000000 2023-09-04 12:48:36,558 epoch 3 - iter 882/1476 - loss 0.09045454 - time (sec): 96.57 - samples/sec: 1036.22 - lr: 0.000041 - momentum: 0.000000 2023-09-04 12:48:52,241 epoch 3 - iter 1029/1476 - loss 0.08793924 - time (sec): 112.25 - samples/sec: 1034.49 - lr: 0.000041 - momentum: 0.000000 2023-09-04 12:49:08,955 epoch 3 - iter 1176/1476 - loss 0.09159878 - time (sec): 128.97 - samples/sec: 1038.96 - lr: 0.000040 - momentum: 0.000000 2023-09-04 12:49:24,711 epoch 3 - iter 1323/1476 - loss 0.09066876 - time (sec): 144.72 - samples/sec: 1038.19 - lr: 0.000039 - momentum: 0.000000 2023-09-04 12:49:40,188 epoch 3 - iter 1470/1476 - loss 0.09316735 - time (sec): 160.20 - samples/sec: 1035.06 - lr: 0.000039 - momentum: 0.000000 2023-09-04 12:49:40,758 ---------------------------------------------------------------------------------------------------- 2023-09-04 12:49:40,758 EPOCH 3 done: loss 0.0931 - lr: 0.000039 2023-09-04 12:49:58,561 DEV : loss 0.1488744020462036 - f1-score (micro avg) 0.7835 2023-09-04 12:49:58,602 saving best model 2023-09-04 12:49:59,944 ---------------------------------------------------------------------------------------------------- 2023-09-04 12:50:15,389 epoch 4 - iter 147/1476 - loss 0.05437294 - time (sec): 15.44 - samples/sec: 986.70 - lr: 0.000038 - momentum: 0.000000 2023-09-04 12:50:30,759 epoch 4 - iter 294/1476 - loss 0.05695284 - time (sec): 30.81 - samples/sec: 1009.35 - lr: 0.000038 - momentum: 0.000000 2023-09-04 12:50:46,761 epoch 4 - iter 441/1476 - loss 0.06071564 - time (sec): 46.82 - samples/sec: 1021.33 - lr: 0.000037 - momentum: 0.000000 2023-09-04 12:51:02,028 epoch 4 - iter 588/1476 - loss 0.06089200 - time (sec): 62.08 - samples/sec: 1018.38 - lr: 0.000037 - momentum: 0.000000 2023-09-04 12:51:18,219 epoch 4 - iter 735/1476 - loss 0.06175334 - time (sec): 78.27 - samples/sec: 1020.96 - lr: 0.000036 - momentum: 0.000000 2023-09-04 12:51:35,354 epoch 4 - iter 882/1476 - loss 0.06353415 - time (sec): 95.41 - samples/sec: 1021.01 - lr: 0.000036 - momentum: 0.000000 2023-09-04 12:51:52,839 epoch 4 - iter 1029/1476 - loss 0.06422628 - time (sec): 112.89 - samples/sec: 1030.76 - lr: 0.000035 - momentum: 0.000000 2023-09-04 12:52:08,729 epoch 4 - iter 1176/1476 - loss 0.06363828 - time (sec): 128.78 - samples/sec: 1032.44 - lr: 0.000034 - momentum: 0.000000 2023-09-04 12:52:25,193 epoch 4 - iter 1323/1476 - loss 0.06681448 - time (sec): 145.25 - samples/sec: 1030.85 - lr: 0.000034 - momentum: 0.000000 2023-09-04 12:52:40,592 epoch 4 - iter 1470/1476 - loss 0.06664854 - time (sec): 160.65 - samples/sec: 1031.61 - lr: 0.000033 - momentum: 0.000000 2023-09-04 12:52:41,271 ---------------------------------------------------------------------------------------------------- 2023-09-04 12:52:41,271 EPOCH 4 done: loss 0.0666 - lr: 0.000033 2023-09-04 12:52:59,139 DEV : loss 0.18947024643421173 - f1-score (micro avg) 0.7886 2023-09-04 12:52:59,167 saving best model 2023-09-04 12:53:00,515 ---------------------------------------------------------------------------------------------------- 2023-09-04 12:53:16,577 epoch 5 - iter 147/1476 - loss 0.04297170 - time (sec): 16.06 - samples/sec: 1030.92 - lr: 0.000033 - momentum: 0.000000 2023-09-04 12:53:32,604 epoch 5 - iter 294/1476 - loss 0.04431455 - time (sec): 32.09 - samples/sec: 1037.33 - lr: 0.000032 - momentum: 0.000000 2023-09-04 12:53:48,491 epoch 5 - iter 441/1476 - loss 0.04484576 - time (sec): 47.97 - samples/sec: 1049.34 - lr: 0.000032 - momentum: 0.000000 2023-09-04 12:54:04,010 epoch 5 - iter 588/1476 - loss 0.04376185 - time (sec): 63.49 - samples/sec: 1045.83 - lr: 0.000031 - momentum: 0.000000 2023-09-04 12:54:19,680 epoch 5 - iter 735/1476 - loss 0.04368058 - time (sec): 79.16 - samples/sec: 1042.72 - lr: 0.000031 - momentum: 0.000000 2023-09-04 12:54:35,189 epoch 5 - iter 882/1476 - loss 0.04539124 - time (sec): 94.67 - samples/sec: 1035.86 - lr: 0.000030 - momentum: 0.000000 2023-09-04 12:54:51,200 epoch 5 - iter 1029/1476 - loss 0.04406403 - time (sec): 110.68 - samples/sec: 1033.53 - lr: 0.000029 - momentum: 0.000000 2023-09-04 12:55:08,769 epoch 5 - iter 1176/1476 - loss 0.04473760 - time (sec): 128.25 - samples/sec: 1035.75 - lr: 0.000029 - momentum: 0.000000 2023-09-04 12:55:25,502 epoch 5 - iter 1323/1476 - loss 0.04353630 - time (sec): 144.99 - samples/sec: 1039.51 - lr: 0.000028 - momentum: 0.000000 2023-09-04 12:55:40,746 epoch 5 - iter 1470/1476 - loss 0.04533154 - time (sec): 160.23 - samples/sec: 1035.10 - lr: 0.000028 - momentum: 0.000000 2023-09-04 12:55:41,340 ---------------------------------------------------------------------------------------------------- 2023-09-04 12:55:41,340 EPOCH 5 done: loss 0.0452 - lr: 0.000028 2023-09-04 12:55:59,294 DEV : loss 0.19925478100776672 - f1-score (micro avg) 0.7925 2023-09-04 12:55:59,323 saving best model 2023-09-04 12:56:00,672 ---------------------------------------------------------------------------------------------------- 2023-09-04 12:56:15,887 epoch 6 - iter 147/1476 - loss 0.03566906 - time (sec): 15.21 - samples/sec: 981.26 - lr: 0.000027 - momentum: 0.000000 2023-09-04 12:56:33,489 epoch 6 - iter 294/1476 - loss 0.03245913 - time (sec): 32.82 - samples/sec: 1040.29 - lr: 0.000027 - momentum: 0.000000 2023-09-04 12:56:49,628 epoch 6 - iter 441/1476 - loss 0.03754273 - time (sec): 48.95 - samples/sec: 1042.98 - lr: 0.000026 - momentum: 0.000000 2023-09-04 12:57:05,764 epoch 6 - iter 588/1476 - loss 0.04214083 - time (sec): 65.09 - samples/sec: 1035.81 - lr: 0.000026 - momentum: 0.000000 2023-09-04 12:57:22,270 epoch 6 - iter 735/1476 - loss 0.03917241 - time (sec): 81.60 - samples/sec: 1040.84 - lr: 0.000025 - momentum: 0.000000 2023-09-04 12:57:38,968 epoch 6 - iter 882/1476 - loss 0.03938033 - time (sec): 98.29 - samples/sec: 1039.46 - lr: 0.000024 - momentum: 0.000000 2023-09-04 12:57:53,807 epoch 6 - iter 1029/1476 - loss 0.03736740 - time (sec): 113.13 - samples/sec: 1037.66 - lr: 0.000024 - momentum: 0.000000 2023-09-04 12:58:09,831 epoch 6 - iter 1176/1476 - loss 0.03518827 - time (sec): 129.16 - samples/sec: 1037.13 - lr: 0.000023 - momentum: 0.000000 2023-09-04 12:58:25,528 epoch 6 - iter 1323/1476 - loss 0.03450183 - time (sec): 144.85 - samples/sec: 1034.15 - lr: 0.000023 - momentum: 0.000000 2023-09-04 12:58:41,496 epoch 6 - iter 1470/1476 - loss 0.03399637 - time (sec): 160.82 - samples/sec: 1032.01 - lr: 0.000022 - momentum: 0.000000 2023-09-04 12:58:42,028 ---------------------------------------------------------------------------------------------------- 2023-09-04 12:58:42,029 EPOCH 6 done: loss 0.0339 - lr: 0.000022 2023-09-04 12:58:59,939 DEV : loss 0.2285047173500061 - f1-score (micro avg) 0.7934 2023-09-04 12:58:59,968 saving best model 2023-09-04 12:59:01,300 ---------------------------------------------------------------------------------------------------- 2023-09-04 12:59:18,284 epoch 7 - iter 147/1476 - loss 0.02070811 - time (sec): 16.98 - samples/sec: 1007.35 - lr: 0.000022 - momentum: 0.000000 2023-09-04 12:59:33,102 epoch 7 - iter 294/1476 - loss 0.01999558 - time (sec): 31.80 - samples/sec: 1010.62 - lr: 0.000021 - momentum: 0.000000 2023-09-04 12:59:50,186 epoch 7 - iter 441/1476 - loss 0.02505522 - time (sec): 48.88 - samples/sec: 1034.16 - lr: 0.000021 - momentum: 0.000000 2023-09-04 13:00:07,527 epoch 7 - iter 588/1476 - loss 0.02337657 - time (sec): 66.23 - samples/sec: 1046.75 - lr: 0.000020 - momentum: 0.000000 2023-09-04 13:00:22,653 epoch 7 - iter 735/1476 - loss 0.02335444 - time (sec): 81.35 - samples/sec: 1037.46 - lr: 0.000019 - momentum: 0.000000 2023-09-04 13:00:37,830 epoch 7 - iter 882/1476 - loss 0.02581999 - time (sec): 96.53 - samples/sec: 1033.52 - lr: 0.000019 - momentum: 0.000000 2023-09-04 13:00:53,075 epoch 7 - iter 1029/1476 - loss 0.02632303 - time (sec): 111.77 - samples/sec: 1036.57 - lr: 0.000018 - momentum: 0.000000 2023-09-04 13:01:08,477 epoch 7 - iter 1176/1476 - loss 0.02644004 - time (sec): 127.18 - samples/sec: 1034.32 - lr: 0.000018 - momentum: 0.000000 2023-09-04 13:01:24,297 epoch 7 - iter 1323/1476 - loss 0.02509027 - time (sec): 143.00 - samples/sec: 1032.18 - lr: 0.000017 - momentum: 0.000000 2023-09-04 13:01:41,491 epoch 7 - iter 1470/1476 - loss 0.02464981 - time (sec): 160.19 - samples/sec: 1035.62 - lr: 0.000017 - momentum: 0.000000 2023-09-04 13:01:42,039 ---------------------------------------------------------------------------------------------------- 2023-09-04 13:01:42,040 EPOCH 7 done: loss 0.0246 - lr: 0.000017 2023-09-04 13:01:59,891 DEV : loss 0.2234429568052292 - f1-score (micro avg) 0.8009 2023-09-04 13:01:59,920 saving best model 2023-09-04 13:02:01,248 ---------------------------------------------------------------------------------------------------- 2023-09-04 13:02:17,045 epoch 8 - iter 147/1476 - loss 0.02104970 - time (sec): 15.80 - samples/sec: 1063.41 - lr: 0.000016 - momentum: 0.000000 2023-09-04 13:02:33,539 epoch 8 - iter 294/1476 - loss 0.01905570 - time (sec): 32.29 - samples/sec: 1045.70 - lr: 0.000016 - momentum: 0.000000 2023-09-04 13:02:51,720 epoch 8 - iter 441/1476 - loss 0.02143358 - time (sec): 50.47 - samples/sec: 1053.59 - lr: 0.000015 - momentum: 0.000000 2023-09-04 13:03:06,995 epoch 8 - iter 588/1476 - loss 0.02318846 - time (sec): 65.75 - samples/sec: 1034.19 - lr: 0.000014 - momentum: 0.000000 2023-09-04 13:03:22,725 epoch 8 - iter 735/1476 - loss 0.02159453 - time (sec): 81.48 - samples/sec: 1028.31 - lr: 0.000014 - momentum: 0.000000 2023-09-04 13:03:37,894 epoch 8 - iter 882/1476 - loss 0.02077588 - time (sec): 96.64 - samples/sec: 1026.41 - lr: 0.000013 - momentum: 0.000000 2023-09-04 13:03:53,907 epoch 8 - iter 1029/1476 - loss 0.01921124 - time (sec): 112.66 - samples/sec: 1022.41 - lr: 0.000013 - momentum: 0.000000 2023-09-04 13:04:08,936 epoch 8 - iter 1176/1476 - loss 0.01898697 - time (sec): 127.69 - samples/sec: 1020.65 - lr: 0.000012 - momentum: 0.000000 2023-09-04 13:04:25,679 epoch 8 - iter 1323/1476 - loss 0.01816890 - time (sec): 144.43 - samples/sec: 1025.22 - lr: 0.000012 - momentum: 0.000000 2023-09-04 13:04:42,147 epoch 8 - iter 1470/1476 - loss 0.01707363 - time (sec): 160.90 - samples/sec: 1030.37 - lr: 0.000011 - momentum: 0.000000 2023-09-04 13:04:42,746 ---------------------------------------------------------------------------------------------------- 2023-09-04 13:04:42,746 EPOCH 8 done: loss 0.0171 - lr: 0.000011 2023-09-04 13:05:00,631 DEV : loss 0.2465333193540573 - f1-score (micro avg) 0.8171 2023-09-04 13:05:00,660 saving best model 2023-09-04 13:05:01,997 ---------------------------------------------------------------------------------------------------- 2023-09-04 13:05:17,734 epoch 9 - iter 147/1476 - loss 0.00882656 - time (sec): 15.74 - samples/sec: 989.49 - lr: 0.000011 - momentum: 0.000000 2023-09-04 13:05:34,580 epoch 9 - iter 294/1476 - loss 0.01220668 - time (sec): 32.58 - samples/sec: 1011.69 - lr: 0.000010 - momentum: 0.000000 2023-09-04 13:05:49,519 epoch 9 - iter 441/1476 - loss 0.01041928 - time (sec): 47.52 - samples/sec: 1022.24 - lr: 0.000009 - momentum: 0.000000 2023-09-04 13:06:05,393 epoch 9 - iter 588/1476 - loss 0.01239141 - time (sec): 63.39 - samples/sec: 1029.38 - lr: 0.000009 - momentum: 0.000000 2023-09-04 13:06:22,377 epoch 9 - iter 735/1476 - loss 0.01291391 - time (sec): 80.38 - samples/sec: 1036.93 - lr: 0.000008 - momentum: 0.000000 2023-09-04 13:06:37,918 epoch 9 - iter 882/1476 - loss 0.01164990 - time (sec): 95.92 - samples/sec: 1031.89 - lr: 0.000008 - momentum: 0.000000 2023-09-04 13:06:54,500 epoch 9 - iter 1029/1476 - loss 0.01012837 - time (sec): 112.50 - samples/sec: 1031.72 - lr: 0.000007 - momentum: 0.000000 2023-09-04 13:07:10,209 epoch 9 - iter 1176/1476 - loss 0.01002218 - time (sec): 128.21 - samples/sec: 1026.68 - lr: 0.000007 - momentum: 0.000000 2023-09-04 13:07:25,675 epoch 9 - iter 1323/1476 - loss 0.00974478 - time (sec): 143.68 - samples/sec: 1030.62 - lr: 0.000006 - momentum: 0.000000 2023-09-04 13:07:42,531 epoch 9 - iter 1470/1476 - loss 0.01086558 - time (sec): 160.53 - samples/sec: 1031.34 - lr: 0.000006 - momentum: 0.000000 2023-09-04 13:07:43,222 ---------------------------------------------------------------------------------------------------- 2023-09-04 13:07:43,222 EPOCH 9 done: loss 0.0111 - lr: 0.000006 2023-09-04 13:08:01,008 DEV : loss 0.23992028832435608 - f1-score (micro avg) 0.8026 2023-09-04 13:08:01,037 ---------------------------------------------------------------------------------------------------- 2023-09-04 13:08:17,870 epoch 10 - iter 147/1476 - loss 0.00738594 - time (sec): 16.83 - samples/sec: 1054.01 - lr: 0.000005 - momentum: 0.000000 2023-09-04 13:08:33,223 epoch 10 - iter 294/1476 - loss 0.00717515 - time (sec): 32.19 - samples/sec: 1046.63 - lr: 0.000004 - momentum: 0.000000 2023-09-04 13:08:47,936 epoch 10 - iter 441/1476 - loss 0.00881687 - time (sec): 46.90 - samples/sec: 1050.10 - lr: 0.000004 - momentum: 0.000000 2023-09-04 13:09:03,864 epoch 10 - iter 588/1476 - loss 0.00781440 - time (sec): 62.83 - samples/sec: 1046.86 - lr: 0.000003 - momentum: 0.000000 2023-09-04 13:09:19,628 epoch 10 - iter 735/1476 - loss 0.00754620 - time (sec): 78.59 - samples/sec: 1041.24 - lr: 0.000003 - momentum: 0.000000 2023-09-04 13:09:37,111 epoch 10 - iter 882/1476 - loss 0.00777399 - time (sec): 96.07 - samples/sec: 1048.14 - lr: 0.000002 - momentum: 0.000000 2023-09-04 13:09:52,104 epoch 10 - iter 1029/1476 - loss 0.00746587 - time (sec): 111.07 - samples/sec: 1039.50 - lr: 0.000002 - momentum: 0.000000 2023-09-04 13:10:08,750 epoch 10 - iter 1176/1476 - loss 0.00697918 - time (sec): 127.71 - samples/sec: 1037.41 - lr: 0.000001 - momentum: 0.000000 2023-09-04 13:10:25,067 epoch 10 - iter 1323/1476 - loss 0.00814534 - time (sec): 144.03 - samples/sec: 1037.56 - lr: 0.000001 - momentum: 0.000000 2023-09-04 13:10:41,038 epoch 10 - iter 1470/1476 - loss 0.00775839 - time (sec): 160.00 - samples/sec: 1036.73 - lr: 0.000000 - momentum: 0.000000 2023-09-04 13:10:41,631 ---------------------------------------------------------------------------------------------------- 2023-09-04 13:10:41,631 EPOCH 10 done: loss 0.0077 - lr: 0.000000 2023-09-04 13:10:59,474 DEV : loss 0.2502734065055847 - f1-score (micro avg) 0.807 2023-09-04 13:11:00,040 ---------------------------------------------------------------------------------------------------- 2023-09-04 13:11:00,042 Loading model from best epoch ... 2023-09-04 13:11:01,899 SequenceTagger predicts: Dictionary with 21 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, S-time, B-time, E-time, I-time, S-prod, B-prod, E-prod, I-prod 2023-09-04 13:11:16,645 Results: - F-score (micro) 0.7859 - F-score (macro) 0.6745 - Accuracy 0.6703 By class: precision recall f1-score support loc 0.8637 0.8566 0.8602 858 pers 0.7371 0.7989 0.7668 537 org 0.5739 0.5000 0.5344 132 prod 0.6032 0.6230 0.6129 61 time 0.5556 0.6481 0.5983 54 micro avg 0.7784 0.7935 0.7859 1642 macro avg 0.6667 0.6853 0.6745 1642 weighted avg 0.7792 0.7935 0.7856 1642 2023-09-04 13:11:16,645 ----------------------------------------------------------------------------------------------------