2023-10-17 18:01:41,762 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:01:41,763 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=17, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-17 18:01:41,763 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:01:41,763 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 2023-10-17 18:01:41,763 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:01:41,763 Train: 1166 sentences 2023-10-17 18:01:41,763 (train_with_dev=False, train_with_test=False) 2023-10-17 18:01:41,763 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:01:41,763 Training Params: 2023-10-17 18:01:41,763 - learning_rate: "3e-05" 2023-10-17 18:01:41,763 - mini_batch_size: "8" 2023-10-17 18:01:41,763 - max_epochs: "10" 2023-10-17 18:01:41,763 - shuffle: "True" 2023-10-17 18:01:41,763 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:01:41,763 Plugins: 2023-10-17 18:01:41,763 - TensorboardLogger 2023-10-17 18:01:41,763 - LinearScheduler | warmup_fraction: '0.1' 2023-10-17 18:01:41,763 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:01:41,764 Final evaluation on model from best epoch (best-model.pt) 2023-10-17 18:01:41,764 - metric: "('micro avg', 'f1-score')" 2023-10-17 18:01:41,764 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:01:41,764 Computation: 2023-10-17 18:01:41,764 - compute on device: cuda:0 2023-10-17 18:01:41,764 - embedding storage: none 2023-10-17 18:01:41,764 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:01:41,764 Model training base path: "hmbench-newseye/fi-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3" 2023-10-17 18:01:41,764 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:01:41,764 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:01:41,764 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-17 18:01:43,042 epoch 1 - iter 14/146 - loss 3.53043795 - time (sec): 1.28 - samples/sec: 2887.22 - lr: 0.000003 - momentum: 0.000000 2023-10-17 18:01:44,750 epoch 1 - iter 28/146 - loss 3.22220354 - time (sec): 2.99 - samples/sec: 2903.33 - lr: 0.000006 - momentum: 0.000000 2023-10-17 18:01:46,108 epoch 1 - iter 42/146 - loss 2.82585575 - time (sec): 4.34 - samples/sec: 2974.54 - lr: 0.000008 - momentum: 0.000000 2023-10-17 18:01:47,961 epoch 1 - iter 56/146 - loss 2.33875546 - time (sec): 6.20 - samples/sec: 2863.59 - lr: 0.000011 - momentum: 0.000000 2023-10-17 18:01:49,464 epoch 1 - iter 70/146 - loss 1.98561751 - time (sec): 7.70 - samples/sec: 2859.02 - lr: 0.000014 - momentum: 0.000000 2023-10-17 18:01:51,151 epoch 1 - iter 84/146 - loss 1.73604633 - time (sec): 9.39 - samples/sec: 2822.89 - lr: 0.000017 - momentum: 0.000000 2023-10-17 18:01:52,455 epoch 1 - iter 98/146 - loss 1.56244089 - time (sec): 10.69 - samples/sec: 2869.67 - lr: 0.000020 - momentum: 0.000000 2023-10-17 18:01:53,840 epoch 1 - iter 112/146 - loss 1.45136493 - time (sec): 12.07 - samples/sec: 2856.69 - lr: 0.000023 - momentum: 0.000000 2023-10-17 18:01:55,351 epoch 1 - iter 126/146 - loss 1.32971185 - time (sec): 13.59 - samples/sec: 2849.01 - lr: 0.000026 - momentum: 0.000000 2023-10-17 18:01:56,627 epoch 1 - iter 140/146 - loss 1.24009769 - time (sec): 14.86 - samples/sec: 2874.39 - lr: 0.000029 - momentum: 0.000000 2023-10-17 18:01:57,281 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:01:57,282 EPOCH 1 done: loss 1.2042 - lr: 0.000029 2023-10-17 18:01:58,453 DEV : loss 0.2369639128446579 - f1-score (micro avg) 0.3429 2023-10-17 18:01:58,459 saving best model 2023-10-17 18:01:58,784 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:02:00,122 epoch 2 - iter 14/146 - loss 0.29832645 - time (sec): 1.34 - samples/sec: 3233.20 - lr: 0.000030 - momentum: 0.000000 2023-10-17 18:02:01,215 epoch 2 - iter 28/146 - loss 0.27090210 - time (sec): 2.43 - samples/sec: 3144.62 - lr: 0.000029 - momentum: 0.000000 2023-10-17 18:02:02,734 epoch 2 - iter 42/146 - loss 0.24774915 - time (sec): 3.95 - samples/sec: 3163.37 - lr: 0.000029 - momentum: 0.000000 2023-10-17 18:02:04,573 epoch 2 - iter 56/146 - loss 0.24935139 - time (sec): 5.79 - samples/sec: 3031.59 - lr: 0.000029 - momentum: 0.000000 2023-10-17 18:02:05,827 epoch 2 - iter 70/146 - loss 0.24406299 - time (sec): 7.04 - samples/sec: 3021.97 - lr: 0.000028 - momentum: 0.000000 2023-10-17 18:02:07,349 epoch 2 - iter 84/146 - loss 0.23566930 - time (sec): 8.56 - samples/sec: 2980.19 - lr: 0.000028 - momentum: 0.000000 2023-10-17 18:02:08,992 epoch 2 - iter 98/146 - loss 0.22785994 - time (sec): 10.21 - samples/sec: 2986.86 - lr: 0.000028 - momentum: 0.000000 2023-10-17 18:02:10,368 epoch 2 - iter 112/146 - loss 0.22262977 - time (sec): 11.58 - samples/sec: 2997.58 - lr: 0.000027 - momentum: 0.000000 2023-10-17 18:02:11,870 epoch 2 - iter 126/146 - loss 0.22621003 - time (sec): 13.08 - samples/sec: 3001.66 - lr: 0.000027 - momentum: 0.000000 2023-10-17 18:02:13,129 epoch 2 - iter 140/146 - loss 0.22404336 - time (sec): 14.34 - samples/sec: 2985.96 - lr: 0.000027 - momentum: 0.000000 2023-10-17 18:02:13,639 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:02:13,639 EPOCH 2 done: loss 0.2206 - lr: 0.000027 2023-10-17 18:02:14,920 DEV : loss 0.1299794614315033 - f1-score (micro avg) 0.5919 2023-10-17 18:02:14,926 saving best model 2023-10-17 18:02:15,371 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:02:16,875 epoch 3 - iter 14/146 - loss 0.12717687 - time (sec): 1.50 - samples/sec: 3121.96 - lr: 0.000026 - momentum: 0.000000 2023-10-17 18:02:18,340 epoch 3 - iter 28/146 - loss 0.13016870 - time (sec): 2.97 - samples/sec: 3101.19 - lr: 0.000026 - momentum: 0.000000 2023-10-17 18:02:19,758 epoch 3 - iter 42/146 - loss 0.14516907 - time (sec): 4.39 - samples/sec: 3022.70 - lr: 0.000026 - momentum: 0.000000 2023-10-17 18:02:21,273 epoch 3 - iter 56/146 - loss 0.13349387 - time (sec): 5.90 - samples/sec: 2944.83 - lr: 0.000025 - momentum: 0.000000 2023-10-17 18:02:22,750 epoch 3 - iter 70/146 - loss 0.12864265 - time (sec): 7.38 - samples/sec: 2953.89 - lr: 0.000025 - momentum: 0.000000 2023-10-17 18:02:24,312 epoch 3 - iter 84/146 - loss 0.13230498 - time (sec): 8.94 - samples/sec: 2942.26 - lr: 0.000025 - momentum: 0.000000 2023-10-17 18:02:25,476 epoch 3 - iter 98/146 - loss 0.12924129 - time (sec): 10.10 - samples/sec: 2969.36 - lr: 0.000024 - momentum: 0.000000 2023-10-17 18:02:27,038 epoch 3 - iter 112/146 - loss 0.12219386 - time (sec): 11.66 - samples/sec: 2977.73 - lr: 0.000024 - momentum: 0.000000 2023-10-17 18:02:28,564 epoch 3 - iter 126/146 - loss 0.12044637 - time (sec): 13.19 - samples/sec: 2959.39 - lr: 0.000024 - momentum: 0.000000 2023-10-17 18:02:30,111 epoch 3 - iter 140/146 - loss 0.12007643 - time (sec): 14.74 - samples/sec: 2913.51 - lr: 0.000024 - momentum: 0.000000 2023-10-17 18:02:30,652 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:02:30,652 EPOCH 3 done: loss 0.1233 - lr: 0.000024 2023-10-17 18:02:32,289 DEV : loss 0.10894083976745605 - f1-score (micro avg) 0.7187 2023-10-17 18:02:32,299 saving best model 2023-10-17 18:02:32,723 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:02:34,231 epoch 4 - iter 14/146 - loss 0.07416896 - time (sec): 1.51 - samples/sec: 3010.11 - lr: 0.000023 - momentum: 0.000000 2023-10-17 18:02:35,856 epoch 4 - iter 28/146 - loss 0.07141417 - time (sec): 3.13 - samples/sec: 2907.94 - lr: 0.000023 - momentum: 0.000000 2023-10-17 18:02:37,152 epoch 4 - iter 42/146 - loss 0.08970989 - time (sec): 4.43 - samples/sec: 2908.80 - lr: 0.000022 - momentum: 0.000000 2023-10-17 18:02:38,867 epoch 4 - iter 56/146 - loss 0.07963099 - time (sec): 6.14 - samples/sec: 2874.00 - lr: 0.000022 - momentum: 0.000000 2023-10-17 18:02:40,463 epoch 4 - iter 70/146 - loss 0.07752524 - time (sec): 7.74 - samples/sec: 2873.54 - lr: 0.000022 - momentum: 0.000000 2023-10-17 18:02:41,955 epoch 4 - iter 84/146 - loss 0.08042752 - time (sec): 9.23 - samples/sec: 2870.29 - lr: 0.000021 - momentum: 0.000000 2023-10-17 18:02:43,558 epoch 4 - iter 98/146 - loss 0.08106679 - time (sec): 10.83 - samples/sec: 2818.94 - lr: 0.000021 - momentum: 0.000000 2023-10-17 18:02:45,073 epoch 4 - iter 112/146 - loss 0.08397046 - time (sec): 12.35 - samples/sec: 2831.20 - lr: 0.000021 - momentum: 0.000000 2023-10-17 18:02:46,566 epoch 4 - iter 126/146 - loss 0.08434407 - time (sec): 13.84 - samples/sec: 2828.54 - lr: 0.000021 - momentum: 0.000000 2023-10-17 18:02:47,841 epoch 4 - iter 140/146 - loss 0.08375380 - time (sec): 15.12 - samples/sec: 2812.99 - lr: 0.000020 - momentum: 0.000000 2023-10-17 18:02:48,440 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:02:48,440 EPOCH 4 done: loss 0.0837 - lr: 0.000020 2023-10-17 18:02:49,738 DEV : loss 0.11354158073663712 - f1-score (micro avg) 0.7364 2023-10-17 18:02:49,744 saving best model 2023-10-17 18:02:50,202 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:02:51,702 epoch 5 - iter 14/146 - loss 0.05382494 - time (sec): 1.50 - samples/sec: 2810.34 - lr: 0.000020 - momentum: 0.000000 2023-10-17 18:02:53,009 epoch 5 - iter 28/146 - loss 0.06071119 - time (sec): 2.80 - samples/sec: 2969.28 - lr: 0.000019 - momentum: 0.000000 2023-10-17 18:02:54,527 epoch 5 - iter 42/146 - loss 0.05523519 - time (sec): 4.32 - samples/sec: 3056.07 - lr: 0.000019 - momentum: 0.000000 2023-10-17 18:02:55,986 epoch 5 - iter 56/146 - loss 0.05750211 - time (sec): 5.78 - samples/sec: 2996.72 - lr: 0.000019 - momentum: 0.000000 2023-10-17 18:02:57,546 epoch 5 - iter 70/146 - loss 0.06410589 - time (sec): 7.34 - samples/sec: 2872.91 - lr: 0.000018 - momentum: 0.000000 2023-10-17 18:02:59,083 epoch 5 - iter 84/146 - loss 0.06215775 - time (sec): 8.88 - samples/sec: 2907.99 - lr: 0.000018 - momentum: 0.000000 2023-10-17 18:03:00,391 epoch 5 - iter 98/146 - loss 0.05920684 - time (sec): 10.18 - samples/sec: 2912.85 - lr: 0.000018 - momentum: 0.000000 2023-10-17 18:03:01,906 epoch 5 - iter 112/146 - loss 0.05729003 - time (sec): 11.70 - samples/sec: 2882.89 - lr: 0.000018 - momentum: 0.000000 2023-10-17 18:03:03,453 epoch 5 - iter 126/146 - loss 0.05501850 - time (sec): 13.25 - samples/sec: 2905.99 - lr: 0.000017 - momentum: 0.000000 2023-10-17 18:03:05,100 epoch 5 - iter 140/146 - loss 0.05545585 - time (sec): 14.89 - samples/sec: 2890.13 - lr: 0.000017 - momentum: 0.000000 2023-10-17 18:03:05,589 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:03:05,589 EPOCH 5 done: loss 0.0549 - lr: 0.000017 2023-10-17 18:03:06,918 DEV : loss 0.11634857952594757 - f1-score (micro avg) 0.7364 2023-10-17 18:03:06,925 saving best model 2023-10-17 18:03:07,379 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:03:08,799 epoch 6 - iter 14/146 - loss 0.04238142 - time (sec): 1.41 - samples/sec: 2913.49 - lr: 0.000016 - momentum: 0.000000 2023-10-17 18:03:10,367 epoch 6 - iter 28/146 - loss 0.03655978 - time (sec): 2.98 - samples/sec: 2978.87 - lr: 0.000016 - momentum: 0.000000 2023-10-17 18:03:11,678 epoch 6 - iter 42/146 - loss 0.04211034 - time (sec): 4.29 - samples/sec: 2890.83 - lr: 0.000016 - momentum: 0.000000 2023-10-17 18:03:13,075 epoch 6 - iter 56/146 - loss 0.04107637 - time (sec): 5.69 - samples/sec: 2815.81 - lr: 0.000015 - momentum: 0.000000 2023-10-17 18:03:14,598 epoch 6 - iter 70/146 - loss 0.03835445 - time (sec): 7.21 - samples/sec: 2834.14 - lr: 0.000015 - momentum: 0.000000 2023-10-17 18:03:16,185 epoch 6 - iter 84/146 - loss 0.04034492 - time (sec): 8.80 - samples/sec: 2893.99 - lr: 0.000015 - momentum: 0.000000 2023-10-17 18:03:17,290 epoch 6 - iter 98/146 - loss 0.03978206 - time (sec): 9.90 - samples/sec: 2918.91 - lr: 0.000015 - momentum: 0.000000 2023-10-17 18:03:18,683 epoch 6 - iter 112/146 - loss 0.04031278 - time (sec): 11.30 - samples/sec: 2921.99 - lr: 0.000014 - momentum: 0.000000 2023-10-17 18:03:20,095 epoch 6 - iter 126/146 - loss 0.04005854 - time (sec): 12.71 - samples/sec: 2955.67 - lr: 0.000014 - momentum: 0.000000 2023-10-17 18:03:21,608 epoch 6 - iter 140/146 - loss 0.03903056 - time (sec): 14.22 - samples/sec: 2987.59 - lr: 0.000014 - momentum: 0.000000 2023-10-17 18:03:22,326 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:03:22,326 EPOCH 6 done: loss 0.0384 - lr: 0.000014 2023-10-17 18:03:23,864 DEV : loss 0.11619787663221359 - f1-score (micro avg) 0.7478 2023-10-17 18:03:23,869 saving best model 2023-10-17 18:03:24,312 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:03:25,733 epoch 7 - iter 14/146 - loss 0.03971526 - time (sec): 1.41 - samples/sec: 2868.48 - lr: 0.000013 - momentum: 0.000000 2023-10-17 18:03:26,974 epoch 7 - iter 28/146 - loss 0.04113722 - time (sec): 2.65 - samples/sec: 2856.67 - lr: 0.000013 - momentum: 0.000000 2023-10-17 18:03:28,350 epoch 7 - iter 42/146 - loss 0.03581246 - time (sec): 4.03 - samples/sec: 2894.28 - lr: 0.000012 - momentum: 0.000000 2023-10-17 18:03:29,725 epoch 7 - iter 56/146 - loss 0.03125811 - time (sec): 5.41 - samples/sec: 2955.57 - lr: 0.000012 - momentum: 0.000000 2023-10-17 18:03:31,234 epoch 7 - iter 70/146 - loss 0.03390483 - time (sec): 6.92 - samples/sec: 2998.62 - lr: 0.000012 - momentum: 0.000000 2023-10-17 18:03:32,727 epoch 7 - iter 84/146 - loss 0.03434154 - time (sec): 8.41 - samples/sec: 2922.72 - lr: 0.000012 - momentum: 0.000000 2023-10-17 18:03:34,202 epoch 7 - iter 98/146 - loss 0.03222323 - time (sec): 9.88 - samples/sec: 2921.38 - lr: 0.000011 - momentum: 0.000000 2023-10-17 18:03:35,723 epoch 7 - iter 112/146 - loss 0.03097942 - time (sec): 11.40 - samples/sec: 2894.68 - lr: 0.000011 - momentum: 0.000000 2023-10-17 18:03:37,396 epoch 7 - iter 126/146 - loss 0.03133453 - time (sec): 13.08 - samples/sec: 2874.13 - lr: 0.000011 - momentum: 0.000000 2023-10-17 18:03:38,840 epoch 7 - iter 140/146 - loss 0.02975401 - time (sec): 14.52 - samples/sec: 2914.63 - lr: 0.000010 - momentum: 0.000000 2023-10-17 18:03:39,546 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:03:39,547 EPOCH 7 done: loss 0.0289 - lr: 0.000010 2023-10-17 18:03:40,808 DEV : loss 0.12387975305318832 - f1-score (micro avg) 0.7793 2023-10-17 18:03:40,813 saving best model 2023-10-17 18:03:41,234 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:03:42,601 epoch 8 - iter 14/146 - loss 0.02000203 - time (sec): 1.37 - samples/sec: 3044.44 - lr: 0.000010 - momentum: 0.000000 2023-10-17 18:03:44,171 epoch 8 - iter 28/146 - loss 0.01956968 - time (sec): 2.94 - samples/sec: 2897.32 - lr: 0.000009 - momentum: 0.000000 2023-10-17 18:03:45,590 epoch 8 - iter 42/146 - loss 0.01928448 - time (sec): 4.35 - samples/sec: 2916.95 - lr: 0.000009 - momentum: 0.000000 2023-10-17 18:03:47,084 epoch 8 - iter 56/146 - loss 0.02405329 - time (sec): 5.85 - samples/sec: 2973.71 - lr: 0.000009 - momentum: 0.000000 2023-10-17 18:03:48,535 epoch 8 - iter 70/146 - loss 0.02559984 - time (sec): 7.30 - samples/sec: 2977.54 - lr: 0.000009 - momentum: 0.000000 2023-10-17 18:03:49,980 epoch 8 - iter 84/146 - loss 0.02430293 - time (sec): 8.74 - samples/sec: 2997.30 - lr: 0.000008 - momentum: 0.000000 2023-10-17 18:03:51,652 epoch 8 - iter 98/146 - loss 0.02276596 - time (sec): 10.42 - samples/sec: 2971.35 - lr: 0.000008 - momentum: 0.000000 2023-10-17 18:03:52,912 epoch 8 - iter 112/146 - loss 0.02276090 - time (sec): 11.68 - samples/sec: 2946.39 - lr: 0.000008 - momentum: 0.000000 2023-10-17 18:03:54,513 epoch 8 - iter 126/146 - loss 0.02259700 - time (sec): 13.28 - samples/sec: 2960.07 - lr: 0.000007 - momentum: 0.000000 2023-10-17 18:03:55,891 epoch 8 - iter 140/146 - loss 0.02211640 - time (sec): 14.66 - samples/sec: 2938.42 - lr: 0.000007 - momentum: 0.000000 2023-10-17 18:03:56,387 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:03:56,387 EPOCH 8 done: loss 0.0219 - lr: 0.000007 2023-10-17 18:03:57,620 DEV : loss 0.12991590797901154 - f1-score (micro avg) 0.7723 2023-10-17 18:03:57,625 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:03:58,998 epoch 9 - iter 14/146 - loss 0.02121112 - time (sec): 1.37 - samples/sec: 2778.46 - lr: 0.000006 - momentum: 0.000000 2023-10-17 18:04:00,753 epoch 9 - iter 28/146 - loss 0.02062558 - time (sec): 3.13 - samples/sec: 2796.60 - lr: 0.000006 - momentum: 0.000000 2023-10-17 18:04:02,551 epoch 9 - iter 42/146 - loss 0.02599310 - time (sec): 4.92 - samples/sec: 2863.48 - lr: 0.000006 - momentum: 0.000000 2023-10-17 18:04:03,939 epoch 9 - iter 56/146 - loss 0.02190832 - time (sec): 6.31 - samples/sec: 2853.47 - lr: 0.000006 - momentum: 0.000000 2023-10-17 18:04:05,179 epoch 9 - iter 70/146 - loss 0.02164107 - time (sec): 7.55 - samples/sec: 2878.04 - lr: 0.000005 - momentum: 0.000000 2023-10-17 18:04:06,666 epoch 9 - iter 84/146 - loss 0.02212824 - time (sec): 9.04 - samples/sec: 2896.08 - lr: 0.000005 - momentum: 0.000000 2023-10-17 18:04:08,036 epoch 9 - iter 98/146 - loss 0.02065961 - time (sec): 10.41 - samples/sec: 2874.83 - lr: 0.000005 - momentum: 0.000000 2023-10-17 18:04:09,463 epoch 9 - iter 112/146 - loss 0.02049381 - time (sec): 11.84 - samples/sec: 2930.78 - lr: 0.000004 - momentum: 0.000000 2023-10-17 18:04:10,874 epoch 9 - iter 126/146 - loss 0.01964867 - time (sec): 13.25 - samples/sec: 2891.82 - lr: 0.000004 - momentum: 0.000000 2023-10-17 18:04:12,171 epoch 9 - iter 140/146 - loss 0.01888946 - time (sec): 14.54 - samples/sec: 2895.64 - lr: 0.000004 - momentum: 0.000000 2023-10-17 18:04:12,832 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:04:12,832 EPOCH 9 done: loss 0.0180 - lr: 0.000004 2023-10-17 18:04:14,602 DEV : loss 0.1360742151737213 - f1-score (micro avg) 0.7604 2023-10-17 18:04:14,609 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:04:16,465 epoch 10 - iter 14/146 - loss 0.02531276 - time (sec): 1.85 - samples/sec: 2708.78 - lr: 0.000003 - momentum: 0.000000 2023-10-17 18:04:18,092 epoch 10 - iter 28/146 - loss 0.01920590 - time (sec): 3.48 - samples/sec: 2742.03 - lr: 0.000003 - momentum: 0.000000 2023-10-17 18:04:19,521 epoch 10 - iter 42/146 - loss 0.01738091 - time (sec): 4.91 - samples/sec: 2769.85 - lr: 0.000003 - momentum: 0.000000 2023-10-17 18:04:20,878 epoch 10 - iter 56/146 - loss 0.01529737 - time (sec): 6.27 - samples/sec: 2803.00 - lr: 0.000002 - momentum: 0.000000 2023-10-17 18:04:22,307 epoch 10 - iter 70/146 - loss 0.01440336 - time (sec): 7.70 - samples/sec: 2783.17 - lr: 0.000002 - momentum: 0.000000 2023-10-17 18:04:23,876 epoch 10 - iter 84/146 - loss 0.01454689 - time (sec): 9.27 - samples/sec: 2737.19 - lr: 0.000002 - momentum: 0.000000 2023-10-17 18:04:25,406 epoch 10 - iter 98/146 - loss 0.01355592 - time (sec): 10.80 - samples/sec: 2754.28 - lr: 0.000001 - momentum: 0.000000 2023-10-17 18:04:27,080 epoch 10 - iter 112/146 - loss 0.01381817 - time (sec): 12.47 - samples/sec: 2767.65 - lr: 0.000001 - momentum: 0.000000 2023-10-17 18:04:28,326 epoch 10 - iter 126/146 - loss 0.01558305 - time (sec): 13.72 - samples/sec: 2798.28 - lr: 0.000001 - momentum: 0.000000 2023-10-17 18:04:29,771 epoch 10 - iter 140/146 - loss 0.01583718 - time (sec): 15.16 - samples/sec: 2800.43 - lr: 0.000000 - momentum: 0.000000 2023-10-17 18:04:30,590 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:04:30,591 EPOCH 10 done: loss 0.0152 - lr: 0.000000 2023-10-17 18:04:31,868 DEV : loss 0.1429719179868698 - f1-score (micro avg) 0.7533 2023-10-17 18:04:32,223 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:04:32,224 Loading model from best epoch ... 2023-10-17 18:04:33,624 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 2023-10-17 18:04:36,114 Results: - F-score (micro) 0.7598 - F-score (macro) 0.6737 - Accuracy 0.6318 By class: precision recall f1-score support PER 0.8260 0.8592 0.8423 348 LOC 0.6483 0.8123 0.7211 261 ORG 0.4250 0.3269 0.3696 52 HumanProd 0.8000 0.7273 0.7619 22 micro avg 0.7263 0.7965 0.7598 683 macro avg 0.6748 0.6814 0.6737 683 weighted avg 0.7267 0.7965 0.7574 683 2023-10-17 18:04:36,114 ----------------------------------------------------------------------------------------------------