2023-10-10 22:38:50,022 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:38:50,024 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() )" 2023-10-10 22:38:50,025 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:38:50,025 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 2023-10-10 22:38:50,025 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:38:50,025 Train: 1166 sentences 2023-10-10 22:38:50,025 (train_with_dev=False, train_with_test=False) 2023-10-10 22:38:50,025 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:38:50,025 Training Params: 2023-10-10 22:38:50,025 - learning_rate: "0.00016" 2023-10-10 22:38:50,025 - mini_batch_size: "4" 2023-10-10 22:38:50,025 - max_epochs: "10" 2023-10-10 22:38:50,025 - shuffle: "True" 2023-10-10 22:38:50,025 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:38:50,025 Plugins: 2023-10-10 22:38:50,026 - TensorboardLogger 2023-10-10 22:38:50,026 - LinearScheduler | warmup_fraction: '0.1' 2023-10-10 22:38:50,026 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:38:50,026 Final evaluation on model from best epoch (best-model.pt) 2023-10-10 22:38:50,026 - metric: "('micro avg', 'f1-score')" 2023-10-10 22:38:50,026 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:38:50,026 Computation: 2023-10-10 22:38:50,026 - compute on device: cuda:0 2023-10-10 22:38:50,026 - embedding storage: none 2023-10-10 22:38:50,026 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:38:50,026 Model training base path: "hmbench-newseye/fi-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-1" 2023-10-10 22:38:50,026 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:38:50,026 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:38:50,026 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-10 22:38:59,893 epoch 1 - iter 29/292 - loss 2.82904011 - time (sec): 9.86 - samples/sec: 517.82 - lr: 0.000015 - momentum: 0.000000 2023-10-10 22:39:08,951 epoch 1 - iter 58/292 - loss 2.82009083 - time (sec): 18.92 - samples/sec: 482.18 - lr: 0.000031 - momentum: 0.000000 2023-10-10 22:39:19,259 epoch 1 - iter 87/292 - loss 2.79608502 - time (sec): 29.23 - samples/sec: 501.01 - lr: 0.000047 - momentum: 0.000000 2023-10-10 22:39:28,698 epoch 1 - iter 116/292 - loss 2.75939629 - time (sec): 38.67 - samples/sec: 482.24 - lr: 0.000063 - momentum: 0.000000 2023-10-10 22:39:38,409 epoch 1 - iter 145/292 - loss 2.68289368 - time (sec): 48.38 - samples/sec: 465.91 - lr: 0.000079 - momentum: 0.000000 2023-10-10 22:39:47,654 epoch 1 - iter 174/292 - loss 2.58924155 - time (sec): 57.63 - samples/sec: 455.94 - lr: 0.000095 - momentum: 0.000000 2023-10-10 22:39:57,761 epoch 1 - iter 203/292 - loss 2.45043910 - time (sec): 67.73 - samples/sec: 458.95 - lr: 0.000111 - momentum: 0.000000 2023-10-10 22:40:08,012 epoch 1 - iter 232/292 - loss 2.32335868 - time (sec): 77.98 - samples/sec: 458.42 - lr: 0.000127 - momentum: 0.000000 2023-10-10 22:40:17,693 epoch 1 - iter 261/292 - loss 2.20131381 - time (sec): 87.66 - samples/sec: 455.90 - lr: 0.000142 - momentum: 0.000000 2023-10-10 22:40:27,577 epoch 1 - iter 290/292 - loss 2.08347326 - time (sec): 97.55 - samples/sec: 453.75 - lr: 0.000158 - momentum: 0.000000 2023-10-10 22:40:28,055 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:40:28,055 EPOCH 1 done: loss 2.0787 - lr: 0.000158 2023-10-10 22:40:33,670 DEV : loss 0.6691536903381348 - f1-score (micro avg) 0.0 2023-10-10 22:40:33,679 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:40:43,569 epoch 2 - iter 29/292 - loss 0.73979818 - time (sec): 9.89 - samples/sec: 473.00 - lr: 0.000158 - momentum: 0.000000 2023-10-10 22:40:53,455 epoch 2 - iter 58/292 - loss 0.79010563 - time (sec): 19.77 - samples/sec: 464.71 - lr: 0.000157 - momentum: 0.000000 2023-10-10 22:41:03,434 epoch 2 - iter 87/292 - loss 0.68326802 - time (sec): 29.75 - samples/sec: 460.46 - lr: 0.000155 - momentum: 0.000000 2023-10-10 22:41:13,798 epoch 2 - iter 116/292 - loss 0.60934981 - time (sec): 40.12 - samples/sec: 460.43 - lr: 0.000153 - momentum: 0.000000 2023-10-10 22:41:24,470 epoch 2 - iter 145/292 - loss 0.58125380 - time (sec): 50.79 - samples/sec: 463.84 - lr: 0.000151 - momentum: 0.000000 2023-10-10 22:41:34,361 epoch 2 - iter 174/292 - loss 0.54661045 - time (sec): 60.68 - samples/sec: 456.43 - lr: 0.000149 - momentum: 0.000000 2023-10-10 22:41:44,174 epoch 2 - iter 203/292 - loss 0.53389626 - time (sec): 70.49 - samples/sec: 447.60 - lr: 0.000148 - momentum: 0.000000 2023-10-10 22:41:54,010 epoch 2 - iter 232/292 - loss 0.51870783 - time (sec): 80.33 - samples/sec: 443.48 - lr: 0.000146 - momentum: 0.000000 2023-10-10 22:42:03,637 epoch 2 - iter 261/292 - loss 0.49765927 - time (sec): 89.96 - samples/sec: 445.87 - lr: 0.000144 - momentum: 0.000000 2023-10-10 22:42:13,850 epoch 2 - iter 290/292 - loss 0.49732765 - time (sec): 100.17 - samples/sec: 442.98 - lr: 0.000142 - momentum: 0.000000 2023-10-10 22:42:14,221 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:42:14,222 EPOCH 2 done: loss 0.4968 - lr: 0.000142 2023-10-10 22:42:20,227 DEV : loss 0.28831687569618225 - f1-score (micro avg) 0.1468 2023-10-10 22:42:20,237 saving best model 2023-10-10 22:42:21,190 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:42:30,312 epoch 3 - iter 29/292 - loss 0.37250554 - time (sec): 9.12 - samples/sec: 407.06 - lr: 0.000141 - momentum: 0.000000 2023-10-10 22:42:41,368 epoch 3 - iter 58/292 - loss 0.28788479 - time (sec): 20.18 - samples/sec: 440.44 - lr: 0.000139 - momentum: 0.000000 2023-10-10 22:42:50,894 epoch 3 - iter 87/292 - loss 0.31377648 - time (sec): 29.70 - samples/sec: 440.72 - lr: 0.000137 - momentum: 0.000000 2023-10-10 22:43:00,160 epoch 3 - iter 116/292 - loss 0.30860177 - time (sec): 38.97 - samples/sec: 439.06 - lr: 0.000135 - momentum: 0.000000 2023-10-10 22:43:09,738 epoch 3 - iter 145/292 - loss 0.30045066 - time (sec): 48.54 - samples/sec: 444.89 - lr: 0.000133 - momentum: 0.000000 2023-10-10 22:43:18,762 epoch 3 - iter 174/292 - loss 0.30780945 - time (sec): 57.57 - samples/sec: 444.68 - lr: 0.000132 - momentum: 0.000000 2023-10-10 22:43:29,441 epoch 3 - iter 203/292 - loss 0.31651504 - time (sec): 68.25 - samples/sec: 455.95 - lr: 0.000130 - momentum: 0.000000 2023-10-10 22:43:39,300 epoch 3 - iter 232/292 - loss 0.31485026 - time (sec): 78.11 - samples/sec: 456.80 - lr: 0.000128 - momentum: 0.000000 2023-10-10 22:43:49,319 epoch 3 - iter 261/292 - loss 0.30698124 - time (sec): 88.13 - samples/sec: 459.46 - lr: 0.000126 - momentum: 0.000000 2023-10-10 22:43:58,539 epoch 3 - iter 290/292 - loss 0.30215031 - time (sec): 97.35 - samples/sec: 455.07 - lr: 0.000125 - momentum: 0.000000 2023-10-10 22:43:58,987 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:43:58,988 EPOCH 3 done: loss 0.3015 - lr: 0.000125 2023-10-10 22:44:05,062 DEV : loss 0.21253274381160736 - f1-score (micro avg) 0.4454 2023-10-10 22:44:05,072 saving best model 2023-10-10 22:44:14,113 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:44:24,977 epoch 4 - iter 29/292 - loss 0.20247520 - time (sec): 10.86 - samples/sec: 460.60 - lr: 0.000123 - momentum: 0.000000 2023-10-10 22:44:34,949 epoch 4 - iter 58/292 - loss 0.18905862 - time (sec): 20.83 - samples/sec: 437.66 - lr: 0.000121 - momentum: 0.000000 2023-10-10 22:44:45,648 epoch 4 - iter 87/292 - loss 0.22813786 - time (sec): 31.53 - samples/sec: 444.87 - lr: 0.000119 - momentum: 0.000000 2023-10-10 22:44:55,775 epoch 4 - iter 116/292 - loss 0.23168511 - time (sec): 41.66 - samples/sec: 441.17 - lr: 0.000117 - momentum: 0.000000 2023-10-10 22:45:06,379 epoch 4 - iter 145/292 - loss 0.22417524 - time (sec): 52.26 - samples/sec: 437.63 - lr: 0.000116 - momentum: 0.000000 2023-10-10 22:45:16,848 epoch 4 - iter 174/292 - loss 0.22225430 - time (sec): 62.73 - samples/sec: 431.91 - lr: 0.000114 - momentum: 0.000000 2023-10-10 22:45:27,869 epoch 4 - iter 203/292 - loss 0.21451680 - time (sec): 73.75 - samples/sec: 430.08 - lr: 0.000112 - momentum: 0.000000 2023-10-10 22:45:37,709 epoch 4 - iter 232/292 - loss 0.21129060 - time (sec): 83.59 - samples/sec: 431.92 - lr: 0.000110 - momentum: 0.000000 2023-10-10 22:45:47,385 epoch 4 - iter 261/292 - loss 0.20883694 - time (sec): 93.27 - samples/sec: 425.79 - lr: 0.000109 - momentum: 0.000000 2023-10-10 22:45:57,486 epoch 4 - iter 290/292 - loss 0.20421114 - time (sec): 103.37 - samples/sec: 427.91 - lr: 0.000107 - momentum: 0.000000 2023-10-10 22:45:57,953 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:45:57,954 EPOCH 4 done: loss 0.2034 - lr: 0.000107 2023-10-10 22:46:03,949 DEV : loss 0.16762110590934753 - f1-score (micro avg) 0.636 2023-10-10 22:46:03,958 saving best model 2023-10-10 22:46:12,656 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:46:22,429 epoch 5 - iter 29/292 - loss 0.15228624 - time (sec): 9.77 - samples/sec: 444.71 - lr: 0.000105 - momentum: 0.000000 2023-10-10 22:46:33,178 epoch 5 - iter 58/292 - loss 0.16824065 - time (sec): 20.52 - samples/sec: 456.34 - lr: 0.000103 - momentum: 0.000000 2023-10-10 22:46:42,550 epoch 5 - iter 87/292 - loss 0.16516610 - time (sec): 29.89 - samples/sec: 454.50 - lr: 0.000101 - momentum: 0.000000 2023-10-10 22:46:52,308 epoch 5 - iter 116/292 - loss 0.14358303 - time (sec): 39.65 - samples/sec: 451.98 - lr: 0.000100 - momentum: 0.000000 2023-10-10 22:47:02,466 epoch 5 - iter 145/292 - loss 0.14189975 - time (sec): 49.81 - samples/sec: 456.77 - lr: 0.000098 - momentum: 0.000000 2023-10-10 22:47:11,868 epoch 5 - iter 174/292 - loss 0.13815073 - time (sec): 59.21 - samples/sec: 447.30 - lr: 0.000096 - momentum: 0.000000 2023-10-10 22:47:22,402 epoch 5 - iter 203/292 - loss 0.13687599 - time (sec): 69.74 - samples/sec: 450.72 - lr: 0.000094 - momentum: 0.000000 2023-10-10 22:47:32,919 epoch 5 - iter 232/292 - loss 0.13663132 - time (sec): 80.26 - samples/sec: 448.91 - lr: 0.000093 - momentum: 0.000000 2023-10-10 22:47:42,044 epoch 5 - iter 261/292 - loss 0.13323399 - time (sec): 89.38 - samples/sec: 444.34 - lr: 0.000091 - momentum: 0.000000 2023-10-10 22:47:52,280 epoch 5 - iter 290/292 - loss 0.13201349 - time (sec): 99.62 - samples/sec: 445.37 - lr: 0.000089 - momentum: 0.000000 2023-10-10 22:47:52,695 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:47:52,696 EPOCH 5 done: loss 0.1319 - lr: 0.000089 2023-10-10 22:47:58,790 DEV : loss 0.1449754387140274 - f1-score (micro avg) 0.7834 2023-10-10 22:47:58,800 saving best model 2023-10-10 22:48:08,053 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:48:18,620 epoch 6 - iter 29/292 - loss 0.09780191 - time (sec): 10.56 - samples/sec: 508.71 - lr: 0.000087 - momentum: 0.000000 2023-10-10 22:48:27,513 epoch 6 - iter 58/292 - loss 0.10813143 - time (sec): 19.46 - samples/sec: 468.11 - lr: 0.000085 - momentum: 0.000000 2023-10-10 22:48:38,005 epoch 6 - iter 87/292 - loss 0.09566991 - time (sec): 29.95 - samples/sec: 464.73 - lr: 0.000084 - momentum: 0.000000 2023-10-10 22:48:47,930 epoch 6 - iter 116/292 - loss 0.09010732 - time (sec): 39.87 - samples/sec: 463.92 - lr: 0.000082 - momentum: 0.000000 2023-10-10 22:48:57,497 epoch 6 - iter 145/292 - loss 0.09093065 - time (sec): 49.44 - samples/sec: 463.49 - lr: 0.000080 - momentum: 0.000000 2023-10-10 22:49:06,968 epoch 6 - iter 174/292 - loss 0.09436299 - time (sec): 58.91 - samples/sec: 460.52 - lr: 0.000078 - momentum: 0.000000 2023-10-10 22:49:16,957 epoch 6 - iter 203/292 - loss 0.09269008 - time (sec): 68.90 - samples/sec: 448.05 - lr: 0.000077 - momentum: 0.000000 2023-10-10 22:49:27,267 epoch 6 - iter 232/292 - loss 0.09292884 - time (sec): 79.21 - samples/sec: 441.90 - lr: 0.000075 - momentum: 0.000000 2023-10-10 22:49:38,103 epoch 6 - iter 261/292 - loss 0.09295591 - time (sec): 90.05 - samples/sec: 442.24 - lr: 0.000073 - momentum: 0.000000 2023-10-10 22:49:48,309 epoch 6 - iter 290/292 - loss 0.09196060 - time (sec): 100.25 - samples/sec: 440.11 - lr: 0.000071 - momentum: 0.000000 2023-10-10 22:49:48,896 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:49:48,897 EPOCH 6 done: loss 0.0923 - lr: 0.000071 2023-10-10 22:49:55,279 DEV : loss 0.1280345916748047 - f1-score (micro avg) 0.7843 2023-10-10 22:49:55,289 saving best model 2023-10-10 22:49:58,465 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:50:09,512 epoch 7 - iter 29/292 - loss 0.07453827 - time (sec): 11.04 - samples/sec: 443.98 - lr: 0.000069 - momentum: 0.000000 2023-10-10 22:50:19,617 epoch 7 - iter 58/292 - loss 0.06721636 - time (sec): 21.15 - samples/sec: 411.62 - lr: 0.000068 - momentum: 0.000000 2023-10-10 22:50:29,934 epoch 7 - iter 87/292 - loss 0.07117217 - time (sec): 31.46 - samples/sec: 417.23 - lr: 0.000066 - momentum: 0.000000 2023-10-10 22:50:40,481 epoch 7 - iter 116/292 - loss 0.06362150 - time (sec): 42.01 - samples/sec: 439.08 - lr: 0.000064 - momentum: 0.000000 2023-10-10 22:50:50,693 epoch 7 - iter 145/292 - loss 0.06235246 - time (sec): 52.22 - samples/sec: 430.49 - lr: 0.000062 - momentum: 0.000000 2023-10-10 22:51:00,958 epoch 7 - iter 174/292 - loss 0.06840778 - time (sec): 62.49 - samples/sec: 423.77 - lr: 0.000061 - momentum: 0.000000 2023-10-10 22:51:10,514 epoch 7 - iter 203/292 - loss 0.06625399 - time (sec): 72.05 - samples/sec: 423.90 - lr: 0.000059 - momentum: 0.000000 2023-10-10 22:51:21,312 epoch 7 - iter 232/292 - loss 0.06499604 - time (sec): 82.84 - samples/sec: 428.42 - lr: 0.000057 - momentum: 0.000000 2023-10-10 22:51:30,775 epoch 7 - iter 261/292 - loss 0.06666186 - time (sec): 92.31 - samples/sec: 431.34 - lr: 0.000055 - momentum: 0.000000 2023-10-10 22:51:40,482 epoch 7 - iter 290/292 - loss 0.06904362 - time (sec): 102.01 - samples/sec: 433.76 - lr: 0.000054 - momentum: 0.000000 2023-10-10 22:51:40,947 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:51:40,947 EPOCH 7 done: loss 0.0696 - lr: 0.000054 2023-10-10 22:51:46,921 DEV : loss 0.13766349852085114 - f1-score (micro avg) 0.757 2023-10-10 22:51:46,929 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:51:56,914 epoch 8 - iter 29/292 - loss 0.04980003 - time (sec): 9.98 - samples/sec: 506.44 - lr: 0.000052 - momentum: 0.000000 2023-10-10 22:52:06,746 epoch 8 - iter 58/292 - loss 0.05123584 - time (sec): 19.82 - samples/sec: 500.27 - lr: 0.000050 - momentum: 0.000000 2023-10-10 22:52:16,597 epoch 8 - iter 87/292 - loss 0.05568077 - time (sec): 29.67 - samples/sec: 485.27 - lr: 0.000048 - momentum: 0.000000 2023-10-10 22:52:25,412 epoch 8 - iter 116/292 - loss 0.05299151 - time (sec): 38.48 - samples/sec: 468.70 - lr: 0.000046 - momentum: 0.000000 2023-10-10 22:52:35,578 epoch 8 - iter 145/292 - loss 0.05256239 - time (sec): 48.65 - samples/sec: 468.99 - lr: 0.000045 - momentum: 0.000000 2023-10-10 22:52:44,728 epoch 8 - iter 174/292 - loss 0.05533329 - time (sec): 57.80 - samples/sec: 467.26 - lr: 0.000043 - momentum: 0.000000 2023-10-10 22:52:53,698 epoch 8 - iter 203/292 - loss 0.05536529 - time (sec): 66.77 - samples/sec: 463.07 - lr: 0.000041 - momentum: 0.000000 2023-10-10 22:53:03,670 epoch 8 - iter 232/292 - loss 0.05502038 - time (sec): 76.74 - samples/sec: 465.90 - lr: 0.000039 - momentum: 0.000000 2023-10-10 22:53:13,047 epoch 8 - iter 261/292 - loss 0.05399818 - time (sec): 86.12 - samples/sec: 461.30 - lr: 0.000038 - momentum: 0.000000 2023-10-10 22:53:23,422 epoch 8 - iter 290/292 - loss 0.05648797 - time (sec): 96.49 - samples/sec: 458.97 - lr: 0.000036 - momentum: 0.000000 2023-10-10 22:53:23,902 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:53:23,903 EPOCH 8 done: loss 0.0564 - lr: 0.000036 2023-10-10 22:53:29,955 DEV : loss 0.12616726756095886 - f1-score (micro avg) 0.7716 2023-10-10 22:53:29,964 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:53:39,918 epoch 9 - iter 29/292 - loss 0.04779883 - time (sec): 9.95 - samples/sec: 468.37 - lr: 0.000034 - momentum: 0.000000 2023-10-10 22:53:49,660 epoch 9 - iter 58/292 - loss 0.04408728 - time (sec): 19.69 - samples/sec: 475.59 - lr: 0.000032 - momentum: 0.000000 2023-10-10 22:53:59,507 epoch 9 - iter 87/292 - loss 0.04629535 - time (sec): 29.54 - samples/sec: 467.97 - lr: 0.000030 - momentum: 0.000000 2023-10-10 22:54:10,142 epoch 9 - iter 116/292 - loss 0.04202133 - time (sec): 40.18 - samples/sec: 457.87 - lr: 0.000029 - momentum: 0.000000 2023-10-10 22:54:19,541 epoch 9 - iter 145/292 - loss 0.04383999 - time (sec): 49.57 - samples/sec: 451.70 - lr: 0.000027 - momentum: 0.000000 2023-10-10 22:54:30,079 epoch 9 - iter 174/292 - loss 0.04672478 - time (sec): 60.11 - samples/sec: 452.65 - lr: 0.000025 - momentum: 0.000000 2023-10-10 22:54:39,735 epoch 9 - iter 203/292 - loss 0.04641879 - time (sec): 69.77 - samples/sec: 444.73 - lr: 0.000023 - momentum: 0.000000 2023-10-10 22:54:50,153 epoch 9 - iter 232/292 - loss 0.04595521 - time (sec): 80.19 - samples/sec: 449.04 - lr: 0.000022 - momentum: 0.000000 2023-10-10 22:55:00,375 epoch 9 - iter 261/292 - loss 0.04464917 - time (sec): 90.41 - samples/sec: 445.85 - lr: 0.000020 - momentum: 0.000000 2023-10-10 22:55:09,855 epoch 9 - iter 290/292 - loss 0.04722935 - time (sec): 99.89 - samples/sec: 442.99 - lr: 0.000018 - momentum: 0.000000 2023-10-10 22:55:10,343 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:55:10,343 EPOCH 9 done: loss 0.0472 - lr: 0.000018 2023-10-10 22:55:16,251 DEV : loss 0.1275288611650467 - f1-score (micro avg) 0.7846 2023-10-10 22:55:16,268 saving best model 2023-10-10 22:55:21,311 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:55:31,028 epoch 10 - iter 29/292 - loss 0.03389118 - time (sec): 9.71 - samples/sec: 475.91 - lr: 0.000016 - momentum: 0.000000 2023-10-10 22:55:40,275 epoch 10 - iter 58/292 - loss 0.04001796 - time (sec): 18.96 - samples/sec: 451.56 - lr: 0.000014 - momentum: 0.000000 2023-10-10 22:55:49,745 epoch 10 - iter 87/292 - loss 0.03855770 - time (sec): 28.43 - samples/sec: 451.16 - lr: 0.000013 - momentum: 0.000000 2023-10-10 22:56:00,674 epoch 10 - iter 116/292 - loss 0.03345664 - time (sec): 39.36 - samples/sec: 461.08 - lr: 0.000011 - momentum: 0.000000 2023-10-10 22:56:10,848 epoch 10 - iter 145/292 - loss 0.03410413 - time (sec): 49.53 - samples/sec: 463.14 - lr: 0.000009 - momentum: 0.000000 2023-10-10 22:56:21,318 epoch 10 - iter 174/292 - loss 0.03396462 - time (sec): 60.00 - samples/sec: 455.00 - lr: 0.000007 - momentum: 0.000000 2023-10-10 22:56:32,199 epoch 10 - iter 203/292 - loss 0.03618776 - time (sec): 70.88 - samples/sec: 451.04 - lr: 0.000006 - momentum: 0.000000 2023-10-10 22:56:42,084 epoch 10 - iter 232/292 - loss 0.04153492 - time (sec): 80.77 - samples/sec: 445.45 - lr: 0.000004 - momentum: 0.000000 2023-10-10 22:56:51,564 epoch 10 - iter 261/292 - loss 0.04016965 - time (sec): 90.25 - samples/sec: 442.69 - lr: 0.000002 - momentum: 0.000000 2023-10-10 22:57:01,710 epoch 10 - iter 290/292 - loss 0.04323998 - time (sec): 100.40 - samples/sec: 440.51 - lr: 0.000000 - momentum: 0.000000 2023-10-10 22:57:02,248 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:57:02,249 EPOCH 10 done: loss 0.0434 - lr: 0.000000 2023-10-10 22:57:08,238 DEV : loss 0.1296752691268921 - f1-score (micro avg) 0.8017 2023-10-10 22:57:08,247 saving best model 2023-10-10 22:57:13,206 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:57:13,209 Loading model from best epoch ... 2023-10-10 22:57:17,160 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 2023-10-10 22:57:30,053 Results: - F-score (micro) 0.7252 - F-score (macro) 0.6708 - Accuracy 0.587 By class: precision recall f1-score support PER 0.7920 0.8534 0.8216 348 LOC 0.5819 0.7625 0.6600 261 ORG 0.4000 0.3846 0.3922 52 HumanProd 0.8500 0.7727 0.8095 22 micro avg 0.6773 0.7804 0.7252 683 macro avg 0.6560 0.6933 0.6708 683 weighted avg 0.6837 0.7804 0.7268 683 2023-10-10 22:57:30,054 ----------------------------------------------------------------------------------------------------