2023-10-17 10:23:49,346 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:23:49,348 Model: "SequenceTagger( (embeddings): TransformerWordEmbeddings( (model): ElectraModel( (embeddings): ElectraEmbeddings( (word_embeddings): Embedding(32001, 768) (position_embeddings): Embedding(512, 768) (token_type_embeddings): Embedding(2, 768) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) (encoder): ElectraEncoder( (layer): ModuleList( (0-11): 12 x ElectraLayer( (attention): ElectraAttention( (self): ElectraSelfAttention( (query): Linear(in_features=768, out_features=768, bias=True) (key): Linear(in_features=768, out_features=768, bias=True) (value): Linear(in_features=768, out_features=768, bias=True) (dropout): Dropout(p=0.1, inplace=False) ) (output): ElectraSelfOutput( (dense): Linear(in_features=768, out_features=768, bias=True) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) (intermediate): ElectraIntermediate( (dense): Linear(in_features=768, out_features=3072, bias=True) (intermediate_act_fn): GELUActivation() ) (output): ElectraOutput( (dense): Linear(in_features=3072, out_features=768, bias=True) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) ) ) ) (locked_dropout): LockedDropout(p=0.5) (linear): Linear(in_features=768, out_features=13, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-17 10:23:49,348 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:23:49,348 MultiCorpus: 14465 train + 1392 dev + 2432 test sentences - NER_HIPE_2022 Corpus: 14465 train + 1392 dev + 2432 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/letemps/fr/with_doc_seperator 2023-10-17 10:23:49,348 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:23:49,348 Train: 14465 sentences 2023-10-17 10:23:49,348 (train_with_dev=False, train_with_test=False) 2023-10-17 10:23:49,348 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:23:49,348 Training Params: 2023-10-17 10:23:49,348 - learning_rate: "3e-05" 2023-10-17 10:23:49,348 - mini_batch_size: "8" 2023-10-17 10:23:49,348 - max_epochs: "10" 2023-10-17 10:23:49,348 - shuffle: "True" 2023-10-17 10:23:49,348 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:23:49,348 Plugins: 2023-10-17 10:23:49,348 - TensorboardLogger 2023-10-17 10:23:49,348 - LinearScheduler | warmup_fraction: '0.1' 2023-10-17 10:23:49,348 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:23:49,348 Final evaluation on model from best epoch (best-model.pt) 2023-10-17 10:23:49,348 - metric: "('micro avg', 'f1-score')" 2023-10-17 10:23:49,348 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:23:49,349 Computation: 2023-10-17 10:23:49,349 - compute on device: cuda:0 2023-10-17 10:23:49,349 - embedding storage: none 2023-10-17 10:23:49,349 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:23:49,349 Model training base path: "hmbench-letemps/fr-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-1" 2023-10-17 10:23:49,349 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:23:49,349 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:23:49,349 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-17 10:24:02,854 epoch 1 - iter 180/1809 - loss 1.97077795 - time (sec): 13.50 - samples/sec: 2869.63 - lr: 0.000003 - momentum: 0.000000 2023-10-17 10:24:15,874 epoch 1 - iter 360/1809 - loss 1.13419673 - time (sec): 26.52 - samples/sec: 2872.06 - lr: 0.000006 - momentum: 0.000000 2023-10-17 10:24:29,572 epoch 1 - iter 540/1809 - loss 0.80989982 - time (sec): 40.22 - samples/sec: 2841.00 - lr: 0.000009 - momentum: 0.000000 2023-10-17 10:24:43,353 epoch 1 - iter 720/1809 - loss 0.64354116 - time (sec): 54.00 - samples/sec: 2825.49 - lr: 0.000012 - momentum: 0.000000 2023-10-17 10:24:57,139 epoch 1 - iter 900/1809 - loss 0.54278891 - time (sec): 67.79 - samples/sec: 2798.57 - lr: 0.000015 - momentum: 0.000000 2023-10-17 10:25:11,974 epoch 1 - iter 1080/1809 - loss 0.47491789 - time (sec): 82.62 - samples/sec: 2741.90 - lr: 0.000018 - momentum: 0.000000 2023-10-17 10:25:26,384 epoch 1 - iter 1260/1809 - loss 0.42566960 - time (sec): 97.03 - samples/sec: 2718.52 - lr: 0.000021 - momentum: 0.000000 2023-10-17 10:25:41,389 epoch 1 - iter 1440/1809 - loss 0.38559522 - time (sec): 112.04 - samples/sec: 2704.90 - lr: 0.000024 - momentum: 0.000000 2023-10-17 10:25:54,544 epoch 1 - iter 1620/1809 - loss 0.35386098 - time (sec): 125.19 - samples/sec: 2725.72 - lr: 0.000027 - momentum: 0.000000 2023-10-17 10:26:07,351 epoch 1 - iter 1800/1809 - loss 0.32934241 - time (sec): 138.00 - samples/sec: 2738.94 - lr: 0.000030 - momentum: 0.000000 2023-10-17 10:26:08,063 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:26:08,063 EPOCH 1 done: loss 0.3284 - lr: 0.000030 2023-10-17 10:26:14,169 DEV : loss 0.09768907725811005 - f1-score (micro avg) 0.6108 2023-10-17 10:26:14,210 saving best model 2023-10-17 10:26:14,712 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:26:27,936 epoch 2 - iter 180/1809 - loss 0.10126158 - time (sec): 13.22 - samples/sec: 2761.32 - lr: 0.000030 - momentum: 0.000000 2023-10-17 10:26:41,008 epoch 2 - iter 360/1809 - loss 0.09521427 - time (sec): 26.29 - samples/sec: 2883.92 - lr: 0.000029 - momentum: 0.000000 2023-10-17 10:26:54,116 epoch 2 - iter 540/1809 - loss 0.09155082 - time (sec): 39.40 - samples/sec: 2914.13 - lr: 0.000029 - momentum: 0.000000 2023-10-17 10:27:07,181 epoch 2 - iter 720/1809 - loss 0.09337842 - time (sec): 52.47 - samples/sec: 2885.06 - lr: 0.000029 - momentum: 0.000000 2023-10-17 10:27:21,201 epoch 2 - iter 900/1809 - loss 0.09093030 - time (sec): 66.49 - samples/sec: 2821.99 - lr: 0.000028 - momentum: 0.000000 2023-10-17 10:27:34,922 epoch 2 - iter 1080/1809 - loss 0.08938417 - time (sec): 80.21 - samples/sec: 2806.14 - lr: 0.000028 - momentum: 0.000000 2023-10-17 10:27:48,461 epoch 2 - iter 1260/1809 - loss 0.08853541 - time (sec): 93.75 - samples/sec: 2802.82 - lr: 0.000028 - momentum: 0.000000 2023-10-17 10:28:02,067 epoch 2 - iter 1440/1809 - loss 0.08910538 - time (sec): 107.35 - samples/sec: 2806.40 - lr: 0.000027 - momentum: 0.000000 2023-10-17 10:28:15,988 epoch 2 - iter 1620/1809 - loss 0.08699630 - time (sec): 121.27 - samples/sec: 2817.16 - lr: 0.000027 - momentum: 0.000000 2023-10-17 10:28:29,006 epoch 2 - iter 1800/1809 - loss 0.08639068 - time (sec): 134.29 - samples/sec: 2816.30 - lr: 0.000027 - momentum: 0.000000 2023-10-17 10:28:29,643 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:28:29,644 EPOCH 2 done: loss 0.0866 - lr: 0.000027 2023-10-17 10:28:36,718 DEV : loss 0.10609198361635208 - f1-score (micro avg) 0.6388 2023-10-17 10:28:36,759 saving best model 2023-10-17 10:28:37,363 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:28:52,654 epoch 3 - iter 180/1809 - loss 0.06507567 - time (sec): 15.29 - samples/sec: 2455.09 - lr: 0.000026 - momentum: 0.000000 2023-10-17 10:29:07,332 epoch 3 - iter 360/1809 - loss 0.06327625 - time (sec): 29.97 - samples/sec: 2571.73 - lr: 0.000026 - momentum: 0.000000 2023-10-17 10:29:22,132 epoch 3 - iter 540/1809 - loss 0.06394436 - time (sec): 44.77 - samples/sec: 2564.70 - lr: 0.000026 - momentum: 0.000000 2023-10-17 10:29:36,446 epoch 3 - iter 720/1809 - loss 0.06297439 - time (sec): 59.08 - samples/sec: 2604.56 - lr: 0.000025 - momentum: 0.000000 2023-10-17 10:29:49,726 epoch 3 - iter 900/1809 - loss 0.06209821 - time (sec): 72.36 - samples/sec: 2647.74 - lr: 0.000025 - momentum: 0.000000 2023-10-17 10:30:02,972 epoch 3 - iter 1080/1809 - loss 0.06128483 - time (sec): 85.61 - samples/sec: 2688.49 - lr: 0.000025 - momentum: 0.000000 2023-10-17 10:30:16,883 epoch 3 - iter 1260/1809 - loss 0.06138333 - time (sec): 99.52 - samples/sec: 2691.44 - lr: 0.000024 - momentum: 0.000000 2023-10-17 10:30:31,017 epoch 3 - iter 1440/1809 - loss 0.06203921 - time (sec): 113.65 - samples/sec: 2679.12 - lr: 0.000024 - momentum: 0.000000 2023-10-17 10:30:43,946 epoch 3 - iter 1620/1809 - loss 0.06242292 - time (sec): 126.58 - samples/sec: 2700.92 - lr: 0.000024 - momentum: 0.000000 2023-10-17 10:30:55,919 epoch 3 - iter 1800/1809 - loss 0.06256995 - time (sec): 138.55 - samples/sec: 2729.61 - lr: 0.000023 - momentum: 0.000000 2023-10-17 10:30:56,446 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:30:56,446 EPOCH 3 done: loss 0.0626 - lr: 0.000023 2023-10-17 10:31:02,865 DEV : loss 0.12596164643764496 - f1-score (micro avg) 0.6405 2023-10-17 10:31:02,910 saving best model 2023-10-17 10:31:03,530 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:31:17,877 epoch 4 - iter 180/1809 - loss 0.04168279 - time (sec): 14.34 - samples/sec: 2707.50 - lr: 0.000023 - momentum: 0.000000 2023-10-17 10:31:31,828 epoch 4 - iter 360/1809 - loss 0.03962130 - time (sec): 28.30 - samples/sec: 2707.68 - lr: 0.000023 - momentum: 0.000000 2023-10-17 10:31:46,220 epoch 4 - iter 540/1809 - loss 0.04238080 - time (sec): 42.69 - samples/sec: 2714.96 - lr: 0.000022 - momentum: 0.000000 2023-10-17 10:31:59,189 epoch 4 - iter 720/1809 - loss 0.04283943 - time (sec): 55.66 - samples/sec: 2743.08 - lr: 0.000022 - momentum: 0.000000 2023-10-17 10:32:12,306 epoch 4 - iter 900/1809 - loss 0.04535463 - time (sec): 68.77 - samples/sec: 2782.02 - lr: 0.000022 - momentum: 0.000000 2023-10-17 10:32:25,119 epoch 4 - iter 1080/1809 - loss 0.04573226 - time (sec): 81.59 - samples/sec: 2784.61 - lr: 0.000021 - momentum: 0.000000 2023-10-17 10:32:38,107 epoch 4 - iter 1260/1809 - loss 0.04529284 - time (sec): 94.57 - samples/sec: 2803.28 - lr: 0.000021 - momentum: 0.000000 2023-10-17 10:32:51,646 epoch 4 - iter 1440/1809 - loss 0.04597535 - time (sec): 108.11 - samples/sec: 2807.64 - lr: 0.000021 - momentum: 0.000000 2023-10-17 10:33:05,014 epoch 4 - iter 1620/1809 - loss 0.04539057 - time (sec): 121.48 - samples/sec: 2801.07 - lr: 0.000020 - momentum: 0.000000 2023-10-17 10:33:18,630 epoch 4 - iter 1800/1809 - loss 0.04608749 - time (sec): 135.10 - samples/sec: 2798.18 - lr: 0.000020 - momentum: 0.000000 2023-10-17 10:33:19,371 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:33:19,371 EPOCH 4 done: loss 0.0461 - lr: 0.000020 2023-10-17 10:33:26,556 DEV : loss 0.17203289270401 - f1-score (micro avg) 0.6513 2023-10-17 10:33:26,598 saving best model 2023-10-17 10:33:27,228 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:33:41,243 epoch 5 - iter 180/1809 - loss 0.02768648 - time (sec): 14.01 - samples/sec: 2708.81 - lr: 0.000020 - momentum: 0.000000 2023-10-17 10:33:54,303 epoch 5 - iter 360/1809 - loss 0.03102599 - time (sec): 27.07 - samples/sec: 2738.17 - lr: 0.000019 - momentum: 0.000000 2023-10-17 10:34:08,359 epoch 5 - iter 540/1809 - loss 0.03185567 - time (sec): 41.13 - samples/sec: 2734.73 - lr: 0.000019 - momentum: 0.000000 2023-10-17 10:34:21,848 epoch 5 - iter 720/1809 - loss 0.03061780 - time (sec): 54.62 - samples/sec: 2737.94 - lr: 0.000019 - momentum: 0.000000 2023-10-17 10:34:36,525 epoch 5 - iter 900/1809 - loss 0.03119659 - time (sec): 69.29 - samples/sec: 2723.36 - lr: 0.000018 - momentum: 0.000000 2023-10-17 10:34:50,562 epoch 5 - iter 1080/1809 - loss 0.03101663 - time (sec): 83.33 - samples/sec: 2720.38 - lr: 0.000018 - momentum: 0.000000 2023-10-17 10:35:04,954 epoch 5 - iter 1260/1809 - loss 0.03070119 - time (sec): 97.72 - samples/sec: 2718.56 - lr: 0.000018 - momentum: 0.000000 2023-10-17 10:35:19,209 epoch 5 - iter 1440/1809 - loss 0.03061179 - time (sec): 111.98 - samples/sec: 2706.42 - lr: 0.000017 - momentum: 0.000000 2023-10-17 10:35:33,200 epoch 5 - iter 1620/1809 - loss 0.03213924 - time (sec): 125.97 - samples/sec: 2700.86 - lr: 0.000017 - momentum: 0.000000 2023-10-17 10:35:46,610 epoch 5 - iter 1800/1809 - loss 0.03232409 - time (sec): 139.38 - samples/sec: 2712.26 - lr: 0.000017 - momentum: 0.000000 2023-10-17 10:35:47,324 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:35:47,324 EPOCH 5 done: loss 0.0324 - lr: 0.000017 2023-10-17 10:35:53,774 DEV : loss 0.2933090031147003 - f1-score (micro avg) 0.6564 2023-10-17 10:35:53,824 saving best model 2023-10-17 10:35:54,466 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:36:08,260 epoch 6 - iter 180/1809 - loss 0.02526723 - time (sec): 13.79 - samples/sec: 2751.63 - lr: 0.000016 - momentum: 0.000000 2023-10-17 10:36:21,717 epoch 6 - iter 360/1809 - loss 0.02266318 - time (sec): 27.25 - samples/sec: 2791.79 - lr: 0.000016 - momentum: 0.000000 2023-10-17 10:36:35,235 epoch 6 - iter 540/1809 - loss 0.02250374 - time (sec): 40.77 - samples/sec: 2768.29 - lr: 0.000016 - momentum: 0.000000 2023-10-17 10:36:48,648 epoch 6 - iter 720/1809 - loss 0.02114013 - time (sec): 54.18 - samples/sec: 2787.83 - lr: 0.000015 - momentum: 0.000000 2023-10-17 10:37:02,292 epoch 6 - iter 900/1809 - loss 0.02208737 - time (sec): 67.82 - samples/sec: 2787.14 - lr: 0.000015 - momentum: 0.000000 2023-10-17 10:37:15,201 epoch 6 - iter 1080/1809 - loss 0.02271038 - time (sec): 80.73 - samples/sec: 2800.43 - lr: 0.000015 - momentum: 0.000000 2023-10-17 10:37:29,400 epoch 6 - iter 1260/1809 - loss 0.02373796 - time (sec): 94.93 - samples/sec: 2783.97 - lr: 0.000014 - momentum: 0.000000 2023-10-17 10:37:43,390 epoch 6 - iter 1440/1809 - loss 0.02351275 - time (sec): 108.92 - samples/sec: 2769.13 - lr: 0.000014 - momentum: 0.000000 2023-10-17 10:37:56,318 epoch 6 - iter 1620/1809 - loss 0.02349186 - time (sec): 121.85 - samples/sec: 2791.97 - lr: 0.000014 - momentum: 0.000000 2023-10-17 10:38:09,258 epoch 6 - iter 1800/1809 - loss 0.02321458 - time (sec): 134.79 - samples/sec: 2801.94 - lr: 0.000013 - momentum: 0.000000 2023-10-17 10:38:09,960 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:38:09,961 EPOCH 6 done: loss 0.0231 - lr: 0.000013 2023-10-17 10:38:16,483 DEV : loss 0.317108690738678 - f1-score (micro avg) 0.6555 2023-10-17 10:38:16,533 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:38:30,845 epoch 7 - iter 180/1809 - loss 0.01666041 - time (sec): 14.31 - samples/sec: 2797.44 - lr: 0.000013 - momentum: 0.000000 2023-10-17 10:38:45,010 epoch 7 - iter 360/1809 - loss 0.01535944 - time (sec): 28.48 - samples/sec: 2763.71 - lr: 0.000013 - momentum: 0.000000 2023-10-17 10:38:59,190 epoch 7 - iter 540/1809 - loss 0.01585990 - time (sec): 42.66 - samples/sec: 2712.38 - lr: 0.000012 - momentum: 0.000000 2023-10-17 10:39:12,131 epoch 7 - iter 720/1809 - loss 0.01588240 - time (sec): 55.60 - samples/sec: 2743.87 - lr: 0.000012 - momentum: 0.000000 2023-10-17 10:39:24,799 epoch 7 - iter 900/1809 - loss 0.01567846 - time (sec): 68.26 - samples/sec: 2772.04 - lr: 0.000012 - momentum: 0.000000 2023-10-17 10:39:37,658 epoch 7 - iter 1080/1809 - loss 0.01573739 - time (sec): 81.12 - samples/sec: 2782.75 - lr: 0.000011 - momentum: 0.000000 2023-10-17 10:39:51,060 epoch 7 - iter 1260/1809 - loss 0.01492397 - time (sec): 94.53 - samples/sec: 2794.91 - lr: 0.000011 - momentum: 0.000000 2023-10-17 10:40:04,433 epoch 7 - iter 1440/1809 - loss 0.01512705 - time (sec): 107.90 - samples/sec: 2804.37 - lr: 0.000011 - momentum: 0.000000 2023-10-17 10:40:18,776 epoch 7 - iter 1620/1809 - loss 0.01537735 - time (sec): 122.24 - samples/sec: 2781.25 - lr: 0.000010 - momentum: 0.000000 2023-10-17 10:40:33,065 epoch 7 - iter 1800/1809 - loss 0.01577194 - time (sec): 136.53 - samples/sec: 2771.15 - lr: 0.000010 - momentum: 0.000000 2023-10-17 10:40:33,734 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:40:33,735 EPOCH 7 done: loss 0.0157 - lr: 0.000010 2023-10-17 10:40:40,913 DEV : loss 0.35636281967163086 - f1-score (micro avg) 0.6586 2023-10-17 10:40:40,989 saving best model 2023-10-17 10:40:41,627 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:40:55,327 epoch 8 - iter 180/1809 - loss 0.01233300 - time (sec): 13.70 - samples/sec: 2714.96 - lr: 0.000010 - momentum: 0.000000 2023-10-17 10:41:08,947 epoch 8 - iter 360/1809 - loss 0.01007325 - time (sec): 27.32 - samples/sec: 2768.24 - lr: 0.000009 - momentum: 0.000000 2023-10-17 10:41:22,905 epoch 8 - iter 540/1809 - loss 0.01079234 - time (sec): 41.28 - samples/sec: 2745.17 - lr: 0.000009 - momentum: 0.000000 2023-10-17 10:41:36,976 epoch 8 - iter 720/1809 - loss 0.01286652 - time (sec): 55.35 - samples/sec: 2726.46 - lr: 0.000009 - momentum: 0.000000 2023-10-17 10:41:51,431 epoch 8 - iter 900/1809 - loss 0.01195537 - time (sec): 69.80 - samples/sec: 2706.92 - lr: 0.000008 - momentum: 0.000000 2023-10-17 10:42:05,932 epoch 8 - iter 1080/1809 - loss 0.01176322 - time (sec): 84.30 - samples/sec: 2683.32 - lr: 0.000008 - momentum: 0.000000 2023-10-17 10:42:20,154 epoch 8 - iter 1260/1809 - loss 0.01134419 - time (sec): 98.52 - samples/sec: 2682.88 - lr: 0.000008 - momentum: 0.000000 2023-10-17 10:42:33,804 epoch 8 - iter 1440/1809 - loss 0.01129014 - time (sec): 112.18 - samples/sec: 2696.96 - lr: 0.000007 - momentum: 0.000000 2023-10-17 10:42:47,117 epoch 8 - iter 1620/1809 - loss 0.01089610 - time (sec): 125.49 - samples/sec: 2708.67 - lr: 0.000007 - momentum: 0.000000 2023-10-17 10:43:00,413 epoch 8 - iter 1800/1809 - loss 0.01091399 - time (sec): 138.78 - samples/sec: 2725.05 - lr: 0.000007 - momentum: 0.000000 2023-10-17 10:43:01,033 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:43:01,033 EPOCH 8 done: loss 0.0110 - lr: 0.000007 2023-10-17 10:43:07,612 DEV : loss 0.35512134432792664 - f1-score (micro avg) 0.6591 2023-10-17 10:43:07,659 saving best model 2023-10-17 10:43:08,266 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:43:21,490 epoch 9 - iter 180/1809 - loss 0.00903196 - time (sec): 13.22 - samples/sec: 2927.11 - lr: 0.000006 - momentum: 0.000000 2023-10-17 10:43:33,246 epoch 9 - iter 360/1809 - loss 0.00779450 - time (sec): 24.98 - samples/sec: 2999.25 - lr: 0.000006 - momentum: 0.000000 2023-10-17 10:43:45,896 epoch 9 - iter 540/1809 - loss 0.00679913 - time (sec): 37.63 - samples/sec: 2998.42 - lr: 0.000006 - momentum: 0.000000 2023-10-17 10:43:57,445 epoch 9 - iter 720/1809 - loss 0.00667016 - time (sec): 49.18 - samples/sec: 3080.76 - lr: 0.000005 - momentum: 0.000000 2023-10-17 10:44:09,143 epoch 9 - iter 900/1809 - loss 0.00738014 - time (sec): 60.88 - samples/sec: 3111.49 - lr: 0.000005 - momentum: 0.000000 2023-10-17 10:44:21,055 epoch 9 - iter 1080/1809 - loss 0.00679923 - time (sec): 72.79 - samples/sec: 3125.62 - lr: 0.000005 - momentum: 0.000000 2023-10-17 10:44:33,455 epoch 9 - iter 1260/1809 - loss 0.00660081 - time (sec): 85.19 - samples/sec: 3109.31 - lr: 0.000004 - momentum: 0.000000 2023-10-17 10:44:47,147 epoch 9 - iter 1440/1809 - loss 0.00659073 - time (sec): 98.88 - samples/sec: 3062.56 - lr: 0.000004 - momentum: 0.000000 2023-10-17 10:45:01,502 epoch 9 - iter 1620/1809 - loss 0.00679880 - time (sec): 113.23 - samples/sec: 3015.68 - lr: 0.000004 - momentum: 0.000000 2023-10-17 10:45:14,150 epoch 9 - iter 1800/1809 - loss 0.00670581 - time (sec): 125.88 - samples/sec: 3006.68 - lr: 0.000003 - momentum: 0.000000 2023-10-17 10:45:14,787 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:45:14,787 EPOCH 9 done: loss 0.0068 - lr: 0.000003 2023-10-17 10:45:21,264 DEV : loss 0.38741225004196167 - f1-score (micro avg) 0.6648 2023-10-17 10:45:21,310 saving best model 2023-10-17 10:45:21,908 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:45:35,707 epoch 10 - iter 180/1809 - loss 0.00757151 - time (sec): 13.80 - samples/sec: 2784.91 - lr: 0.000003 - momentum: 0.000000 2023-10-17 10:45:49,725 epoch 10 - iter 360/1809 - loss 0.00689107 - time (sec): 27.82 - samples/sec: 2720.52 - lr: 0.000003 - momentum: 0.000000 2023-10-17 10:46:04,659 epoch 10 - iter 540/1809 - loss 0.00613861 - time (sec): 42.75 - samples/sec: 2647.42 - lr: 0.000002 - momentum: 0.000000 2023-10-17 10:46:18,293 epoch 10 - iter 720/1809 - loss 0.00588222 - time (sec): 56.38 - samples/sec: 2663.59 - lr: 0.000002 - momentum: 0.000000 2023-10-17 10:46:32,068 epoch 10 - iter 900/1809 - loss 0.00544332 - time (sec): 70.16 - samples/sec: 2658.14 - lr: 0.000002 - momentum: 0.000000 2023-10-17 10:46:45,517 epoch 10 - iter 1080/1809 - loss 0.00530158 - time (sec): 83.61 - samples/sec: 2688.74 - lr: 0.000001 - momentum: 0.000000 2023-10-17 10:46:59,225 epoch 10 - iter 1260/1809 - loss 0.00497582 - time (sec): 97.32 - samples/sec: 2719.02 - lr: 0.000001 - momentum: 0.000000 2023-10-17 10:47:12,932 epoch 10 - iter 1440/1809 - loss 0.00553063 - time (sec): 111.02 - samples/sec: 2716.47 - lr: 0.000001 - momentum: 0.000000 2023-10-17 10:47:27,001 epoch 10 - iter 1620/1809 - loss 0.00533905 - time (sec): 125.09 - samples/sec: 2716.76 - lr: 0.000000 - momentum: 0.000000 2023-10-17 10:47:41,204 epoch 10 - iter 1800/1809 - loss 0.00529427 - time (sec): 139.29 - samples/sec: 2715.81 - lr: 0.000000 - momentum: 0.000000 2023-10-17 10:47:41,818 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:47:41,818 EPOCH 10 done: loss 0.0054 - lr: 0.000000 2023-10-17 10:47:48,079 DEV : loss 0.40038371086120605 - f1-score (micro avg) 0.6651 2023-10-17 10:47:48,122 saving best model 2023-10-17 10:47:49,288 ---------------------------------------------------------------------------------------------------- 2023-10-17 10:47:49,290 Loading model from best epoch ... 2023-10-17 10:47:51,019 SequenceTagger predicts: Dictionary with 13 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 2023-10-17 10:48:00,015 Results: - F-score (micro) 0.6719 - F-score (macro) 0.5432 - Accuracy 0.5187 By class: precision recall f1-score support loc 0.6531 0.8156 0.7254 591 pers 0.5876 0.7423 0.6559 357 org 0.2931 0.2152 0.2482 79 micro avg 0.6127 0.7439 0.6719 1027 macro avg 0.5113 0.5910 0.5432 1027 weighted avg 0.6026 0.7439 0.6645 1027 2023-10-17 10:48:00,015 ----------------------------------------------------------------------------------------------------