2023-10-27 20:07:01,765 ---------------------------------------------------------------------------------------------------- 2023-10-27 20:07:01,767 Model: "SequenceTagger( (embeddings): TransformerWordEmbeddings( (model): XLMRobertaModel( (embeddings): XLMRobertaEmbeddings( (word_embeddings): Embedding(250003, 1024) (position_embeddings): Embedding(514, 1024, padding_idx=1) (token_type_embeddings): Embedding(1, 1024) (LayerNorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) (encoder): XLMRobertaEncoder( (layer): ModuleList( (0-23): 24 x XLMRobertaLayer( (attention): XLMRobertaAttention( (self): XLMRobertaSelfAttention( (query): Linear(in_features=1024, out_features=1024, bias=True) (key): Linear(in_features=1024, out_features=1024, bias=True) (value): Linear(in_features=1024, out_features=1024, bias=True) (dropout): Dropout(p=0.1, inplace=False) ) (output): XLMRobertaSelfOutput( (dense): Linear(in_features=1024, out_features=1024, bias=True) (LayerNorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) (intermediate): XLMRobertaIntermediate( (dense): Linear(in_features=1024, out_features=4096, bias=True) (intermediate_act_fn): GELUActivation() ) (output): XLMRobertaOutput( (dense): Linear(in_features=4096, out_features=1024, bias=True) (LayerNorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) ) (pooler): XLMRobertaPooler( (dense): Linear(in_features=1024, out_features=1024, bias=True) (activation): Tanh() ) ) ) (locked_dropout): LockedDropout(p=0.5) (linear): Linear(in_features=1024, out_features=17, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-27 20:07:01,767 ---------------------------------------------------------------------------------------------------- 2023-10-27 20:07:01,767 Corpus: 14903 train + 3449 dev + 3658 test sentences 2023-10-27 20:07:01,767 ---------------------------------------------------------------------------------------------------- 2023-10-27 20:07:01,767 Train: 14903 sentences 2023-10-27 20:07:01,767 (train_with_dev=False, train_with_test=False) 2023-10-27 20:07:01,767 ---------------------------------------------------------------------------------------------------- 2023-10-27 20:07:01,767 Training Params: 2023-10-27 20:07:01,767 - learning_rate: "5e-06" 2023-10-27 20:07:01,767 - mini_batch_size: "4" 2023-10-27 20:07:01,767 - max_epochs: "10" 2023-10-27 20:07:01,767 - shuffle: "True" 2023-10-27 20:07:01,767 ---------------------------------------------------------------------------------------------------- 2023-10-27 20:07:01,768 Plugins: 2023-10-27 20:07:01,768 - TensorboardLogger 2023-10-27 20:07:01,768 - LinearScheduler | warmup_fraction: '0.1' 2023-10-27 20:07:01,768 ---------------------------------------------------------------------------------------------------- 2023-10-27 20:07:01,768 Final evaluation on model from best epoch (best-model.pt) 2023-10-27 20:07:01,768 - metric: "('micro avg', 'f1-score')" 2023-10-27 20:07:01,768 ---------------------------------------------------------------------------------------------------- 2023-10-27 20:07:01,768 Computation: 2023-10-27 20:07:01,768 - compute on device: cuda:0 2023-10-27 20:07:01,768 - embedding storage: none 2023-10-27 20:07:01,768 ---------------------------------------------------------------------------------------------------- 2023-10-27 20:07:01,768 Model training base path: "flair-clean-conll-lr5e-06-bs4-5" 2023-10-27 20:07:01,768 ---------------------------------------------------------------------------------------------------- 2023-10-27 20:07:01,768 ---------------------------------------------------------------------------------------------------- 2023-10-27 20:07:01,768 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-27 20:07:47,370 epoch 1 - iter 372/3726 - loss 2.80593129 - time (sec): 45.60 - samples/sec: 438.05 - lr: 0.000000 - momentum: 0.000000 2023-10-27 20:08:32,644 epoch 1 - iter 744/3726 - loss 1.85094677 - time (sec): 90.87 - samples/sec: 444.69 - lr: 0.000001 - momentum: 0.000000 2023-10-27 20:09:17,954 epoch 1 - iter 1116/3726 - loss 1.42058830 - time (sec): 136.18 - samples/sec: 444.93 - lr: 0.000001 - momentum: 0.000000 2023-10-27 20:10:03,236 epoch 1 - iter 1488/3726 - loss 1.16393809 - time (sec): 181.47 - samples/sec: 444.38 - lr: 0.000002 - momentum: 0.000000 2023-10-27 20:10:48,524 epoch 1 - iter 1860/3726 - loss 0.98551923 - time (sec): 226.75 - samples/sec: 446.04 - lr: 0.000002 - momentum: 0.000000 2023-10-27 20:11:33,918 epoch 1 - iter 2232/3726 - loss 0.84790959 - time (sec): 272.15 - samples/sec: 450.27 - lr: 0.000003 - momentum: 0.000000 2023-10-27 20:12:19,596 epoch 1 - iter 2604/3726 - loss 0.74598188 - time (sec): 317.83 - samples/sec: 451.02 - lr: 0.000003 - momentum: 0.000000 2023-10-27 20:13:05,071 epoch 1 - iter 2976/3726 - loss 0.66540477 - time (sec): 363.30 - samples/sec: 451.42 - lr: 0.000004 - momentum: 0.000000 2023-10-27 20:13:50,701 epoch 1 - iter 3348/3726 - loss 0.60748783 - time (sec): 408.93 - samples/sec: 449.21 - lr: 0.000004 - momentum: 0.000000 2023-10-27 20:14:35,944 epoch 1 - iter 3720/3726 - loss 0.55800133 - time (sec): 454.17 - samples/sec: 449.55 - lr: 0.000005 - momentum: 0.000000 2023-10-27 20:14:36,662 ---------------------------------------------------------------------------------------------------- 2023-10-27 20:14:36,663 EPOCH 1 done: loss 0.5571 - lr: 0.000005 2023-10-27 20:15:00,976 DEV : loss 0.08272701501846313 - f1-score (micro avg) 0.9305 2023-10-27 20:15:01,029 saving best model 2023-10-27 20:15:02,837 ---------------------------------------------------------------------------------------------------- 2023-10-27 20:15:49,496 epoch 2 - iter 372/3726 - loss 0.09075236 - time (sec): 46.66 - samples/sec: 446.77 - lr: 0.000005 - momentum: 0.000000 2023-10-27 20:16:35,366 epoch 2 - iter 744/3726 - loss 0.09661752 - time (sec): 92.53 - samples/sec: 440.88 - lr: 0.000005 - momentum: 0.000000 2023-10-27 20:17:20,750 epoch 2 - iter 1116/3726 - loss 0.09592189 - time (sec): 137.91 - samples/sec: 442.98 - lr: 0.000005 - momentum: 0.000000 2023-10-27 20:18:05,959 epoch 2 - iter 1488/3726 - loss 0.09322980 - time (sec): 183.12 - samples/sec: 443.25 - lr: 0.000005 - momentum: 0.000000 2023-10-27 20:18:51,330 epoch 2 - iter 1860/3726 - loss 0.08941354 - time (sec): 228.49 - samples/sec: 443.63 - lr: 0.000005 - momentum: 0.000000 2023-10-27 20:19:36,865 epoch 2 - iter 2232/3726 - loss 0.08782526 - time (sec): 274.03 - samples/sec: 444.12 - lr: 0.000005 - momentum: 0.000000 2023-10-27 20:20:21,760 epoch 2 - iter 2604/3726 - loss 0.08763755 - time (sec): 318.92 - samples/sec: 447.07 - lr: 0.000005 - momentum: 0.000000 2023-10-27 20:21:07,307 epoch 2 - iter 2976/3726 - loss 0.08459677 - time (sec): 364.47 - samples/sec: 449.80 - lr: 0.000005 - momentum: 0.000000 2023-10-27 20:21:52,612 epoch 2 - iter 3348/3726 - loss 0.08200724 - time (sec): 409.77 - samples/sec: 449.39 - lr: 0.000005 - momentum: 0.000000 2023-10-27 20:22:37,887 epoch 2 - iter 3720/3726 - loss 0.08090953 - time (sec): 455.05 - samples/sec: 448.96 - lr: 0.000004 - momentum: 0.000000 2023-10-27 20:22:38,602 ---------------------------------------------------------------------------------------------------- 2023-10-27 20:22:38,603 EPOCH 2 done: loss 0.0808 - lr: 0.000004 2023-10-27 20:23:01,816 DEV : loss 0.0558977946639061 - f1-score (micro avg) 0.9643 2023-10-27 20:23:01,871 saving best model 2023-10-27 20:23:04,562 ---------------------------------------------------------------------------------------------------- 2023-10-27 20:23:50,075 epoch 3 - iter 372/3726 - loss 0.04925132 - time (sec): 45.51 - samples/sec: 436.29 - lr: 0.000004 - momentum: 0.000000 2023-10-27 20:24:35,505 epoch 3 - iter 744/3726 - loss 0.05096433 - time (sec): 90.94 - samples/sec: 441.79 - lr: 0.000004 - momentum: 0.000000 2023-10-27 20:25:21,376 epoch 3 - iter 1116/3726 - loss 0.05345821 - time (sec): 136.81 - samples/sec: 444.14 - lr: 0.000004 - momentum: 0.000000 2023-10-27 20:26:06,796 epoch 3 - iter 1488/3726 - loss 0.05364040 - time (sec): 182.23 - samples/sec: 444.47 - lr: 0.000004 - momentum: 0.000000 2023-10-27 20:26:52,221 epoch 3 - iter 1860/3726 - loss 0.05380637 - time (sec): 227.66 - samples/sec: 447.62 - lr: 0.000004 - momentum: 0.000000 2023-10-27 20:27:37,903 epoch 3 - iter 2232/3726 - loss 0.05332169 - time (sec): 273.34 - samples/sec: 448.66 - lr: 0.000004 - momentum: 0.000000 2023-10-27 20:28:24,443 epoch 3 - iter 2604/3726 - loss 0.05365144 - time (sec): 319.88 - samples/sec: 446.15 - lr: 0.000004 - momentum: 0.000000 2023-10-27 20:29:09,620 epoch 3 - iter 2976/3726 - loss 0.05262140 - time (sec): 365.06 - samples/sec: 447.36 - lr: 0.000004 - momentum: 0.000000 2023-10-27 20:29:55,368 epoch 3 - iter 3348/3726 - loss 0.05205947 - time (sec): 410.80 - samples/sec: 448.23 - lr: 0.000004 - momentum: 0.000000 2023-10-27 20:30:40,872 epoch 3 - iter 3720/3726 - loss 0.05122877 - time (sec): 456.31 - samples/sec: 447.79 - lr: 0.000004 - momentum: 0.000000 2023-10-27 20:30:41,552 ---------------------------------------------------------------------------------------------------- 2023-10-27 20:30:41,553 EPOCH 3 done: loss 0.0512 - lr: 0.000004 2023-10-27 20:31:04,451 DEV : loss 0.04910625144839287 - f1-score (micro avg) 0.969 2023-10-27 20:31:04,505 saving best model 2023-10-27 20:31:07,070 ---------------------------------------------------------------------------------------------------- 2023-10-27 20:31:52,530 epoch 4 - iter 372/3726 - loss 0.03342383 - time (sec): 45.46 - samples/sec: 455.54 - lr: 0.000004 - momentum: 0.000000 2023-10-27 20:32:37,688 epoch 4 - iter 744/3726 - loss 0.03205166 - time (sec): 90.62 - samples/sec: 456.58 - lr: 0.000004 - momentum: 0.000000 2023-10-27 20:33:24,378 epoch 4 - iter 1116/3726 - loss 0.03357460 - time (sec): 137.31 - samples/sec: 451.87 - lr: 0.000004 - momentum: 0.000000 2023-10-27 20:34:10,145 epoch 4 - iter 1488/3726 - loss 0.03617981 - time (sec): 183.07 - samples/sec: 454.34 - lr: 0.000004 - momentum: 0.000000 2023-10-27 20:34:56,203 epoch 4 - iter 1860/3726 - loss 0.03561724 - time (sec): 229.13 - samples/sec: 450.05 - lr: 0.000004 - momentum: 0.000000 2023-10-27 20:35:42,349 epoch 4 - iter 2232/3726 - loss 0.03513961 - time (sec): 275.28 - samples/sec: 447.95 - lr: 0.000004 - momentum: 0.000000 2023-10-27 20:36:27,872 epoch 4 - iter 2604/3726 - loss 0.03560568 - time (sec): 320.80 - samples/sec: 446.14 - lr: 0.000004 - momentum: 0.000000 2023-10-27 20:37:13,373 epoch 4 - iter 2976/3726 - loss 0.03572121 - time (sec): 366.30 - samples/sec: 445.85 - lr: 0.000003 - momentum: 0.000000 2023-10-27 20:37:59,045 epoch 4 - iter 3348/3726 - loss 0.03483182 - time (sec): 411.97 - samples/sec: 446.11 - lr: 0.000003 - momentum: 0.000000 2023-10-27 20:38:44,300 epoch 4 - iter 3720/3726 - loss 0.03483957 - time (sec): 457.23 - samples/sec: 447.06 - lr: 0.000003 - momentum: 0.000000 2023-10-27 20:38:44,988 ---------------------------------------------------------------------------------------------------- 2023-10-27 20:38:44,988 EPOCH 4 done: loss 0.0348 - lr: 0.000003 2023-10-27 20:39:07,891 DEV : loss 0.04652674123644829 - f1-score (micro avg) 0.9705 2023-10-27 20:39:07,943 saving best model 2023-10-27 20:39:10,583 ---------------------------------------------------------------------------------------------------- 2023-10-27 20:39:56,189 epoch 5 - iter 372/3726 - loss 0.03173966 - time (sec): 45.60 - samples/sec: 442.22 - lr: 0.000003 - momentum: 0.000000 2023-10-27 20:40:42,588 epoch 5 - iter 744/3726 - loss 0.03324355 - time (sec): 92.00 - samples/sec: 441.81 - lr: 0.000003 - momentum: 0.000000 2023-10-27 20:41:28,491 epoch 5 - iter 1116/3726 - loss 0.03176114 - time (sec): 137.91 - samples/sec: 446.35 - lr: 0.000003 - momentum: 0.000000 2023-10-27 20:42:14,724 epoch 5 - iter 1488/3726 - loss 0.02967819 - time (sec): 184.14 - samples/sec: 446.19 - lr: 0.000003 - momentum: 0.000000 2023-10-27 20:42:59,527 epoch 5 - iter 1860/3726 - loss 0.03079490 - time (sec): 228.94 - samples/sec: 446.65 - lr: 0.000003 - momentum: 0.000000 2023-10-27 20:43:44,886 epoch 5 - iter 2232/3726 - loss 0.02966149 - time (sec): 274.30 - samples/sec: 445.62 - lr: 0.000003 - momentum: 0.000000 2023-10-27 20:44:30,379 epoch 5 - iter 2604/3726 - loss 0.03065570 - time (sec): 319.79 - samples/sec: 446.65 - lr: 0.000003 - momentum: 0.000000 2023-10-27 20:45:15,785 epoch 5 - iter 2976/3726 - loss 0.03042225 - time (sec): 365.20 - samples/sec: 446.89 - lr: 0.000003 - momentum: 0.000000 2023-10-27 20:46:01,755 epoch 5 - iter 3348/3726 - loss 0.03020845 - time (sec): 411.17 - samples/sec: 446.33 - lr: 0.000003 - momentum: 0.000000 2023-10-27 20:46:47,177 epoch 5 - iter 3720/3726 - loss 0.02983678 - time (sec): 456.59 - samples/sec: 447.55 - lr: 0.000003 - momentum: 0.000000 2023-10-27 20:46:47,923 ---------------------------------------------------------------------------------------------------- 2023-10-27 20:46:47,924 EPOCH 5 done: loss 0.0299 - lr: 0.000003 2023-10-27 20:47:10,884 DEV : loss 0.050089359283447266 - f1-score (micro avg) 0.9712 2023-10-27 20:47:10,938 saving best model 2023-10-27 20:47:13,597 ---------------------------------------------------------------------------------------------------- 2023-10-27 20:47:59,020 epoch 6 - iter 372/3726 - loss 0.02248010 - time (sec): 45.42 - samples/sec: 453.45 - lr: 0.000003 - momentum: 0.000000 2023-10-27 20:48:44,494 epoch 6 - iter 744/3726 - loss 0.01889729 - time (sec): 90.89 - samples/sec: 449.45 - lr: 0.000003 - momentum: 0.000000 2023-10-27 20:49:30,409 epoch 6 - iter 1116/3726 - loss 0.01902198 - time (sec): 136.81 - samples/sec: 446.61 - lr: 0.000003 - momentum: 0.000000 2023-10-27 20:50:16,297 epoch 6 - iter 1488/3726 - loss 0.02004643 - time (sec): 182.70 - samples/sec: 443.97 - lr: 0.000003 - momentum: 0.000000 2023-10-27 20:51:01,691 epoch 6 - iter 1860/3726 - loss 0.01983142 - time (sec): 228.09 - samples/sec: 444.93 - lr: 0.000003 - momentum: 0.000000 2023-10-27 20:51:47,098 epoch 6 - iter 2232/3726 - loss 0.02028738 - time (sec): 273.50 - samples/sec: 446.24 - lr: 0.000002 - momentum: 0.000000 2023-10-27 20:52:33,330 epoch 6 - iter 2604/3726 - loss 0.02002001 - time (sec): 319.73 - samples/sec: 446.18 - lr: 0.000002 - momentum: 0.000000 2023-10-27 20:53:19,605 epoch 6 - iter 2976/3726 - loss 0.02081204 - time (sec): 366.01 - samples/sec: 445.36 - lr: 0.000002 - momentum: 0.000000 2023-10-27 20:54:05,958 epoch 6 - iter 3348/3726 - loss 0.02032741 - time (sec): 412.36 - samples/sec: 445.83 - lr: 0.000002 - momentum: 0.000000 2023-10-27 20:54:52,761 epoch 6 - iter 3720/3726 - loss 0.02044719 - time (sec): 459.16 - samples/sec: 444.72 - lr: 0.000002 - momentum: 0.000000 2023-10-27 20:54:53,517 ---------------------------------------------------------------------------------------------------- 2023-10-27 20:54:53,517 EPOCH 6 done: loss 0.0205 - lr: 0.000002 2023-10-27 20:55:17,099 DEV : loss 0.04764683172106743 - f1-score (micro avg) 0.9742 2023-10-27 20:55:17,154 saving best model 2023-10-27 20:55:19,755 ---------------------------------------------------------------------------------------------------- 2023-10-27 20:56:05,378 epoch 7 - iter 372/3726 - loss 0.02366929 - time (sec): 45.62 - samples/sec: 447.27 - lr: 0.000002 - momentum: 0.000000 2023-10-27 20:56:51,118 epoch 7 - iter 744/3726 - loss 0.02311710 - time (sec): 91.36 - samples/sec: 439.82 - lr: 0.000002 - momentum: 0.000000 2023-10-27 20:57:36,527 epoch 7 - iter 1116/3726 - loss 0.02129467 - time (sec): 136.77 - samples/sec: 445.39 - lr: 0.000002 - momentum: 0.000000 2023-10-27 20:58:21,642 epoch 7 - iter 1488/3726 - loss 0.02001426 - time (sec): 181.88 - samples/sec: 447.89 - lr: 0.000002 - momentum: 0.000000 2023-10-27 20:59:08,061 epoch 7 - iter 1860/3726 - loss 0.01894813 - time (sec): 228.30 - samples/sec: 445.16 - lr: 0.000002 - momentum: 0.000000 2023-10-27 20:59:53,051 epoch 7 - iter 2232/3726 - loss 0.01829151 - time (sec): 273.29 - samples/sec: 443.22 - lr: 0.000002 - momentum: 0.000000 2023-10-27 21:00:38,919 epoch 7 - iter 2604/3726 - loss 0.01783981 - time (sec): 319.16 - samples/sec: 442.58 - lr: 0.000002 - momentum: 0.000000 2023-10-27 21:01:26,075 epoch 7 - iter 2976/3726 - loss 0.01776618 - time (sec): 366.32 - samples/sec: 442.68 - lr: 0.000002 - momentum: 0.000000 2023-10-27 21:02:13,955 epoch 7 - iter 3348/3726 - loss 0.01772398 - time (sec): 414.20 - samples/sec: 442.24 - lr: 0.000002 - momentum: 0.000000 2023-10-27 21:03:01,418 epoch 7 - iter 3720/3726 - loss 0.01723102 - time (sec): 461.66 - samples/sec: 442.49 - lr: 0.000002 - momentum: 0.000000 2023-10-27 21:03:02,183 ---------------------------------------------------------------------------------------------------- 2023-10-27 21:03:02,184 EPOCH 7 done: loss 0.0177 - lr: 0.000002 2023-10-27 21:03:26,361 DEV : loss 0.05419960245490074 - f1-score (micro avg) 0.9746 2023-10-27 21:03:26,416 saving best model 2023-10-27 21:03:29,497 ---------------------------------------------------------------------------------------------------- 2023-10-27 21:04:16,746 epoch 8 - iter 372/3726 - loss 0.01736122 - time (sec): 47.25 - samples/sec: 425.05 - lr: 0.000002 - momentum: 0.000000 2023-10-27 21:05:04,978 epoch 8 - iter 744/3726 - loss 0.01398385 - time (sec): 95.48 - samples/sec: 422.25 - lr: 0.000002 - momentum: 0.000000 2023-10-27 21:05:52,318 epoch 8 - iter 1116/3726 - loss 0.01274088 - time (sec): 142.82 - samples/sec: 424.69 - lr: 0.000002 - momentum: 0.000000 2023-10-27 21:06:39,260 epoch 8 - iter 1488/3726 - loss 0.01328050 - time (sec): 189.76 - samples/sec: 424.24 - lr: 0.000001 - momentum: 0.000000 2023-10-27 21:07:26,410 epoch 8 - iter 1860/3726 - loss 0.01227844 - time (sec): 236.91 - samples/sec: 427.47 - lr: 0.000001 - momentum: 0.000000 2023-10-27 21:08:13,172 epoch 8 - iter 2232/3726 - loss 0.01171643 - time (sec): 283.67 - samples/sec: 428.79 - lr: 0.000001 - momentum: 0.000000 2023-10-27 21:09:00,982 epoch 8 - iter 2604/3726 - loss 0.01235731 - time (sec): 331.48 - samples/sec: 428.53 - lr: 0.000001 - momentum: 0.000000 2023-10-27 21:09:51,061 epoch 8 - iter 2976/3726 - loss 0.01221098 - time (sec): 381.56 - samples/sec: 426.25 - lr: 0.000001 - momentum: 0.000000 2023-10-27 21:10:41,073 epoch 8 - iter 3348/3726 - loss 0.01227301 - time (sec): 431.57 - samples/sec: 426.32 - lr: 0.000001 - momentum: 0.000000 2023-10-27 21:11:30,225 epoch 8 - iter 3720/3726 - loss 0.01200497 - time (sec): 480.72 - samples/sec: 424.99 - lr: 0.000001 - momentum: 0.000000 2023-10-27 21:11:31,013 ---------------------------------------------------------------------------------------------------- 2023-10-27 21:11:31,013 EPOCH 8 done: loss 0.0120 - lr: 0.000001 2023-10-27 21:11:56,731 DEV : loss 0.05550903454422951 - f1-score (micro avg) 0.9746 2023-10-27 21:11:56,806 ---------------------------------------------------------------------------------------------------- 2023-10-27 21:12:46,264 epoch 9 - iter 372/3726 - loss 0.00563766 - time (sec): 49.46 - samples/sec: 405.90 - lr: 0.000001 - momentum: 0.000000 2023-10-27 21:13:36,129 epoch 9 - iter 744/3726 - loss 0.00454582 - time (sec): 99.32 - samples/sec: 411.65 - lr: 0.000001 - momentum: 0.000000 2023-10-27 21:14:26,227 epoch 9 - iter 1116/3726 - loss 0.00553718 - time (sec): 149.42 - samples/sec: 408.79 - lr: 0.000001 - momentum: 0.000000 2023-10-27 21:15:15,608 epoch 9 - iter 1488/3726 - loss 0.00675128 - time (sec): 198.80 - samples/sec: 409.19 - lr: 0.000001 - momentum: 0.000000 2023-10-27 21:16:05,355 epoch 9 - iter 1860/3726 - loss 0.00722006 - time (sec): 248.55 - samples/sec: 412.39 - lr: 0.000001 - momentum: 0.000000 2023-10-27 21:16:54,643 epoch 9 - iter 2232/3726 - loss 0.00736249 - time (sec): 297.83 - samples/sec: 411.62 - lr: 0.000001 - momentum: 0.000000 2023-10-27 21:17:45,323 epoch 9 - iter 2604/3726 - loss 0.00786494 - time (sec): 348.51 - samples/sec: 410.57 - lr: 0.000001 - momentum: 0.000000 2023-10-27 21:18:35,586 epoch 9 - iter 2976/3726 - loss 0.00784383 - time (sec): 398.78 - samples/sec: 410.83 - lr: 0.000001 - momentum: 0.000000 2023-10-27 21:19:25,103 epoch 9 - iter 3348/3726 - loss 0.00763218 - time (sec): 448.29 - samples/sec: 409.68 - lr: 0.000001 - momentum: 0.000000 2023-10-27 21:20:15,011 epoch 9 - iter 3720/3726 - loss 0.00729659 - time (sec): 498.20 - samples/sec: 409.86 - lr: 0.000001 - momentum: 0.000000 2023-10-27 21:20:15,793 ---------------------------------------------------------------------------------------------------- 2023-10-27 21:20:15,793 EPOCH 9 done: loss 0.0073 - lr: 0.000001 2023-10-27 21:20:41,491 DEV : loss 0.056521423161029816 - f1-score (micro avg) 0.9737 2023-10-27 21:20:41,559 ---------------------------------------------------------------------------------------------------- 2023-10-27 21:21:31,209 epoch 10 - iter 372/3726 - loss 0.00974983 - time (sec): 49.65 - samples/sec: 405.18 - lr: 0.000001 - momentum: 0.000000 2023-10-27 21:22:20,871 epoch 10 - iter 744/3726 - loss 0.00598632 - time (sec): 99.31 - samples/sec: 409.05 - lr: 0.000000 - momentum: 0.000000 2023-10-27 21:23:10,531 epoch 10 - iter 1116/3726 - loss 0.00650431 - time (sec): 148.97 - samples/sec: 415.14 - lr: 0.000000 - momentum: 0.000000 2023-10-27 21:24:00,035 epoch 10 - iter 1488/3726 - loss 0.00622288 - time (sec): 198.47 - samples/sec: 415.10 - lr: 0.000000 - momentum: 0.000000 2023-10-27 21:24:50,574 epoch 10 - iter 1860/3726 - loss 0.00650968 - time (sec): 249.01 - samples/sec: 412.96 - lr: 0.000000 - momentum: 0.000000 2023-10-27 21:25:40,130 epoch 10 - iter 2232/3726 - loss 0.00707901 - time (sec): 298.57 - samples/sec: 412.19 - lr: 0.000000 - momentum: 0.000000 2023-10-27 21:26:30,069 epoch 10 - iter 2604/3726 - loss 0.00707633 - time (sec): 348.51 - samples/sec: 408.38 - lr: 0.000000 - momentum: 0.000000 2023-10-27 21:27:20,314 epoch 10 - iter 2976/3726 - loss 0.00672126 - time (sec): 398.75 - samples/sec: 409.67 - lr: 0.000000 - momentum: 0.000000 2023-10-27 21:28:09,594 epoch 10 - iter 3348/3726 - loss 0.00659107 - time (sec): 448.03 - samples/sec: 408.65 - lr: 0.000000 - momentum: 0.000000 2023-10-27 21:28:58,954 epoch 10 - iter 3720/3726 - loss 0.00650167 - time (sec): 497.39 - samples/sec: 410.70 - lr: 0.000000 - momentum: 0.000000 2023-10-27 21:28:59,752 ---------------------------------------------------------------------------------------------------- 2023-10-27 21:28:59,752 EPOCH 10 done: loss 0.0065 - lr: 0.000000 2023-10-27 21:29:25,442 DEV : loss 0.05730742961168289 - f1-score (micro avg) 0.9744 2023-10-27 21:29:28,531 ---------------------------------------------------------------------------------------------------- 2023-10-27 21:29:28,534 Loading model from best epoch ... 2023-10-27 21:29:38,713 SequenceTagger predicts: Dictionary with 17 tags: O, S-ORG, B-ORG, E-ORG, I-ORG, S-PER, B-PER, E-PER, I-PER, S-LOC, B-LOC, E-LOC, I-LOC, S-MISC, B-MISC, E-MISC, I-MISC 2023-10-27 21:30:03,801 Results: - F-score (micro) 0.9699 - F-score (macro) 0.9647 - Accuracy 0.9567 By class: precision recall f1-score support ORG 0.9662 0.9738 0.9700 1909 PER 0.9956 0.9937 0.9947 1591 LOC 0.9701 0.9632 0.9666 1413 MISC 0.9170 0.9384 0.9276 812 micro avg 0.9682 0.9717 0.9699 5725 macro avg 0.9622 0.9673 0.9647 5725 weighted avg 0.9683 0.9717 0.9700 5725 2023-10-27 21:30:03,801 ----------------------------------------------------------------------------------------------------