2023-10-11 01:45:47,177 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:45:47,179 Model: "SequenceTagger( (embeddings): ByT5Embeddings( (model): T5EncoderModel( (shared): Embedding(384, 1472) (encoder): T5Stack( (embed_tokens): Embedding(384, 1472) (block): ModuleList( (0): T5Block( (layer): ModuleList( (0): T5LayerSelfAttention( (SelfAttention): T5Attention( (q): Linear(in_features=1472, out_features=384, bias=False) (k): Linear(in_features=1472, out_features=384, bias=False) (v): Linear(in_features=1472, out_features=384, bias=False) (o): Linear(in_features=384, out_features=1472, bias=False) (relative_attention_bias): Embedding(32, 6) ) (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) (1): T5LayerFF( (DenseReluDense): T5DenseGatedActDense( (wi_0): Linear(in_features=1472, out_features=3584, bias=False) (wi_1): Linear(in_features=1472, out_features=3584, bias=False) (wo): Linear(in_features=3584, out_features=1472, bias=False) (dropout): Dropout(p=0.1, inplace=False) (act): NewGELUActivation() ) (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) (1-11): 11 x T5Block( (layer): ModuleList( (0): T5LayerSelfAttention( (SelfAttention): T5Attention( (q): Linear(in_features=1472, out_features=384, bias=False) (k): Linear(in_features=1472, out_features=384, bias=False) (v): Linear(in_features=1472, out_features=384, bias=False) (o): Linear(in_features=384, out_features=1472, bias=False) ) (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) (1): T5LayerFF( (DenseReluDense): T5DenseGatedActDense( (wi_0): Linear(in_features=1472, out_features=3584, bias=False) (wi_1): Linear(in_features=1472, out_features=3584, bias=False) (wo): Linear(in_features=3584, out_features=1472, bias=False) (dropout): Dropout(p=0.1, inplace=False) (act): NewGELUActivation() ) (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) ) (final_layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) (locked_dropout): LockedDropout(p=0.5) (linear): Linear(in_features=1472, out_features=17, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-11 01:45:47,179 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:45:47,179 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 2023-10-11 01:45:47,179 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:45:47,180 Train: 1166 sentences 2023-10-11 01:45:47,180 (train_with_dev=False, train_with_test=False) 2023-10-11 01:45:47,180 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:45:47,180 Training Params: 2023-10-11 01:45:47,180 - learning_rate: "0.00016" 2023-10-11 01:45:47,180 - mini_batch_size: "8" 2023-10-11 01:45:47,180 - max_epochs: "10" 2023-10-11 01:45:47,180 - shuffle: "True" 2023-10-11 01:45:47,180 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:45:47,180 Plugins: 2023-10-11 01:45:47,180 - TensorboardLogger 2023-10-11 01:45:47,180 - LinearScheduler | warmup_fraction: '0.1' 2023-10-11 01:45:47,180 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:45:47,180 Final evaluation on model from best epoch (best-model.pt) 2023-10-11 01:45:47,180 - metric: "('micro avg', 'f1-score')" 2023-10-11 01:45:47,181 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:45:47,181 Computation: 2023-10-11 01:45:47,181 - compute on device: cuda:0 2023-10-11 01:45:47,181 - embedding storage: none 2023-10-11 01:45:47,181 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:45:47,181 Model training base path: "hmbench-newseye/fi-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-4" 2023-10-11 01:45:47,181 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:45:47,181 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:45:47,181 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-11 01:45:55,978 epoch 1 - iter 14/146 - loss 2.84799591 - time (sec): 8.80 - samples/sec: 511.07 - lr: 0.000014 - momentum: 0.000000 2023-10-11 01:46:04,528 epoch 1 - iter 28/146 - loss 2.84070737 - time (sec): 17.35 - samples/sec: 495.81 - lr: 0.000030 - momentum: 0.000000 2023-10-11 01:46:12,675 epoch 1 - iter 42/146 - loss 2.82989666 - time (sec): 25.49 - samples/sec: 486.19 - lr: 0.000045 - momentum: 0.000000 2023-10-11 01:46:21,414 epoch 1 - iter 56/146 - loss 2.80652957 - time (sec): 34.23 - samples/sec: 492.38 - lr: 0.000060 - momentum: 0.000000 2023-10-11 01:46:29,846 epoch 1 - iter 70/146 - loss 2.76874544 - time (sec): 42.66 - samples/sec: 483.58 - lr: 0.000076 - momentum: 0.000000 2023-10-11 01:46:38,429 epoch 1 - iter 84/146 - loss 2.70903645 - time (sec): 51.25 - samples/sec: 482.63 - lr: 0.000091 - momentum: 0.000000 2023-10-11 01:46:47,869 epoch 1 - iter 98/146 - loss 2.62745353 - time (sec): 60.69 - samples/sec: 494.94 - lr: 0.000106 - momentum: 0.000000 2023-10-11 01:46:57,104 epoch 1 - iter 112/146 - loss 2.54807138 - time (sec): 69.92 - samples/sec: 494.47 - lr: 0.000122 - momentum: 0.000000 2023-10-11 01:47:05,555 epoch 1 - iter 126/146 - loss 2.47027428 - time (sec): 78.37 - samples/sec: 493.29 - lr: 0.000137 - momentum: 0.000000 2023-10-11 01:47:13,913 epoch 1 - iter 140/146 - loss 2.39191033 - time (sec): 86.73 - samples/sec: 489.96 - lr: 0.000152 - momentum: 0.000000 2023-10-11 01:47:17,745 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:47:17,745 EPOCH 1 done: loss 2.3510 - lr: 0.000152 2023-10-11 01:47:22,883 DEV : loss 1.284703254699707 - f1-score (micro avg) 0.0 2023-10-11 01:47:22,892 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:47:31,469 epoch 2 - iter 14/146 - loss 1.28576135 - time (sec): 8.57 - samples/sec: 484.20 - lr: 0.000158 - momentum: 0.000000 2023-10-11 01:47:40,053 epoch 2 - iter 28/146 - loss 1.19350508 - time (sec): 17.16 - samples/sec: 484.75 - lr: 0.000157 - momentum: 0.000000 2023-10-11 01:47:48,140 epoch 2 - iter 42/146 - loss 1.11461490 - time (sec): 25.25 - samples/sec: 487.08 - lr: 0.000155 - momentum: 0.000000 2023-10-11 01:47:56,860 epoch 2 - iter 56/146 - loss 1.01345969 - time (sec): 33.97 - samples/sec: 500.64 - lr: 0.000153 - momentum: 0.000000 2023-10-11 01:48:05,311 epoch 2 - iter 70/146 - loss 0.95460535 - time (sec): 42.42 - samples/sec: 501.66 - lr: 0.000152 - momentum: 0.000000 2023-10-11 01:48:13,908 epoch 2 - iter 84/146 - loss 0.90405139 - time (sec): 51.01 - samples/sec: 503.12 - lr: 0.000150 - momentum: 0.000000 2023-10-11 01:48:22,825 epoch 2 - iter 98/146 - loss 0.90347909 - time (sec): 59.93 - samples/sec: 508.28 - lr: 0.000148 - momentum: 0.000000 2023-10-11 01:48:31,635 epoch 2 - iter 112/146 - loss 0.87398082 - time (sec): 68.74 - samples/sec: 508.77 - lr: 0.000147 - momentum: 0.000000 2023-10-11 01:48:40,026 epoch 2 - iter 126/146 - loss 0.85172335 - time (sec): 77.13 - samples/sec: 505.74 - lr: 0.000145 - momentum: 0.000000 2023-10-11 01:48:48,193 epoch 2 - iter 140/146 - loss 0.81628531 - time (sec): 85.30 - samples/sec: 501.54 - lr: 0.000143 - momentum: 0.000000 2023-10-11 01:48:51,600 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:48:51,600 EPOCH 2 done: loss 0.8079 - lr: 0.000143 2023-10-11 01:48:57,050 DEV : loss 0.40683862566947937 - f1-score (micro avg) 0.0 2023-10-11 01:48:57,058 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:49:05,067 epoch 3 - iter 14/146 - loss 0.52447831 - time (sec): 8.01 - samples/sec: 448.66 - lr: 0.000141 - momentum: 0.000000 2023-10-11 01:49:13,729 epoch 3 - iter 28/146 - loss 0.46662481 - time (sec): 16.67 - samples/sec: 490.31 - lr: 0.000139 - momentum: 0.000000 2023-10-11 01:49:21,835 epoch 3 - iter 42/146 - loss 0.44349831 - time (sec): 24.77 - samples/sec: 491.32 - lr: 0.000137 - momentum: 0.000000 2023-10-11 01:49:30,221 epoch 3 - iter 56/146 - loss 0.48544884 - time (sec): 33.16 - samples/sec: 494.60 - lr: 0.000136 - momentum: 0.000000 2023-10-11 01:49:39,305 epoch 3 - iter 70/146 - loss 0.45625358 - time (sec): 42.24 - samples/sec: 505.20 - lr: 0.000134 - momentum: 0.000000 2023-10-11 01:49:48,487 epoch 3 - iter 84/146 - loss 0.44065904 - time (sec): 51.43 - samples/sec: 508.94 - lr: 0.000132 - momentum: 0.000000 2023-10-11 01:49:56,651 epoch 3 - iter 98/146 - loss 0.42778850 - time (sec): 59.59 - samples/sec: 504.64 - lr: 0.000131 - momentum: 0.000000 2023-10-11 01:50:04,334 epoch 3 - iter 112/146 - loss 0.41962827 - time (sec): 67.27 - samples/sec: 496.11 - lr: 0.000129 - momentum: 0.000000 2023-10-11 01:50:13,000 epoch 3 - iter 126/146 - loss 0.40808767 - time (sec): 75.94 - samples/sec: 496.87 - lr: 0.000127 - momentum: 0.000000 2023-10-11 01:50:22,043 epoch 3 - iter 140/146 - loss 0.40388098 - time (sec): 84.98 - samples/sec: 499.65 - lr: 0.000125 - momentum: 0.000000 2023-10-11 01:50:25,855 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:50:25,855 EPOCH 3 done: loss 0.4001 - lr: 0.000125 2023-10-11 01:50:31,482 DEV : loss 0.2588852643966675 - f1-score (micro avg) 0.276 2023-10-11 01:50:31,490 saving best model 2023-10-11 01:50:32,587 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:50:41,007 epoch 4 - iter 14/146 - loss 0.24164107 - time (sec): 8.42 - samples/sec: 488.85 - lr: 0.000123 - momentum: 0.000000 2023-10-11 01:50:49,321 epoch 4 - iter 28/146 - loss 0.29146948 - time (sec): 16.73 - samples/sec: 500.12 - lr: 0.000121 - momentum: 0.000000 2023-10-11 01:50:57,496 epoch 4 - iter 42/146 - loss 0.28541708 - time (sec): 24.91 - samples/sec: 500.91 - lr: 0.000120 - momentum: 0.000000 2023-10-11 01:51:05,765 epoch 4 - iter 56/146 - loss 0.27322829 - time (sec): 33.18 - samples/sec: 502.12 - lr: 0.000118 - momentum: 0.000000 2023-10-11 01:51:13,936 epoch 4 - iter 70/146 - loss 0.27476808 - time (sec): 41.35 - samples/sec: 499.65 - lr: 0.000116 - momentum: 0.000000 2023-10-11 01:51:22,618 epoch 4 - iter 84/146 - loss 0.29740996 - time (sec): 50.03 - samples/sec: 502.25 - lr: 0.000115 - momentum: 0.000000 2023-10-11 01:51:31,373 epoch 4 - iter 98/146 - loss 0.30483075 - time (sec): 58.78 - samples/sec: 504.22 - lr: 0.000113 - momentum: 0.000000 2023-10-11 01:51:39,388 epoch 4 - iter 112/146 - loss 0.30252875 - time (sec): 66.80 - samples/sec: 498.73 - lr: 0.000111 - momentum: 0.000000 2023-10-11 01:51:47,732 epoch 4 - iter 126/146 - loss 0.29906065 - time (sec): 75.14 - samples/sec: 495.80 - lr: 0.000109 - momentum: 0.000000 2023-10-11 01:51:56,969 epoch 4 - iter 140/146 - loss 0.29067698 - time (sec): 84.38 - samples/sec: 501.64 - lr: 0.000108 - momentum: 0.000000 2023-10-11 01:52:00,782 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:52:00,782 EPOCH 4 done: loss 0.2895 - lr: 0.000108 2023-10-11 01:52:06,509 DEV : loss 0.2072305679321289 - f1-score (micro avg) 0.4454 2023-10-11 01:52:06,517 saving best model 2023-10-11 01:52:09,093 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:52:18,749 epoch 5 - iter 14/146 - loss 0.30683651 - time (sec): 9.65 - samples/sec: 572.55 - lr: 0.000105 - momentum: 0.000000 2023-10-11 01:52:27,752 epoch 5 - iter 28/146 - loss 0.25854309 - time (sec): 18.65 - samples/sec: 545.82 - lr: 0.000104 - momentum: 0.000000 2023-10-11 01:52:35,868 epoch 5 - iter 42/146 - loss 0.25470375 - time (sec): 26.77 - samples/sec: 517.15 - lr: 0.000102 - momentum: 0.000000 2023-10-11 01:52:44,048 epoch 5 - iter 56/146 - loss 0.24566226 - time (sec): 34.95 - samples/sec: 511.09 - lr: 0.000100 - momentum: 0.000000 2023-10-11 01:52:52,786 epoch 5 - iter 70/146 - loss 0.24130166 - time (sec): 43.69 - samples/sec: 514.10 - lr: 0.000099 - momentum: 0.000000 2023-10-11 01:53:00,635 epoch 5 - iter 84/146 - loss 0.24670391 - time (sec): 51.54 - samples/sec: 507.09 - lr: 0.000097 - momentum: 0.000000 2023-10-11 01:53:09,102 epoch 5 - iter 98/146 - loss 0.23499454 - time (sec): 60.00 - samples/sec: 505.81 - lr: 0.000095 - momentum: 0.000000 2023-10-11 01:53:17,232 epoch 5 - iter 112/146 - loss 0.22971880 - time (sec): 68.13 - samples/sec: 503.34 - lr: 0.000093 - momentum: 0.000000 2023-10-11 01:53:25,052 epoch 5 - iter 126/146 - loss 0.22515418 - time (sec): 75.95 - samples/sec: 498.05 - lr: 0.000092 - momentum: 0.000000 2023-10-11 01:53:33,478 epoch 5 - iter 140/146 - loss 0.22370291 - time (sec): 84.38 - samples/sec: 498.10 - lr: 0.000090 - momentum: 0.000000 2023-10-11 01:53:37,472 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:53:37,472 EPOCH 5 done: loss 0.2231 - lr: 0.000090 2023-10-11 01:53:43,406 DEV : loss 0.18223470449447632 - f1-score (micro avg) 0.5156 2023-10-11 01:53:43,416 saving best model 2023-10-11 01:53:45,986 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:53:55,576 epoch 6 - iter 14/146 - loss 0.21654807 - time (sec): 9.59 - samples/sec: 550.21 - lr: 0.000088 - momentum: 0.000000 2023-10-11 01:54:04,225 epoch 6 - iter 28/146 - loss 0.17901612 - time (sec): 18.23 - samples/sec: 535.03 - lr: 0.000086 - momentum: 0.000000 2023-10-11 01:54:12,266 epoch 6 - iter 42/146 - loss 0.17103159 - time (sec): 26.28 - samples/sec: 511.31 - lr: 0.000084 - momentum: 0.000000 2023-10-11 01:54:21,012 epoch 6 - iter 56/146 - loss 0.15995008 - time (sec): 35.02 - samples/sec: 508.91 - lr: 0.000083 - momentum: 0.000000 2023-10-11 01:54:29,434 epoch 6 - iter 70/146 - loss 0.16068347 - time (sec): 43.44 - samples/sec: 494.68 - lr: 0.000081 - momentum: 0.000000 2023-10-11 01:54:38,141 epoch 6 - iter 84/146 - loss 0.17578810 - time (sec): 52.15 - samples/sec: 494.74 - lr: 0.000079 - momentum: 0.000000 2023-10-11 01:54:48,217 epoch 6 - iter 98/146 - loss 0.16726653 - time (sec): 62.23 - samples/sec: 502.60 - lr: 0.000077 - momentum: 0.000000 2023-10-11 01:54:57,061 epoch 6 - iter 112/146 - loss 0.16948194 - time (sec): 71.07 - samples/sec: 491.93 - lr: 0.000076 - momentum: 0.000000 2023-10-11 01:55:06,565 epoch 6 - iter 126/146 - loss 0.16941279 - time (sec): 80.57 - samples/sec: 480.60 - lr: 0.000074 - momentum: 0.000000 2023-10-11 01:55:16,251 epoch 6 - iter 140/146 - loss 0.16809766 - time (sec): 90.26 - samples/sec: 472.65 - lr: 0.000072 - momentum: 0.000000 2023-10-11 01:55:20,370 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:55:20,371 EPOCH 6 done: loss 0.1675 - lr: 0.000072 2023-10-11 01:55:26,683 DEV : loss 0.15008610486984253 - f1-score (micro avg) 0.63 2023-10-11 01:55:26,693 saving best model 2023-10-11 01:55:34,594 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:55:43,486 epoch 7 - iter 14/146 - loss 0.15174053 - time (sec): 8.89 - samples/sec: 406.96 - lr: 0.000070 - momentum: 0.000000 2023-10-11 01:55:52,482 epoch 7 - iter 28/146 - loss 0.13072503 - time (sec): 17.88 - samples/sec: 450.28 - lr: 0.000068 - momentum: 0.000000 2023-10-11 01:56:01,097 epoch 7 - iter 42/146 - loss 0.13179409 - time (sec): 26.50 - samples/sec: 442.24 - lr: 0.000067 - momentum: 0.000000 2023-10-11 01:56:11,670 epoch 7 - iter 56/146 - loss 0.12405555 - time (sec): 37.07 - samples/sec: 442.31 - lr: 0.000065 - momentum: 0.000000 2023-10-11 01:56:21,134 epoch 7 - iter 70/146 - loss 0.12815382 - time (sec): 46.54 - samples/sec: 461.77 - lr: 0.000063 - momentum: 0.000000 2023-10-11 01:56:29,304 epoch 7 - iter 84/146 - loss 0.13476612 - time (sec): 54.71 - samples/sec: 460.73 - lr: 0.000061 - momentum: 0.000000 2023-10-11 01:56:38,680 epoch 7 - iter 98/146 - loss 0.13398473 - time (sec): 64.08 - samples/sec: 469.26 - lr: 0.000060 - momentum: 0.000000 2023-10-11 01:56:47,734 epoch 7 - iter 112/146 - loss 0.13242540 - time (sec): 73.14 - samples/sec: 468.51 - lr: 0.000058 - momentum: 0.000000 2023-10-11 01:56:56,401 epoch 7 - iter 126/146 - loss 0.12813846 - time (sec): 81.80 - samples/sec: 469.65 - lr: 0.000056 - momentum: 0.000000 2023-10-11 01:57:05,171 epoch 7 - iter 140/146 - loss 0.13180592 - time (sec): 90.57 - samples/sec: 471.51 - lr: 0.000055 - momentum: 0.000000 2023-10-11 01:57:08,863 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:57:08,863 EPOCH 7 done: loss 0.1298 - lr: 0.000055 2023-10-11 01:57:14,752 DEV : loss 0.14380821585655212 - f1-score (micro avg) 0.6953 2023-10-11 01:57:14,762 saving best model 2023-10-11 01:57:24,219 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:57:33,086 epoch 8 - iter 14/146 - loss 0.12831157 - time (sec): 8.86 - samples/sec: 474.89 - lr: 0.000052 - momentum: 0.000000 2023-10-11 01:57:42,240 epoch 8 - iter 28/146 - loss 0.10696556 - time (sec): 18.02 - samples/sec: 495.38 - lr: 0.000051 - momentum: 0.000000 2023-10-11 01:57:50,350 epoch 8 - iter 42/146 - loss 0.10681912 - time (sec): 26.13 - samples/sec: 473.89 - lr: 0.000049 - momentum: 0.000000 2023-10-11 01:57:59,658 epoch 8 - iter 56/146 - loss 0.11273648 - time (sec): 35.43 - samples/sec: 471.03 - lr: 0.000047 - momentum: 0.000000 2023-10-11 01:58:09,418 epoch 8 - iter 70/146 - loss 0.10310746 - time (sec): 45.19 - samples/sec: 478.00 - lr: 0.000045 - momentum: 0.000000 2023-10-11 01:58:18,799 epoch 8 - iter 84/146 - loss 0.10691084 - time (sec): 54.58 - samples/sec: 480.45 - lr: 0.000044 - momentum: 0.000000 2023-10-11 01:58:27,931 epoch 8 - iter 98/146 - loss 0.10879791 - time (sec): 63.71 - samples/sec: 482.22 - lr: 0.000042 - momentum: 0.000000 2023-10-11 01:58:36,616 epoch 8 - iter 112/146 - loss 0.10888580 - time (sec): 72.39 - samples/sec: 478.46 - lr: 0.000040 - momentum: 0.000000 2023-10-11 01:58:45,450 epoch 8 - iter 126/146 - loss 0.10711927 - time (sec): 81.23 - samples/sec: 471.58 - lr: 0.000039 - momentum: 0.000000 2023-10-11 01:58:55,017 epoch 8 - iter 140/146 - loss 0.10900138 - time (sec): 90.79 - samples/sec: 470.50 - lr: 0.000037 - momentum: 0.000000 2023-10-11 01:58:58,974 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:58:58,975 EPOCH 8 done: loss 0.1099 - lr: 0.000037 2023-10-11 01:59:04,887 DEV : loss 0.13584169745445251 - f1-score (micro avg) 0.7712 2023-10-11 01:59:04,897 saving best model 2023-10-11 01:59:11,021 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:59:19,814 epoch 9 - iter 14/146 - loss 0.08415165 - time (sec): 8.79 - samples/sec: 475.62 - lr: 0.000035 - momentum: 0.000000 2023-10-11 01:59:28,488 epoch 9 - iter 28/146 - loss 0.09301484 - time (sec): 17.46 - samples/sec: 472.65 - lr: 0.000033 - momentum: 0.000000 2023-10-11 01:59:37,554 epoch 9 - iter 42/146 - loss 0.08023038 - time (sec): 26.53 - samples/sec: 483.47 - lr: 0.000031 - momentum: 0.000000 2023-10-11 01:59:46,510 epoch 9 - iter 56/146 - loss 0.08357733 - time (sec): 35.48 - samples/sec: 480.43 - lr: 0.000029 - momentum: 0.000000 2023-10-11 01:59:55,072 epoch 9 - iter 70/146 - loss 0.08569746 - time (sec): 44.05 - samples/sec: 485.46 - lr: 0.000028 - momentum: 0.000000 2023-10-11 02:00:03,593 epoch 9 - iter 84/146 - loss 0.09152495 - time (sec): 52.57 - samples/sec: 487.52 - lr: 0.000026 - momentum: 0.000000 2023-10-11 02:00:12,273 epoch 9 - iter 98/146 - loss 0.09293580 - time (sec): 61.25 - samples/sec: 487.05 - lr: 0.000024 - momentum: 0.000000 2023-10-11 02:00:21,641 epoch 9 - iter 112/146 - loss 0.09202586 - time (sec): 70.62 - samples/sec: 494.33 - lr: 0.000023 - momentum: 0.000000 2023-10-11 02:00:30,382 epoch 9 - iter 126/146 - loss 0.09556408 - time (sec): 79.36 - samples/sec: 489.70 - lr: 0.000021 - momentum: 0.000000 2023-10-11 02:00:39,304 epoch 9 - iter 140/146 - loss 0.09594021 - time (sec): 88.28 - samples/sec: 488.90 - lr: 0.000019 - momentum: 0.000000 2023-10-11 02:00:42,560 ---------------------------------------------------------------------------------------------------- 2023-10-11 02:00:42,560 EPOCH 9 done: loss 0.0978 - lr: 0.000019 2023-10-11 02:00:48,591 DEV : loss 0.13022395968437195 - f1-score (micro avg) 0.7804 2023-10-11 02:00:48,600 saving best model 2023-10-11 02:00:57,702 ---------------------------------------------------------------------------------------------------- 2023-10-11 02:01:06,932 epoch 10 - iter 14/146 - loss 0.09334435 - time (sec): 9.23 - samples/sec: 547.15 - lr: 0.000017 - momentum: 0.000000 2023-10-11 02:01:15,693 epoch 10 - iter 28/146 - loss 0.09187192 - time (sec): 17.99 - samples/sec: 518.99 - lr: 0.000015 - momentum: 0.000000 2023-10-11 02:01:24,775 epoch 10 - iter 42/146 - loss 0.08406622 - time (sec): 27.07 - samples/sec: 515.09 - lr: 0.000013 - momentum: 0.000000 2023-10-11 02:01:32,761 epoch 10 - iter 56/146 - loss 0.09189053 - time (sec): 35.05 - samples/sec: 499.56 - lr: 0.000012 - momentum: 0.000000 2023-10-11 02:01:41,775 epoch 10 - iter 70/146 - loss 0.09213464 - time (sec): 44.07 - samples/sec: 501.69 - lr: 0.000010 - momentum: 0.000000 2023-10-11 02:01:51,096 epoch 10 - iter 84/146 - loss 0.08675532 - time (sec): 53.39 - samples/sec: 503.47 - lr: 0.000008 - momentum: 0.000000 2023-10-11 02:01:59,298 epoch 10 - iter 98/146 - loss 0.08770465 - time (sec): 61.59 - samples/sec: 496.31 - lr: 0.000007 - momentum: 0.000000 2023-10-11 02:02:07,665 epoch 10 - iter 112/146 - loss 0.08846977 - time (sec): 69.96 - samples/sec: 489.99 - lr: 0.000005 - momentum: 0.000000 2023-10-11 02:02:16,158 epoch 10 - iter 126/146 - loss 0.08936363 - time (sec): 78.45 - samples/sec: 488.05 - lr: 0.000003 - momentum: 0.000000 2023-10-11 02:02:24,963 epoch 10 - iter 140/146 - loss 0.09235372 - time (sec): 87.26 - samples/sec: 490.59 - lr: 0.000002 - momentum: 0.000000 2023-10-11 02:02:28,301 ---------------------------------------------------------------------------------------------------- 2023-10-11 02:02:28,301 EPOCH 10 done: loss 0.0909 - lr: 0.000002 2023-10-11 02:02:34,184 DEV : loss 0.1293526142835617 - f1-score (micro avg) 0.7702 2023-10-11 02:02:35,042 ---------------------------------------------------------------------------------------------------- 2023-10-11 02:02:35,044 Loading model from best epoch ... 2023-10-11 02:02:39,160 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 2023-10-11 02:02:52,020 Results: - F-score (micro) 0.7274 - F-score (macro) 0.6818 - Accuracy 0.5901 By class: precision recall f1-score support PER 0.7745 0.8190 0.7961 348 LOC 0.6258 0.7816 0.6951 261 ORG 0.3800 0.3654 0.3725 52 HumanProd 0.8636 0.8636 0.8636 22 micro avg 0.6880 0.7716 0.7274 683 macro avg 0.6610 0.7074 0.6818 683 weighted avg 0.6905 0.7716 0.7274 683 2023-10-11 02:02:52,020 ----------------------------------------------------------------------------------------------------