2023-10-16 17:53:10,355 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:53:10,356 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=17, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-16 17:53:10,356 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:53:10,356 MultiCorpus: 1166 train + 165 dev + 415 test sentences - NER_HIPE_2022 Corpus: 1166 train + 165 dev + 415 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fi/with_doc_seperator 2023-10-16 17:53:10,356 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:53:10,356 Train: 1166 sentences 2023-10-16 17:53:10,356 (train_with_dev=False, train_with_test=False) 2023-10-16 17:53:10,356 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:53:10,356 Training Params: 2023-10-16 17:53:10,356 - learning_rate: "3e-05" 2023-10-16 17:53:10,356 - mini_batch_size: "4" 2023-10-16 17:53:10,356 - max_epochs: "10" 2023-10-16 17:53:10,356 - shuffle: "True" 2023-10-16 17:53:10,356 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:53:10,356 Plugins: 2023-10-16 17:53:10,356 - LinearScheduler | warmup_fraction: '0.1' 2023-10-16 17:53:10,356 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:53:10,356 Final evaluation on model from best epoch (best-model.pt) 2023-10-16 17:53:10,356 - metric: "('micro avg', 'f1-score')" 2023-10-16 17:53:10,356 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:53:10,356 Computation: 2023-10-16 17:53:10,356 - compute on device: cuda:0 2023-10-16 17:53:10,356 - embedding storage: none 2023-10-16 17:53:10,356 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:53:10,356 Model training base path: "hmbench-newseye/fi-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-1" 2023-10-16 17:53:10,356 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:53:10,357 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:53:13,018 epoch 1 - iter 29/292 - loss 3.01670138 - time (sec): 2.66 - samples/sec: 1590.03 - lr: 0.000003 - momentum: 0.000000 2023-10-16 17:53:14,948 epoch 1 - iter 58/292 - loss 2.54580984 - time (sec): 4.59 - samples/sec: 2158.32 - lr: 0.000006 - momentum: 0.000000 2023-10-16 17:53:16,645 epoch 1 - iter 87/292 - loss 2.07822180 - time (sec): 6.29 - samples/sec: 2183.16 - lr: 0.000009 - momentum: 0.000000 2023-10-16 17:53:18,313 epoch 1 - iter 116/292 - loss 1.77670253 - time (sec): 7.96 - samples/sec: 2206.58 - lr: 0.000012 - momentum: 0.000000 2023-10-16 17:53:20,188 epoch 1 - iter 145/292 - loss 1.48355404 - time (sec): 9.83 - samples/sec: 2309.09 - lr: 0.000015 - momentum: 0.000000 2023-10-16 17:53:21,905 epoch 1 - iter 174/292 - loss 1.31635425 - time (sec): 11.55 - samples/sec: 2320.51 - lr: 0.000018 - momentum: 0.000000 2023-10-16 17:53:23,540 epoch 1 - iter 203/292 - loss 1.20996793 - time (sec): 13.18 - samples/sec: 2379.01 - lr: 0.000021 - momentum: 0.000000 2023-10-16 17:53:25,115 epoch 1 - iter 232/292 - loss 1.10738448 - time (sec): 14.76 - samples/sec: 2389.64 - lr: 0.000024 - momentum: 0.000000 2023-10-16 17:53:26,860 epoch 1 - iter 261/292 - loss 1.01693990 - time (sec): 16.50 - samples/sec: 2405.80 - lr: 0.000027 - momentum: 0.000000 2023-10-16 17:53:28,671 epoch 1 - iter 290/292 - loss 0.94433148 - time (sec): 18.31 - samples/sec: 2415.89 - lr: 0.000030 - momentum: 0.000000 2023-10-16 17:53:28,769 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:53:28,769 EPOCH 1 done: loss 0.9414 - lr: 0.000030 2023-10-16 17:53:29,849 DEV : loss 0.2120908498764038 - f1-score (micro avg) 0.4492 2023-10-16 17:53:29,853 saving best model 2023-10-16 17:53:30,340 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:53:32,028 epoch 2 - iter 29/292 - loss 0.18672523 - time (sec): 1.69 - samples/sec: 2637.53 - lr: 0.000030 - momentum: 0.000000 2023-10-16 17:53:33,606 epoch 2 - iter 58/292 - loss 0.22053055 - time (sec): 3.27 - samples/sec: 2719.56 - lr: 0.000029 - momentum: 0.000000 2023-10-16 17:53:35,118 epoch 2 - iter 87/292 - loss 0.23027135 - time (sec): 4.78 - samples/sec: 2749.72 - lr: 0.000029 - momentum: 0.000000 2023-10-16 17:53:36,675 epoch 2 - iter 116/292 - loss 0.20606513 - time (sec): 6.33 - samples/sec: 2752.78 - lr: 0.000029 - momentum: 0.000000 2023-10-16 17:53:38,235 epoch 2 - iter 145/292 - loss 0.20734738 - time (sec): 7.89 - samples/sec: 2671.83 - lr: 0.000028 - momentum: 0.000000 2023-10-16 17:53:39,802 epoch 2 - iter 174/292 - loss 0.20460916 - time (sec): 9.46 - samples/sec: 2668.26 - lr: 0.000028 - momentum: 0.000000 2023-10-16 17:53:41,754 epoch 2 - iter 203/292 - loss 0.21184347 - time (sec): 11.41 - samples/sec: 2680.86 - lr: 0.000028 - momentum: 0.000000 2023-10-16 17:53:43,491 epoch 2 - iter 232/292 - loss 0.20796678 - time (sec): 13.15 - samples/sec: 2667.97 - lr: 0.000027 - momentum: 0.000000 2023-10-16 17:53:45,327 epoch 2 - iter 261/292 - loss 0.20176642 - time (sec): 14.99 - samples/sec: 2667.45 - lr: 0.000027 - momentum: 0.000000 2023-10-16 17:53:46,957 epoch 2 - iter 290/292 - loss 0.19939226 - time (sec): 16.62 - samples/sec: 2657.51 - lr: 0.000027 - momentum: 0.000000 2023-10-16 17:53:47,055 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:53:47,056 EPOCH 2 done: loss 0.1997 - lr: 0.000027 2023-10-16 17:53:48,273 DEV : loss 0.13168571889400482 - f1-score (micro avg) 0.6189 2023-10-16 17:53:48,277 saving best model 2023-10-16 17:53:48,826 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:53:50,649 epoch 3 - iter 29/292 - loss 0.15474519 - time (sec): 1.82 - samples/sec: 3032.01 - lr: 0.000026 - momentum: 0.000000 2023-10-16 17:53:52,245 epoch 3 - iter 58/292 - loss 0.13264738 - time (sec): 3.41 - samples/sec: 2593.52 - lr: 0.000026 - momentum: 0.000000 2023-10-16 17:53:53,805 epoch 3 - iter 87/292 - loss 0.13055608 - time (sec): 4.97 - samples/sec: 2655.91 - lr: 0.000026 - momentum: 0.000000 2023-10-16 17:53:55,758 epoch 3 - iter 116/292 - loss 0.12609661 - time (sec): 6.93 - samples/sec: 2574.92 - lr: 0.000025 - momentum: 0.000000 2023-10-16 17:53:57,465 epoch 3 - iter 145/292 - loss 0.11999825 - time (sec): 8.63 - samples/sec: 2669.39 - lr: 0.000025 - momentum: 0.000000 2023-10-16 17:53:59,053 epoch 3 - iter 174/292 - loss 0.11593147 - time (sec): 10.22 - samples/sec: 2668.14 - lr: 0.000025 - momentum: 0.000000 2023-10-16 17:54:00,553 epoch 3 - iter 203/292 - loss 0.11595612 - time (sec): 11.72 - samples/sec: 2624.30 - lr: 0.000024 - momentum: 0.000000 2023-10-16 17:54:02,144 epoch 3 - iter 232/292 - loss 0.11529465 - time (sec): 13.31 - samples/sec: 2613.51 - lr: 0.000024 - momentum: 0.000000 2023-10-16 17:54:03,755 epoch 3 - iter 261/292 - loss 0.11497819 - time (sec): 14.92 - samples/sec: 2629.55 - lr: 0.000024 - momentum: 0.000000 2023-10-16 17:54:05,520 epoch 3 - iter 290/292 - loss 0.11247568 - time (sec): 16.69 - samples/sec: 2648.07 - lr: 0.000023 - momentum: 0.000000 2023-10-16 17:54:05,615 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:54:05,615 EPOCH 3 done: loss 0.1132 - lr: 0.000023 2023-10-16 17:54:06,825 DEV : loss 0.12188015133142471 - f1-score (micro avg) 0.7049 2023-10-16 17:54:06,829 saving best model 2023-10-16 17:54:07,368 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:54:09,077 epoch 4 - iter 29/292 - loss 0.07659714 - time (sec): 1.71 - samples/sec: 2724.79 - lr: 0.000023 - momentum: 0.000000 2023-10-16 17:54:10,758 epoch 4 - iter 58/292 - loss 0.08198423 - time (sec): 3.39 - samples/sec: 2706.29 - lr: 0.000023 - momentum: 0.000000 2023-10-16 17:54:12,342 epoch 4 - iter 87/292 - loss 0.07174156 - time (sec): 4.97 - samples/sec: 2704.35 - lr: 0.000022 - momentum: 0.000000 2023-10-16 17:54:14,000 epoch 4 - iter 116/292 - loss 0.07292734 - time (sec): 6.63 - samples/sec: 2706.79 - lr: 0.000022 - momentum: 0.000000 2023-10-16 17:54:15,587 epoch 4 - iter 145/292 - loss 0.07248743 - time (sec): 8.22 - samples/sec: 2666.65 - lr: 0.000022 - momentum: 0.000000 2023-10-16 17:54:17,256 epoch 4 - iter 174/292 - loss 0.07455389 - time (sec): 9.88 - samples/sec: 2672.34 - lr: 0.000021 - momentum: 0.000000 2023-10-16 17:54:18,864 epoch 4 - iter 203/292 - loss 0.07414268 - time (sec): 11.49 - samples/sec: 2648.26 - lr: 0.000021 - momentum: 0.000000 2023-10-16 17:54:20,561 epoch 4 - iter 232/292 - loss 0.07838926 - time (sec): 13.19 - samples/sec: 2687.21 - lr: 0.000021 - momentum: 0.000000 2023-10-16 17:54:22,083 epoch 4 - iter 261/292 - loss 0.07356868 - time (sec): 14.71 - samples/sec: 2688.42 - lr: 0.000020 - momentum: 0.000000 2023-10-16 17:54:23,717 epoch 4 - iter 290/292 - loss 0.07163331 - time (sec): 16.35 - samples/sec: 2709.19 - lr: 0.000020 - momentum: 0.000000 2023-10-16 17:54:23,801 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:54:23,801 EPOCH 4 done: loss 0.0721 - lr: 0.000020 2023-10-16 17:54:24,988 DEV : loss 0.11146893352270126 - f1-score (micro avg) 0.7448 2023-10-16 17:54:24,992 saving best model 2023-10-16 17:54:25,594 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:54:27,306 epoch 5 - iter 29/292 - loss 0.05024086 - time (sec): 1.71 - samples/sec: 2953.50 - lr: 0.000020 - momentum: 0.000000 2023-10-16 17:54:28,952 epoch 5 - iter 58/292 - loss 0.05493652 - time (sec): 3.36 - samples/sec: 2757.32 - lr: 0.000019 - momentum: 0.000000 2023-10-16 17:54:30,633 epoch 5 - iter 87/292 - loss 0.05660584 - time (sec): 5.04 - samples/sec: 2735.59 - lr: 0.000019 - momentum: 0.000000 2023-10-16 17:54:32,295 epoch 5 - iter 116/292 - loss 0.05164633 - time (sec): 6.70 - samples/sec: 2713.03 - lr: 0.000019 - momentum: 0.000000 2023-10-16 17:54:34,049 epoch 5 - iter 145/292 - loss 0.04760992 - time (sec): 8.45 - samples/sec: 2700.32 - lr: 0.000018 - momentum: 0.000000 2023-10-16 17:54:35,733 epoch 5 - iter 174/292 - loss 0.05141896 - time (sec): 10.14 - samples/sec: 2731.30 - lr: 0.000018 - momentum: 0.000000 2023-10-16 17:54:37,256 epoch 5 - iter 203/292 - loss 0.05190871 - time (sec): 11.66 - samples/sec: 2715.80 - lr: 0.000018 - momentum: 0.000000 2023-10-16 17:54:38,859 epoch 5 - iter 232/292 - loss 0.05203336 - time (sec): 13.26 - samples/sec: 2699.38 - lr: 0.000017 - momentum: 0.000000 2023-10-16 17:54:40,414 epoch 5 - iter 261/292 - loss 0.05031939 - time (sec): 14.82 - samples/sec: 2694.27 - lr: 0.000017 - momentum: 0.000000 2023-10-16 17:54:42,108 epoch 5 - iter 290/292 - loss 0.05098912 - time (sec): 16.51 - samples/sec: 2682.83 - lr: 0.000017 - momentum: 0.000000 2023-10-16 17:54:42,199 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:54:42,200 EPOCH 5 done: loss 0.0516 - lr: 0.000017 2023-10-16 17:54:43,417 DEV : loss 0.12832774221897125 - f1-score (micro avg) 0.73 2023-10-16 17:54:43,422 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:54:45,097 epoch 6 - iter 29/292 - loss 0.05950145 - time (sec): 1.67 - samples/sec: 2499.26 - lr: 0.000016 - momentum: 0.000000 2023-10-16 17:54:46,754 epoch 6 - iter 58/292 - loss 0.04628346 - time (sec): 3.33 - samples/sec: 2599.83 - lr: 0.000016 - momentum: 0.000000 2023-10-16 17:54:48,212 epoch 6 - iter 87/292 - loss 0.04890591 - time (sec): 4.79 - samples/sec: 2578.40 - lr: 0.000016 - momentum: 0.000000 2023-10-16 17:54:49,916 epoch 6 - iter 116/292 - loss 0.04392787 - time (sec): 6.49 - samples/sec: 2612.30 - lr: 0.000015 - momentum: 0.000000 2023-10-16 17:54:51,424 epoch 6 - iter 145/292 - loss 0.04177698 - time (sec): 8.00 - samples/sec: 2569.81 - lr: 0.000015 - momentum: 0.000000 2023-10-16 17:54:53,231 epoch 6 - iter 174/292 - loss 0.04185515 - time (sec): 9.81 - samples/sec: 2600.73 - lr: 0.000015 - momentum: 0.000000 2023-10-16 17:54:54,950 epoch 6 - iter 203/292 - loss 0.04627777 - time (sec): 11.53 - samples/sec: 2669.00 - lr: 0.000014 - momentum: 0.000000 2023-10-16 17:54:56,490 epoch 6 - iter 232/292 - loss 0.04466345 - time (sec): 13.07 - samples/sec: 2675.35 - lr: 0.000014 - momentum: 0.000000 2023-10-16 17:54:58,193 epoch 6 - iter 261/292 - loss 0.04324020 - time (sec): 14.77 - samples/sec: 2697.60 - lr: 0.000014 - momentum: 0.000000 2023-10-16 17:54:59,804 epoch 6 - iter 290/292 - loss 0.04093259 - time (sec): 16.38 - samples/sec: 2700.29 - lr: 0.000013 - momentum: 0.000000 2023-10-16 17:54:59,896 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:54:59,897 EPOCH 6 done: loss 0.0408 - lr: 0.000013 2023-10-16 17:55:01,125 DEV : loss 0.14976172149181366 - f1-score (micro avg) 0.74 2023-10-16 17:55:01,129 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:55:02,678 epoch 7 - iter 29/292 - loss 0.02344868 - time (sec): 1.55 - samples/sec: 2823.98 - lr: 0.000013 - momentum: 0.000000 2023-10-16 17:55:04,236 epoch 7 - iter 58/292 - loss 0.01915777 - time (sec): 3.11 - samples/sec: 2719.21 - lr: 0.000013 - momentum: 0.000000 2023-10-16 17:55:05,773 epoch 7 - iter 87/292 - loss 0.03000931 - time (sec): 4.64 - samples/sec: 2680.51 - lr: 0.000012 - momentum: 0.000000 2023-10-16 17:55:07,689 epoch 7 - iter 116/292 - loss 0.03479154 - time (sec): 6.56 - samples/sec: 2718.57 - lr: 0.000012 - momentum: 0.000000 2023-10-16 17:55:09,461 epoch 7 - iter 145/292 - loss 0.03189663 - time (sec): 8.33 - samples/sec: 2703.43 - lr: 0.000012 - momentum: 0.000000 2023-10-16 17:55:11,006 epoch 7 - iter 174/292 - loss 0.03017231 - time (sec): 9.88 - samples/sec: 2650.72 - lr: 0.000011 - momentum: 0.000000 2023-10-16 17:55:12,695 epoch 7 - iter 203/292 - loss 0.02853514 - time (sec): 11.56 - samples/sec: 2617.59 - lr: 0.000011 - momentum: 0.000000 2023-10-16 17:55:14,460 epoch 7 - iter 232/292 - loss 0.03262212 - time (sec): 13.33 - samples/sec: 2637.76 - lr: 0.000011 - momentum: 0.000000 2023-10-16 17:55:16,144 epoch 7 - iter 261/292 - loss 0.03145605 - time (sec): 15.01 - samples/sec: 2643.78 - lr: 0.000010 - momentum: 0.000000 2023-10-16 17:55:17,765 epoch 7 - iter 290/292 - loss 0.03161094 - time (sec): 16.63 - samples/sec: 2662.24 - lr: 0.000010 - momentum: 0.000000 2023-10-16 17:55:17,864 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:55:17,864 EPOCH 7 done: loss 0.0317 - lr: 0.000010 2023-10-16 17:55:19,109 DEV : loss 0.15802569687366486 - f1-score (micro avg) 0.7669 2023-10-16 17:55:19,115 saving best model 2023-10-16 17:55:19,736 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:55:21,251 epoch 8 - iter 29/292 - loss 0.02308704 - time (sec): 1.51 - samples/sec: 2423.33 - lr: 0.000010 - momentum: 0.000000 2023-10-16 17:55:23,003 epoch 8 - iter 58/292 - loss 0.01751107 - time (sec): 3.27 - samples/sec: 2645.43 - lr: 0.000009 - momentum: 0.000000 2023-10-16 17:55:24,636 epoch 8 - iter 87/292 - loss 0.01868607 - time (sec): 4.90 - samples/sec: 2645.19 - lr: 0.000009 - momentum: 0.000000 2023-10-16 17:55:26,347 epoch 8 - iter 116/292 - loss 0.01794405 - time (sec): 6.61 - samples/sec: 2669.32 - lr: 0.000009 - momentum: 0.000000 2023-10-16 17:55:28,103 epoch 8 - iter 145/292 - loss 0.02452178 - time (sec): 8.37 - samples/sec: 2670.62 - lr: 0.000008 - momentum: 0.000000 2023-10-16 17:55:29,621 epoch 8 - iter 174/292 - loss 0.02379010 - time (sec): 9.88 - samples/sec: 2633.50 - lr: 0.000008 - momentum: 0.000000 2023-10-16 17:55:31,201 epoch 8 - iter 203/292 - loss 0.02414199 - time (sec): 11.46 - samples/sec: 2615.66 - lr: 0.000008 - momentum: 0.000000 2023-10-16 17:55:32,900 epoch 8 - iter 232/292 - loss 0.02725617 - time (sec): 13.16 - samples/sec: 2616.60 - lr: 0.000007 - momentum: 0.000000 2023-10-16 17:55:34,529 epoch 8 - iter 261/292 - loss 0.02585742 - time (sec): 14.79 - samples/sec: 2621.20 - lr: 0.000007 - momentum: 0.000000 2023-10-16 17:55:36,363 epoch 8 - iter 290/292 - loss 0.02434377 - time (sec): 16.63 - samples/sec: 2657.48 - lr: 0.000007 - momentum: 0.000000 2023-10-16 17:55:36,485 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:55:36,485 EPOCH 8 done: loss 0.0242 - lr: 0.000007 2023-10-16 17:55:37,720 DEV : loss 0.16654258966445923 - f1-score (micro avg) 0.742 2023-10-16 17:55:37,725 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:55:39,403 epoch 9 - iter 29/292 - loss 0.01417823 - time (sec): 1.68 - samples/sec: 2715.08 - lr: 0.000006 - momentum: 0.000000 2023-10-16 17:55:41,051 epoch 9 - iter 58/292 - loss 0.01232230 - time (sec): 3.33 - samples/sec: 2616.24 - lr: 0.000006 - momentum: 0.000000 2023-10-16 17:55:42,622 epoch 9 - iter 87/292 - loss 0.01376009 - time (sec): 4.90 - samples/sec: 2518.21 - lr: 0.000006 - momentum: 0.000000 2023-10-16 17:55:44,385 epoch 9 - iter 116/292 - loss 0.02555969 - time (sec): 6.66 - samples/sec: 2645.87 - lr: 0.000005 - momentum: 0.000000 2023-10-16 17:55:46,074 epoch 9 - iter 145/292 - loss 0.02158313 - time (sec): 8.35 - samples/sec: 2653.79 - lr: 0.000005 - momentum: 0.000000 2023-10-16 17:55:47,637 epoch 9 - iter 174/292 - loss 0.02019502 - time (sec): 9.91 - samples/sec: 2641.31 - lr: 0.000005 - momentum: 0.000000 2023-10-16 17:55:49,335 epoch 9 - iter 203/292 - loss 0.02252725 - time (sec): 11.61 - samples/sec: 2620.54 - lr: 0.000004 - momentum: 0.000000 2023-10-16 17:55:50,943 epoch 9 - iter 232/292 - loss 0.02172381 - time (sec): 13.22 - samples/sec: 2623.35 - lr: 0.000004 - momentum: 0.000000 2023-10-16 17:55:52,771 epoch 9 - iter 261/292 - loss 0.01959173 - time (sec): 15.05 - samples/sec: 2637.08 - lr: 0.000004 - momentum: 0.000000 2023-10-16 17:55:54,506 epoch 9 - iter 290/292 - loss 0.01884339 - time (sec): 16.78 - samples/sec: 2641.24 - lr: 0.000003 - momentum: 0.000000 2023-10-16 17:55:54,588 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:55:54,588 EPOCH 9 done: loss 0.0188 - lr: 0.000003 2023-10-16 17:55:55,813 DEV : loss 0.1669674515724182 - f1-score (micro avg) 0.7447 2023-10-16 17:55:55,819 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:55:57,419 epoch 10 - iter 29/292 - loss 0.00817807 - time (sec): 1.60 - samples/sec: 2510.45 - lr: 0.000003 - momentum: 0.000000 2023-10-16 17:55:59,136 epoch 10 - iter 58/292 - loss 0.00727258 - time (sec): 3.32 - samples/sec: 2788.28 - lr: 0.000003 - momentum: 0.000000 2023-10-16 17:56:01,145 epoch 10 - iter 87/292 - loss 0.01631169 - time (sec): 5.33 - samples/sec: 2808.60 - lr: 0.000002 - momentum: 0.000000 2023-10-16 17:56:02,812 epoch 10 - iter 116/292 - loss 0.01409848 - time (sec): 6.99 - samples/sec: 2845.86 - lr: 0.000002 - momentum: 0.000000 2023-10-16 17:56:04,508 epoch 10 - iter 145/292 - loss 0.01324680 - time (sec): 8.69 - samples/sec: 2765.70 - lr: 0.000002 - momentum: 0.000000 2023-10-16 17:56:06,061 epoch 10 - iter 174/292 - loss 0.01866032 - time (sec): 10.24 - samples/sec: 2731.65 - lr: 0.000001 - momentum: 0.000000 2023-10-16 17:56:07,607 epoch 10 - iter 203/292 - loss 0.01831901 - time (sec): 11.79 - samples/sec: 2690.65 - lr: 0.000001 - momentum: 0.000000 2023-10-16 17:56:09,171 epoch 10 - iter 232/292 - loss 0.01714936 - time (sec): 13.35 - samples/sec: 2679.21 - lr: 0.000001 - momentum: 0.000000 2023-10-16 17:56:10,757 epoch 10 - iter 261/292 - loss 0.01673991 - time (sec): 14.94 - samples/sec: 2656.88 - lr: 0.000000 - momentum: 0.000000 2023-10-16 17:56:12,321 epoch 10 - iter 290/292 - loss 0.01661350 - time (sec): 16.50 - samples/sec: 2666.97 - lr: 0.000000 - momentum: 0.000000 2023-10-16 17:56:12,466 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:56:12,466 EPOCH 10 done: loss 0.0165 - lr: 0.000000 2023-10-16 17:56:13,697 DEV : loss 0.1702689379453659 - f1-score (micro avg) 0.7368 2023-10-16 17:56:14,198 ---------------------------------------------------------------------------------------------------- 2023-10-16 17:56:14,199 Loading model from best epoch ... 2023-10-16 17:56:16,142 SequenceTagger predicts: Dictionary with 17 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd 2023-10-16 17:56:18,744 Results: - F-score (micro) 0.7574 - F-score (macro) 0.6866 - Accuracy 0.6335 By class: precision recall f1-score support PER 0.7936 0.8506 0.8211 348 LOC 0.6677 0.8238 0.7376 261 ORG 0.4130 0.3654 0.3878 52 HumanProd 0.7826 0.8182 0.8000 22 micro avg 0.7173 0.8023 0.7574 683 macro avg 0.6642 0.7145 0.6866 683 weighted avg 0.7161 0.8023 0.7555 683 2023-10-16 17:56:18,744 ----------------------------------------------------------------------------------------------------