2023-09-03 22:19:50,600 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:19:50,601 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() )" 2023-09-03 22:19:50,601 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:19:50,601 MultiCorpus: 3575 train + 1235 dev + 1266 test sentences - NER_HIPE_2022 Corpus: 3575 train + 1235 dev + 1266 test sentences - /app/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/de/with_doc_seperator 2023-09-03 22:19:50,601 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:19:50,601 Train: 3575 sentences 2023-09-03 22:19:50,601 (train_with_dev=False, train_with_test=False) 2023-09-03 22:19:50,601 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:19:50,601 Training Params: 2023-09-03 22:19:50,601 - learning_rate: "3e-05" 2023-09-03 22:19:50,601 - mini_batch_size: "4" 2023-09-03 22:19:50,601 - max_epochs: "10" 2023-09-03 22:19:50,601 - shuffle: "True" 2023-09-03 22:19:50,601 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:19:50,601 Plugins: 2023-09-03 22:19:50,601 - LinearScheduler | warmup_fraction: '0.1' 2023-09-03 22:19:50,601 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:19:50,602 Final evaluation on model from best epoch (best-model.pt) 2023-09-03 22:19:50,602 - metric: "('micro avg', 'f1-score')" 2023-09-03 22:19:50,602 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:19:50,602 Computation: 2023-09-03 22:19:50,602 - compute on device: cuda:0 2023-09-03 22:19:50,602 - embedding storage: none 2023-09-03 22:19:50,602 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:19:50,602 Model training base path: "hmbench-hipe2020/de-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4" 2023-09-03 22:19:50,602 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:19:50,602 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:19:59,927 epoch 1 - iter 89/894 - loss 2.99385041 - time (sec): 9.32 - samples/sec: 948.21 - lr: 0.000003 - momentum: 0.000000 2023-09-03 22:20:09,422 epoch 1 - iter 178/894 - loss 1.89257345 - time (sec): 18.82 - samples/sec: 960.08 - lr: 0.000006 - momentum: 0.000000 2023-09-03 22:20:18,371 epoch 1 - iter 267/894 - loss 1.45863122 - time (sec): 27.77 - samples/sec: 936.11 - lr: 0.000009 - momentum: 0.000000 2023-09-03 22:20:28,030 epoch 1 - iter 356/894 - loss 1.18119843 - time (sec): 37.43 - samples/sec: 933.06 - lr: 0.000012 - momentum: 0.000000 2023-09-03 22:20:37,044 epoch 1 - iter 445/894 - loss 1.01187653 - time (sec): 46.44 - samples/sec: 930.79 - lr: 0.000015 - momentum: 0.000000 2023-09-03 22:20:46,303 epoch 1 - iter 534/894 - loss 0.89949704 - time (sec): 55.70 - samples/sec: 927.61 - lr: 0.000018 - momentum: 0.000000 2023-09-03 22:20:55,424 epoch 1 - iter 623/894 - loss 0.81701336 - time (sec): 64.82 - samples/sec: 923.67 - lr: 0.000021 - momentum: 0.000000 2023-09-03 22:21:04,538 epoch 1 - iter 712/894 - loss 0.74798755 - time (sec): 73.94 - samples/sec: 926.04 - lr: 0.000024 - momentum: 0.000000 2023-09-03 22:21:13,595 epoch 1 - iter 801/894 - loss 0.69082416 - time (sec): 82.99 - samples/sec: 921.59 - lr: 0.000027 - momentum: 0.000000 2023-09-03 22:21:23,775 epoch 1 - iter 890/894 - loss 0.64011871 - time (sec): 93.17 - samples/sec: 925.00 - lr: 0.000030 - momentum: 0.000000 2023-09-03 22:21:24,173 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:21:24,173 EPOCH 1 done: loss 0.6378 - lr: 0.000030 2023-09-03 22:21:35,351 DEV : loss 0.19711647927761078 - f1-score (micro avg) 0.633 2023-09-03 22:21:35,379 saving best model 2023-09-03 22:21:35,850 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:21:45,484 epoch 2 - iter 89/894 - loss 0.20206718 - time (sec): 9.63 - samples/sec: 936.70 - lr: 0.000030 - momentum: 0.000000 2023-09-03 22:21:55,556 epoch 2 - iter 178/894 - loss 0.18754199 - time (sec): 19.70 - samples/sec: 944.01 - lr: 0.000029 - momentum: 0.000000 2023-09-03 22:22:04,434 epoch 2 - iter 267/894 - loss 0.17632130 - time (sec): 28.58 - samples/sec: 927.57 - lr: 0.000029 - momentum: 0.000000 2023-09-03 22:22:13,483 epoch 2 - iter 356/894 - loss 0.18038725 - time (sec): 37.63 - samples/sec: 925.66 - lr: 0.000029 - momentum: 0.000000 2023-09-03 22:22:22,982 epoch 2 - iter 445/894 - loss 0.17588838 - time (sec): 47.13 - samples/sec: 925.32 - lr: 0.000028 - momentum: 0.000000 2023-09-03 22:22:32,035 epoch 2 - iter 534/894 - loss 0.16862741 - time (sec): 56.18 - samples/sec: 931.12 - lr: 0.000028 - momentum: 0.000000 2023-09-03 22:22:40,942 epoch 2 - iter 623/894 - loss 0.16667206 - time (sec): 65.09 - samples/sec: 928.76 - lr: 0.000028 - momentum: 0.000000 2023-09-03 22:22:49,780 epoch 2 - iter 712/894 - loss 0.16552575 - time (sec): 73.93 - samples/sec: 928.53 - lr: 0.000027 - momentum: 0.000000 2023-09-03 22:22:59,680 epoch 2 - iter 801/894 - loss 0.16212332 - time (sec): 83.83 - samples/sec: 924.61 - lr: 0.000027 - momentum: 0.000000 2023-09-03 22:23:08,843 epoch 2 - iter 890/894 - loss 0.15955169 - time (sec): 92.99 - samples/sec: 927.09 - lr: 0.000027 - momentum: 0.000000 2023-09-03 22:23:09,253 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:23:09,253 EPOCH 2 done: loss 0.1594 - lr: 0.000027 2023-09-03 22:23:22,822 DEV : loss 0.14685824513435364 - f1-score (micro avg) 0.7162 2023-09-03 22:23:22,849 saving best model 2023-09-03 22:23:24,156 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:23:33,474 epoch 3 - iter 89/894 - loss 0.11785865 - time (sec): 9.32 - samples/sec: 930.51 - lr: 0.000026 - momentum: 0.000000 2023-09-03 22:23:43,422 epoch 3 - iter 178/894 - loss 0.10855285 - time (sec): 19.27 - samples/sec: 936.04 - lr: 0.000026 - momentum: 0.000000 2023-09-03 22:23:52,612 epoch 3 - iter 267/894 - loss 0.09664498 - time (sec): 28.45 - samples/sec: 929.61 - lr: 0.000026 - momentum: 0.000000 2023-09-03 22:24:01,721 epoch 3 - iter 356/894 - loss 0.09517122 - time (sec): 37.56 - samples/sec: 925.78 - lr: 0.000025 - momentum: 0.000000 2023-09-03 22:24:10,594 epoch 3 - iter 445/894 - loss 0.09228509 - time (sec): 46.44 - samples/sec: 916.67 - lr: 0.000025 - momentum: 0.000000 2023-09-03 22:24:19,755 epoch 3 - iter 534/894 - loss 0.09175790 - time (sec): 55.60 - samples/sec: 920.16 - lr: 0.000025 - momentum: 0.000000 2023-09-03 22:24:28,662 epoch 3 - iter 623/894 - loss 0.09474014 - time (sec): 64.50 - samples/sec: 920.19 - lr: 0.000024 - momentum: 0.000000 2023-09-03 22:24:38,233 epoch 3 - iter 712/894 - loss 0.09076201 - time (sec): 74.08 - samples/sec: 924.70 - lr: 0.000024 - momentum: 0.000000 2023-09-03 22:24:47,203 epoch 3 - iter 801/894 - loss 0.09299959 - time (sec): 83.05 - samples/sec: 924.12 - lr: 0.000024 - momentum: 0.000000 2023-09-03 22:24:57,206 epoch 3 - iter 890/894 - loss 0.09218085 - time (sec): 93.05 - samples/sec: 926.13 - lr: 0.000023 - momentum: 0.000000 2023-09-03 22:24:57,602 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:24:57,602 EPOCH 3 done: loss 0.0923 - lr: 0.000023 2023-09-03 22:25:11,191 DEV : loss 0.14347213506698608 - f1-score (micro avg) 0.7318 2023-09-03 22:25:11,218 saving best model 2023-09-03 22:25:12,561 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:25:21,779 epoch 4 - iter 89/894 - loss 0.05074308 - time (sec): 9.22 - samples/sec: 935.68 - lr: 0.000023 - momentum: 0.000000 2023-09-03 22:25:30,583 epoch 4 - iter 178/894 - loss 0.05957873 - time (sec): 18.02 - samples/sec: 921.32 - lr: 0.000023 - momentum: 0.000000 2023-09-03 22:25:39,872 epoch 4 - iter 267/894 - loss 0.05507104 - time (sec): 27.31 - samples/sec: 924.86 - lr: 0.000022 - momentum: 0.000000 2023-09-03 22:25:49,016 epoch 4 - iter 356/894 - loss 0.05116468 - time (sec): 36.45 - samples/sec: 934.55 - lr: 0.000022 - momentum: 0.000000 2023-09-03 22:25:59,254 epoch 4 - iter 445/894 - loss 0.05499857 - time (sec): 46.69 - samples/sec: 936.71 - lr: 0.000022 - momentum: 0.000000 2023-09-03 22:26:08,615 epoch 4 - iter 534/894 - loss 0.05429197 - time (sec): 56.05 - samples/sec: 936.56 - lr: 0.000021 - momentum: 0.000000 2023-09-03 22:26:17,448 epoch 4 - iter 623/894 - loss 0.05524917 - time (sec): 64.89 - samples/sec: 930.83 - lr: 0.000021 - momentum: 0.000000 2023-09-03 22:26:27,371 epoch 4 - iter 712/894 - loss 0.05641771 - time (sec): 74.81 - samples/sec: 928.13 - lr: 0.000021 - momentum: 0.000000 2023-09-03 22:26:36,995 epoch 4 - iter 801/894 - loss 0.05612467 - time (sec): 84.43 - samples/sec: 925.34 - lr: 0.000020 - momentum: 0.000000 2023-09-03 22:26:46,328 epoch 4 - iter 890/894 - loss 0.05680520 - time (sec): 93.77 - samples/sec: 918.98 - lr: 0.000020 - momentum: 0.000000 2023-09-03 22:26:46,692 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:26:46,692 EPOCH 4 done: loss 0.0567 - lr: 0.000020 2023-09-03 22:27:00,383 DEV : loss 0.1699320524930954 - f1-score (micro avg) 0.7621 2023-09-03 22:27:00,410 saving best model 2023-09-03 22:27:01,760 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:27:11,282 epoch 5 - iter 89/894 - loss 0.04947815 - time (sec): 9.52 - samples/sec: 951.53 - lr: 0.000020 - momentum: 0.000000 2023-09-03 22:27:20,518 epoch 5 - iter 178/894 - loss 0.04168584 - time (sec): 18.76 - samples/sec: 927.73 - lr: 0.000019 - momentum: 0.000000 2023-09-03 22:27:29,905 epoch 5 - iter 267/894 - loss 0.03998089 - time (sec): 28.14 - samples/sec: 934.01 - lr: 0.000019 - momentum: 0.000000 2023-09-03 22:27:39,202 epoch 5 - iter 356/894 - loss 0.04070305 - time (sec): 37.44 - samples/sec: 935.88 - lr: 0.000019 - momentum: 0.000000 2023-09-03 22:27:48,313 epoch 5 - iter 445/894 - loss 0.04091535 - time (sec): 46.55 - samples/sec: 929.95 - lr: 0.000018 - momentum: 0.000000 2023-09-03 22:27:57,683 epoch 5 - iter 534/894 - loss 0.04205066 - time (sec): 55.92 - samples/sec: 925.56 - lr: 0.000018 - momentum: 0.000000 2023-09-03 22:28:07,934 epoch 5 - iter 623/894 - loss 0.03983172 - time (sec): 66.17 - samples/sec: 925.70 - lr: 0.000018 - momentum: 0.000000 2023-09-03 22:28:16,745 epoch 5 - iter 712/894 - loss 0.04011488 - time (sec): 74.98 - samples/sec: 924.13 - lr: 0.000017 - momentum: 0.000000 2023-09-03 22:28:26,052 epoch 5 - iter 801/894 - loss 0.03892639 - time (sec): 84.29 - samples/sec: 921.73 - lr: 0.000017 - momentum: 0.000000 2023-09-03 22:28:35,323 epoch 5 - iter 890/894 - loss 0.03788280 - time (sec): 93.56 - samples/sec: 921.87 - lr: 0.000017 - momentum: 0.000000 2023-09-03 22:28:35,703 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:28:35,703 EPOCH 5 done: loss 0.0378 - lr: 0.000017 2023-09-03 22:28:49,309 DEV : loss 0.19927991926670074 - f1-score (micro avg) 0.7739 2023-09-03 22:28:49,335 saving best model 2023-09-03 22:28:50,661 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:28:59,979 epoch 6 - iter 89/894 - loss 0.01383422 - time (sec): 9.32 - samples/sec: 940.15 - lr: 0.000016 - momentum: 0.000000 2023-09-03 22:29:09,603 epoch 6 - iter 178/894 - loss 0.02356595 - time (sec): 18.94 - samples/sec: 938.76 - lr: 0.000016 - momentum: 0.000000 2023-09-03 22:29:18,543 epoch 6 - iter 267/894 - loss 0.02264157 - time (sec): 27.88 - samples/sec: 940.76 - lr: 0.000016 - momentum: 0.000000 2023-09-03 22:29:28,928 epoch 6 - iter 356/894 - loss 0.02218148 - time (sec): 38.27 - samples/sec: 946.57 - lr: 0.000015 - momentum: 0.000000 2023-09-03 22:29:38,030 epoch 6 - iter 445/894 - loss 0.02336928 - time (sec): 47.37 - samples/sec: 923.12 - lr: 0.000015 - momentum: 0.000000 2023-09-03 22:29:47,093 epoch 6 - iter 534/894 - loss 0.02274038 - time (sec): 56.43 - samples/sec: 922.21 - lr: 0.000015 - momentum: 0.000000 2023-09-03 22:29:56,477 epoch 6 - iter 623/894 - loss 0.02327204 - time (sec): 65.81 - samples/sec: 919.88 - lr: 0.000014 - momentum: 0.000000 2023-09-03 22:30:05,471 epoch 6 - iter 712/894 - loss 0.02357540 - time (sec): 74.81 - samples/sec: 917.58 - lr: 0.000014 - momentum: 0.000000 2023-09-03 22:30:14,730 epoch 6 - iter 801/894 - loss 0.02500668 - time (sec): 84.07 - samples/sec: 922.82 - lr: 0.000014 - momentum: 0.000000 2023-09-03 22:30:23,893 epoch 6 - iter 890/894 - loss 0.02557927 - time (sec): 93.23 - samples/sec: 924.82 - lr: 0.000013 - momentum: 0.000000 2023-09-03 22:30:24,265 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:30:24,265 EPOCH 6 done: loss 0.0256 - lr: 0.000013 2023-09-03 22:30:37,779 DEV : loss 0.2088565230369568 - f1-score (micro avg) 0.7648 2023-09-03 22:30:37,806 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:30:48,296 epoch 7 - iter 89/894 - loss 0.02360744 - time (sec): 10.49 - samples/sec: 954.74 - lr: 0.000013 - momentum: 0.000000 2023-09-03 22:30:57,383 epoch 7 - iter 178/894 - loss 0.01774078 - time (sec): 19.58 - samples/sec: 924.70 - lr: 0.000013 - momentum: 0.000000 2023-09-03 22:31:06,870 epoch 7 - iter 267/894 - loss 0.01531152 - time (sec): 29.06 - samples/sec: 927.50 - lr: 0.000012 - momentum: 0.000000 2023-09-03 22:31:16,280 epoch 7 - iter 356/894 - loss 0.01606620 - time (sec): 38.47 - samples/sec: 932.59 - lr: 0.000012 - momentum: 0.000000 2023-09-03 22:31:25,502 epoch 7 - iter 445/894 - loss 0.01680028 - time (sec): 47.70 - samples/sec: 931.42 - lr: 0.000012 - momentum: 0.000000 2023-09-03 22:31:34,440 epoch 7 - iter 534/894 - loss 0.01699701 - time (sec): 56.63 - samples/sec: 920.02 - lr: 0.000011 - momentum: 0.000000 2023-09-03 22:31:43,554 epoch 7 - iter 623/894 - loss 0.01643865 - time (sec): 65.75 - samples/sec: 925.67 - lr: 0.000011 - momentum: 0.000000 2023-09-03 22:31:52,707 epoch 7 - iter 712/894 - loss 0.01655476 - time (sec): 74.90 - samples/sec: 923.45 - lr: 0.000011 - momentum: 0.000000 2023-09-03 22:32:01,704 epoch 7 - iter 801/894 - loss 0.01718109 - time (sec): 83.90 - samples/sec: 923.77 - lr: 0.000010 - momentum: 0.000000 2023-09-03 22:32:10,864 epoch 7 - iter 890/894 - loss 0.01679469 - time (sec): 93.06 - samples/sec: 926.26 - lr: 0.000010 - momentum: 0.000000 2023-09-03 22:32:11,266 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:32:11,266 EPOCH 7 done: loss 0.0169 - lr: 0.000010 2023-09-03 22:32:24,795 DEV : loss 0.2267816662788391 - f1-score (micro avg) 0.7894 2023-09-03 22:32:24,822 saving best model 2023-09-03 22:32:26,160 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:32:35,446 epoch 8 - iter 89/894 - loss 0.00286651 - time (sec): 9.29 - samples/sec: 942.33 - lr: 0.000010 - momentum: 0.000000 2023-09-03 22:32:44,830 epoch 8 - iter 178/894 - loss 0.00447331 - time (sec): 18.67 - samples/sec: 925.26 - lr: 0.000009 - momentum: 0.000000 2023-09-03 22:32:53,920 epoch 8 - iter 267/894 - loss 0.00854419 - time (sec): 27.76 - samples/sec: 924.15 - lr: 0.000009 - momentum: 0.000000 2023-09-03 22:33:03,017 epoch 8 - iter 356/894 - loss 0.00931466 - time (sec): 36.86 - samples/sec: 918.72 - lr: 0.000009 - momentum: 0.000000 2023-09-03 22:33:12,472 epoch 8 - iter 445/894 - loss 0.00857502 - time (sec): 46.31 - samples/sec: 912.28 - lr: 0.000008 - momentum: 0.000000 2023-09-03 22:33:21,536 epoch 8 - iter 534/894 - loss 0.00924001 - time (sec): 55.37 - samples/sec: 913.74 - lr: 0.000008 - momentum: 0.000000 2023-09-03 22:33:30,700 epoch 8 - iter 623/894 - loss 0.00886184 - time (sec): 64.54 - samples/sec: 916.24 - lr: 0.000008 - momentum: 0.000000 2023-09-03 22:33:40,849 epoch 8 - iter 712/894 - loss 0.00994815 - time (sec): 74.69 - samples/sec: 914.76 - lr: 0.000007 - momentum: 0.000000 2023-09-03 22:33:50,725 epoch 8 - iter 801/894 - loss 0.01039865 - time (sec): 84.56 - samples/sec: 917.22 - lr: 0.000007 - momentum: 0.000000 2023-09-03 22:33:59,890 epoch 8 - iter 890/894 - loss 0.01002659 - time (sec): 93.73 - samples/sec: 920.48 - lr: 0.000007 - momentum: 0.000000 2023-09-03 22:34:00,256 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:34:00,256 EPOCH 8 done: loss 0.0100 - lr: 0.000007 2023-09-03 22:34:13,835 DEV : loss 0.22448322176933289 - f1-score (micro avg) 0.7826 2023-09-03 22:34:13,861 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:34:22,890 epoch 9 - iter 89/894 - loss 0.01763393 - time (sec): 9.03 - samples/sec: 928.84 - lr: 0.000006 - momentum: 0.000000 2023-09-03 22:34:32,548 epoch 9 - iter 178/894 - loss 0.01559000 - time (sec): 18.69 - samples/sec: 919.32 - lr: 0.000006 - momentum: 0.000000 2023-09-03 22:34:41,424 epoch 9 - iter 267/894 - loss 0.01468936 - time (sec): 27.56 - samples/sec: 915.11 - lr: 0.000006 - momentum: 0.000000 2023-09-03 22:34:50,641 epoch 9 - iter 356/894 - loss 0.01471083 - time (sec): 36.78 - samples/sec: 911.52 - lr: 0.000005 - momentum: 0.000000 2023-09-03 22:35:00,334 epoch 9 - iter 445/894 - loss 0.01203964 - time (sec): 46.47 - samples/sec: 915.58 - lr: 0.000005 - momentum: 0.000000 2023-09-03 22:35:09,740 epoch 9 - iter 534/894 - loss 0.01028989 - time (sec): 55.88 - samples/sec: 913.03 - lr: 0.000005 - momentum: 0.000000 2023-09-03 22:35:19,206 epoch 9 - iter 623/894 - loss 0.00998073 - time (sec): 65.34 - samples/sec: 920.57 - lr: 0.000004 - momentum: 0.000000 2023-09-03 22:35:29,444 epoch 9 - iter 712/894 - loss 0.00979519 - time (sec): 75.58 - samples/sec: 925.35 - lr: 0.000004 - momentum: 0.000000 2023-09-03 22:35:38,547 epoch 9 - iter 801/894 - loss 0.00944777 - time (sec): 84.68 - samples/sec: 922.13 - lr: 0.000004 - momentum: 0.000000 2023-09-03 22:35:47,583 epoch 9 - iter 890/894 - loss 0.00870663 - time (sec): 93.72 - samples/sec: 920.90 - lr: 0.000003 - momentum: 0.000000 2023-09-03 22:35:47,951 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:35:47,951 EPOCH 9 done: loss 0.0087 - lr: 0.000003 2023-09-03 22:36:02,217 DEV : loss 0.23423825204372406 - f1-score (micro avg) 0.7907 2023-09-03 22:36:02,244 saving best model 2023-09-03 22:36:03,595 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:36:13,930 epoch 10 - iter 89/894 - loss 0.00999648 - time (sec): 10.33 - samples/sec: 960.32 - lr: 0.000003 - momentum: 0.000000 2023-09-03 22:36:23,753 epoch 10 - iter 178/894 - loss 0.00735189 - time (sec): 20.16 - samples/sec: 929.52 - lr: 0.000003 - momentum: 0.000000 2023-09-03 22:36:32,949 epoch 10 - iter 267/894 - loss 0.00739292 - time (sec): 29.35 - samples/sec: 921.08 - lr: 0.000002 - momentum: 0.000000 2023-09-03 22:36:41,932 epoch 10 - iter 356/894 - loss 0.00680302 - time (sec): 38.33 - samples/sec: 916.98 - lr: 0.000002 - momentum: 0.000000 2023-09-03 22:36:51,269 epoch 10 - iter 445/894 - loss 0.00563722 - time (sec): 47.67 - samples/sec: 919.56 - lr: 0.000002 - momentum: 0.000000 2023-09-03 22:37:00,180 epoch 10 - iter 534/894 - loss 0.00581287 - time (sec): 56.58 - samples/sec: 917.60 - lr: 0.000001 - momentum: 0.000000 2023-09-03 22:37:09,404 epoch 10 - iter 623/894 - loss 0.00560941 - time (sec): 65.81 - samples/sec: 913.81 - lr: 0.000001 - momentum: 0.000000 2023-09-03 22:37:18,917 epoch 10 - iter 712/894 - loss 0.00496472 - time (sec): 75.32 - samples/sec: 915.04 - lr: 0.000001 - momentum: 0.000000 2023-09-03 22:37:27,962 epoch 10 - iter 801/894 - loss 0.00452758 - time (sec): 84.37 - samples/sec: 919.11 - lr: 0.000000 - momentum: 0.000000 2023-09-03 22:37:37,165 epoch 10 - iter 890/894 - loss 0.00451483 - time (sec): 93.57 - samples/sec: 921.41 - lr: 0.000000 - momentum: 0.000000 2023-09-03 22:37:37,576 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:37:37,576 EPOCH 10 done: loss 0.0045 - lr: 0.000000 2023-09-03 22:37:51,158 DEV : loss 0.23175767064094543 - f1-score (micro avg) 0.7889 2023-09-03 22:37:51,694 ---------------------------------------------------------------------------------------------------- 2023-09-03 22:37:51,696 Loading model from best epoch ... 2023-09-03 22:37:53,562 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-prod, B-prod, E-prod, I-prod, S-time, B-time, E-time, I-time 2023-09-03 22:38:04,339 Results: - F-score (micro) 0.7631 - F-score (macro) 0.6803 - Accuracy 0.6371 By class: precision recall f1-score support loc 0.8376 0.8742 0.8555 596 pers 0.6799 0.7718 0.7229 333 org 0.5766 0.4848 0.5267 132 prod 0.6481 0.5303 0.5833 66 time 0.6923 0.7347 0.7129 49 micro avg 0.7502 0.7764 0.7631 1176 macro avg 0.6869 0.6792 0.6803 1176 weighted avg 0.7470 0.7764 0.7598 1176 2023-09-03 22:38:04,339 ----------------------------------------------------------------------------------------------------