2023-10-10 23:17:02,388 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:17:02,390 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 23:17:02,390 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:17:02,390 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 23:17:02,390 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:17:02,390 Train: 1166 sentences 2023-10-10 23:17:02,391 (train_with_dev=False, train_with_test=False) 2023-10-10 23:17:02,391 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:17:02,391 Training Params: 2023-10-10 23:17:02,391 - learning_rate: "0.00016" 2023-10-10 23:17:02,391 - mini_batch_size: "8" 2023-10-10 23:17:02,391 - max_epochs: "10" 2023-10-10 23:17:02,391 - shuffle: "True" 2023-10-10 23:17:02,391 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:17:02,391 Plugins: 2023-10-10 23:17:02,391 - TensorboardLogger 2023-10-10 23:17:02,391 - LinearScheduler | warmup_fraction: '0.1' 2023-10-10 23:17:02,391 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:17:02,391 Final evaluation on model from best epoch (best-model.pt) 2023-10-10 23:17:02,391 - metric: "('micro avg', 'f1-score')" 2023-10-10 23:17:02,391 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:17:02,392 Computation: 2023-10-10 23:17:02,392 - compute on device: cuda:0 2023-10-10 23:17:02,392 - embedding storage: none 2023-10-10 23:17:02,392 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:17:02,392 Model training base path: "hmbench-newseye/fi-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-2" 2023-10-10 23:17:02,392 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:17:02,392 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:17:02,392 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-10 23:17:12,249 epoch 1 - iter 14/146 - loss 2.85068931 - time (sec): 9.85 - samples/sec: 507.67 - lr: 0.000014 - momentum: 0.000000 2023-10-10 23:17:21,019 epoch 1 - iter 28/146 - loss 2.84691647 - time (sec): 18.63 - samples/sec: 469.95 - lr: 0.000030 - momentum: 0.000000 2023-10-10 23:17:30,537 epoch 1 - iter 42/146 - loss 2.83473964 - time (sec): 28.14 - samples/sec: 472.52 - lr: 0.000045 - momentum: 0.000000 2023-10-10 23:17:39,867 epoch 1 - iter 56/146 - loss 2.81949073 - time (sec): 37.47 - samples/sec: 468.02 - lr: 0.000060 - momentum: 0.000000 2023-10-10 23:17:48,437 epoch 1 - iter 70/146 - loss 2.78568641 - time (sec): 46.04 - samples/sec: 463.24 - lr: 0.000076 - momentum: 0.000000 2023-10-10 23:17:56,794 epoch 1 - iter 84/146 - loss 2.72961033 - time (sec): 54.40 - samples/sec: 460.97 - lr: 0.000091 - momentum: 0.000000 2023-10-10 23:18:06,346 epoch 1 - iter 98/146 - loss 2.65786696 - time (sec): 63.95 - samples/sec: 455.80 - lr: 0.000106 - momentum: 0.000000 2023-10-10 23:18:16,236 epoch 1 - iter 112/146 - loss 2.57487198 - time (sec): 73.84 - samples/sec: 452.26 - lr: 0.000122 - momentum: 0.000000 2023-10-10 23:18:26,359 epoch 1 - iter 126/146 - loss 2.47816169 - time (sec): 83.97 - samples/sec: 452.64 - lr: 0.000137 - momentum: 0.000000 2023-10-10 23:18:36,630 epoch 1 - iter 140/146 - loss 2.38111585 - time (sec): 94.24 - samples/sec: 453.04 - lr: 0.000152 - momentum: 0.000000 2023-10-10 23:18:40,373 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:18:40,374 EPOCH 1 done: loss 2.3434 - lr: 0.000152 2023-10-10 23:18:46,381 DEV : loss 1.2865359783172607 - f1-score (micro avg) 0.0 2023-10-10 23:18:46,390 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:18:55,313 epoch 2 - iter 14/146 - loss 1.28415238 - time (sec): 8.92 - samples/sec: 472.05 - lr: 0.000158 - momentum: 0.000000 2023-10-10 23:19:05,700 epoch 2 - iter 28/146 - loss 1.19956105 - time (sec): 19.31 - samples/sec: 464.48 - lr: 0.000157 - momentum: 0.000000 2023-10-10 23:19:16,430 epoch 2 - iter 42/146 - loss 1.13288131 - time (sec): 30.04 - samples/sec: 458.33 - lr: 0.000155 - momentum: 0.000000 2023-10-10 23:19:26,012 epoch 2 - iter 56/146 - loss 1.07129114 - time (sec): 39.62 - samples/sec: 437.61 - lr: 0.000153 - momentum: 0.000000 2023-10-10 23:19:35,741 epoch 2 - iter 70/146 - loss 0.99550129 - time (sec): 49.35 - samples/sec: 438.17 - lr: 0.000152 - momentum: 0.000000 2023-10-10 23:19:44,108 epoch 2 - iter 84/146 - loss 0.96876458 - time (sec): 57.72 - samples/sec: 427.25 - lr: 0.000150 - momentum: 0.000000 2023-10-10 23:19:53,864 epoch 2 - iter 98/146 - loss 0.91377890 - time (sec): 67.47 - samples/sec: 432.59 - lr: 0.000148 - momentum: 0.000000 2023-10-10 23:20:04,816 epoch 2 - iter 112/146 - loss 0.85770767 - time (sec): 78.42 - samples/sec: 432.27 - lr: 0.000147 - momentum: 0.000000 2023-10-10 23:20:15,157 epoch 2 - iter 126/146 - loss 0.81157139 - time (sec): 88.76 - samples/sec: 430.30 - lr: 0.000145 - momentum: 0.000000 2023-10-10 23:20:25,002 epoch 2 - iter 140/146 - loss 0.78208038 - time (sec): 98.61 - samples/sec: 427.37 - lr: 0.000143 - momentum: 0.000000 2023-10-10 23:20:29,385 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:20:29,385 EPOCH 2 done: loss 0.8100 - lr: 0.000143 2023-10-10 23:20:36,275 DEV : loss 0.44310665130615234 - f1-score (micro avg) 0.0 2023-10-10 23:20:36,286 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:20:45,315 epoch 3 - iter 14/146 - loss 0.54953200 - time (sec): 9.03 - samples/sec: 403.67 - lr: 0.000141 - momentum: 0.000000 2023-10-10 23:20:55,048 epoch 3 - iter 28/146 - loss 0.47992717 - time (sec): 18.76 - samples/sec: 414.34 - lr: 0.000139 - momentum: 0.000000 2023-10-10 23:21:05,647 epoch 3 - iter 42/146 - loss 0.56595579 - time (sec): 29.36 - samples/sec: 427.02 - lr: 0.000137 - momentum: 0.000000 2023-10-10 23:21:15,050 epoch 3 - iter 56/146 - loss 0.54451042 - time (sec): 38.76 - samples/sec: 422.88 - lr: 0.000136 - momentum: 0.000000 2023-10-10 23:21:24,347 epoch 3 - iter 70/146 - loss 0.51838610 - time (sec): 48.06 - samples/sec: 429.97 - lr: 0.000134 - momentum: 0.000000 2023-10-10 23:21:33,558 epoch 3 - iter 84/146 - loss 0.50225077 - time (sec): 57.27 - samples/sec: 432.25 - lr: 0.000132 - momentum: 0.000000 2023-10-10 23:21:42,473 epoch 3 - iter 98/146 - loss 0.48025404 - time (sec): 66.19 - samples/sec: 441.08 - lr: 0.000131 - momentum: 0.000000 2023-10-10 23:21:52,350 epoch 3 - iter 112/146 - loss 0.46014641 - time (sec): 76.06 - samples/sec: 445.61 - lr: 0.000129 - momentum: 0.000000 2023-10-10 23:22:02,256 epoch 3 - iter 126/146 - loss 0.44598830 - time (sec): 85.97 - samples/sec: 445.28 - lr: 0.000127 - momentum: 0.000000 2023-10-10 23:22:12,230 epoch 3 - iter 140/146 - loss 0.43394233 - time (sec): 95.94 - samples/sec: 446.51 - lr: 0.000125 - momentum: 0.000000 2023-10-10 23:22:16,255 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:22:16,255 EPOCH 3 done: loss 0.4381 - lr: 0.000125 2023-10-10 23:22:22,564 DEV : loss 0.3388194143772125 - f1-score (micro avg) 0.233 2023-10-10 23:22:22,574 saving best model 2023-10-10 23:22:23,549 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:22:32,339 epoch 4 - iter 14/146 - loss 0.35134012 - time (sec): 8.79 - samples/sec: 464.07 - lr: 0.000123 - momentum: 0.000000 2023-10-10 23:22:41,623 epoch 4 - iter 28/146 - loss 0.42932021 - time (sec): 18.07 - samples/sec: 467.42 - lr: 0.000121 - momentum: 0.000000 2023-10-10 23:22:50,468 epoch 4 - iter 42/146 - loss 0.35560185 - time (sec): 26.92 - samples/sec: 473.02 - lr: 0.000120 - momentum: 0.000000 2023-10-10 23:22:58,955 epoch 4 - iter 56/146 - loss 0.35249099 - time (sec): 35.40 - samples/sec: 469.36 - lr: 0.000118 - momentum: 0.000000 2023-10-10 23:23:07,930 epoch 4 - iter 70/146 - loss 0.35507799 - time (sec): 44.38 - samples/sec: 464.26 - lr: 0.000116 - momentum: 0.000000 2023-10-10 23:23:16,896 epoch 4 - iter 84/146 - loss 0.35517374 - time (sec): 53.34 - samples/sec: 462.94 - lr: 0.000115 - momentum: 0.000000 2023-10-10 23:23:26,090 epoch 4 - iter 98/146 - loss 0.33861094 - time (sec): 62.54 - samples/sec: 465.58 - lr: 0.000113 - momentum: 0.000000 2023-10-10 23:23:35,066 epoch 4 - iter 112/146 - loss 0.33627771 - time (sec): 71.51 - samples/sec: 462.76 - lr: 0.000111 - momentum: 0.000000 2023-10-10 23:23:44,314 epoch 4 - iter 126/146 - loss 0.33904883 - time (sec): 80.76 - samples/sec: 465.33 - lr: 0.000109 - momentum: 0.000000 2023-10-10 23:23:53,980 epoch 4 - iter 140/146 - loss 0.34205571 - time (sec): 90.43 - samples/sec: 469.00 - lr: 0.000108 - momentum: 0.000000 2023-10-10 23:23:58,178 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:23:58,178 EPOCH 4 done: loss 0.3373 - lr: 0.000108 2023-10-10 23:24:04,577 DEV : loss 0.25445958971977234 - f1-score (micro avg) 0.2262 2023-10-10 23:24:04,587 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:24:13,816 epoch 5 - iter 14/146 - loss 0.32229392 - time (sec): 9.23 - samples/sec: 439.47 - lr: 0.000105 - momentum: 0.000000 2023-10-10 23:24:23,678 epoch 5 - iter 28/146 - loss 0.26837025 - time (sec): 19.09 - samples/sec: 464.72 - lr: 0.000104 - momentum: 0.000000 2023-10-10 23:24:32,600 epoch 5 - iter 42/146 - loss 0.26326562 - time (sec): 28.01 - samples/sec: 458.17 - lr: 0.000102 - momentum: 0.000000 2023-10-10 23:24:41,405 epoch 5 - iter 56/146 - loss 0.25728661 - time (sec): 36.82 - samples/sec: 461.08 - lr: 0.000100 - momentum: 0.000000 2023-10-10 23:24:50,289 epoch 5 - iter 70/146 - loss 0.26829976 - time (sec): 45.70 - samples/sec: 456.49 - lr: 0.000099 - momentum: 0.000000 2023-10-10 23:25:00,492 epoch 5 - iter 84/146 - loss 0.29246429 - time (sec): 55.90 - samples/sec: 467.11 - lr: 0.000097 - momentum: 0.000000 2023-10-10 23:25:10,684 epoch 5 - iter 98/146 - loss 0.29542982 - time (sec): 66.10 - samples/sec: 467.01 - lr: 0.000095 - momentum: 0.000000 2023-10-10 23:25:20,273 epoch 5 - iter 112/146 - loss 0.28924318 - time (sec): 75.68 - samples/sec: 467.48 - lr: 0.000093 - momentum: 0.000000 2023-10-10 23:25:28,956 epoch 5 - iter 126/146 - loss 0.28713756 - time (sec): 84.37 - samples/sec: 463.52 - lr: 0.000092 - momentum: 0.000000 2023-10-10 23:25:37,924 epoch 5 - iter 140/146 - loss 0.28490061 - time (sec): 93.34 - samples/sec: 457.23 - lr: 0.000090 - momentum: 0.000000 2023-10-10 23:25:41,869 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:25:41,869 EPOCH 5 done: loss 0.2832 - lr: 0.000090 2023-10-10 23:25:48,099 DEV : loss 0.22307763993740082 - f1-score (micro avg) 0.2994 2023-10-10 23:25:48,108 saving best model 2023-10-10 23:25:56,000 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:26:06,665 epoch 6 - iter 14/146 - loss 0.20640681 - time (sec): 10.66 - samples/sec: 436.38 - lr: 0.000088 - momentum: 0.000000 2023-10-10 23:26:15,474 epoch 6 - iter 28/146 - loss 0.23046142 - time (sec): 19.47 - samples/sec: 443.97 - lr: 0.000086 - momentum: 0.000000 2023-10-10 23:26:24,366 epoch 6 - iter 42/146 - loss 0.21524711 - time (sec): 28.36 - samples/sec: 452.86 - lr: 0.000084 - momentum: 0.000000 2023-10-10 23:26:33,110 epoch 6 - iter 56/146 - loss 0.22748057 - time (sec): 37.11 - samples/sec: 457.24 - lr: 0.000083 - momentum: 0.000000 2023-10-10 23:26:41,853 epoch 6 - iter 70/146 - loss 0.23191428 - time (sec): 45.85 - samples/sec: 462.58 - lr: 0.000081 - momentum: 0.000000 2023-10-10 23:26:50,581 epoch 6 - iter 84/146 - loss 0.23502256 - time (sec): 54.58 - samples/sec: 460.47 - lr: 0.000079 - momentum: 0.000000 2023-10-10 23:27:00,694 epoch 6 - iter 98/146 - loss 0.24896975 - time (sec): 64.69 - samples/sec: 467.42 - lr: 0.000077 - momentum: 0.000000 2023-10-10 23:27:09,453 epoch 6 - iter 112/146 - loss 0.24733184 - time (sec): 73.45 - samples/sec: 465.81 - lr: 0.000076 - momentum: 0.000000 2023-10-10 23:27:17,982 epoch 6 - iter 126/146 - loss 0.24239142 - time (sec): 81.98 - samples/sec: 466.69 - lr: 0.000074 - momentum: 0.000000 2023-10-10 23:27:26,637 epoch 6 - iter 140/146 - loss 0.23759475 - time (sec): 90.64 - samples/sec: 468.61 - lr: 0.000072 - momentum: 0.000000 2023-10-10 23:27:30,382 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:27:30,382 EPOCH 6 done: loss 0.2349 - lr: 0.000072 2023-10-10 23:27:36,129 DEV : loss 0.19796797633171082 - f1-score (micro avg) 0.4681 2023-10-10 23:27:36,139 saving best model 2023-10-10 23:27:43,828 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:27:52,247 epoch 7 - iter 14/146 - loss 0.19250517 - time (sec): 8.41 - samples/sec: 495.70 - lr: 0.000070 - momentum: 0.000000 2023-10-10 23:28:01,366 epoch 7 - iter 28/146 - loss 0.18333791 - time (sec): 17.53 - samples/sec: 525.61 - lr: 0.000068 - momentum: 0.000000 2023-10-10 23:28:09,139 epoch 7 - iter 42/146 - loss 0.18322523 - time (sec): 25.31 - samples/sec: 500.87 - lr: 0.000067 - momentum: 0.000000 2023-10-10 23:28:17,468 epoch 7 - iter 56/146 - loss 0.19443327 - time (sec): 33.64 - samples/sec: 499.47 - lr: 0.000065 - momentum: 0.000000 2023-10-10 23:28:24,988 epoch 7 - iter 70/146 - loss 0.18783057 - time (sec): 41.16 - samples/sec: 487.71 - lr: 0.000063 - momentum: 0.000000 2023-10-10 23:28:33,260 epoch 7 - iter 84/146 - loss 0.19033179 - time (sec): 49.43 - samples/sec: 490.52 - lr: 0.000061 - momentum: 0.000000 2023-10-10 23:28:43,265 epoch 7 - iter 98/146 - loss 0.19173980 - time (sec): 59.43 - samples/sec: 494.50 - lr: 0.000060 - momentum: 0.000000 2023-10-10 23:28:52,163 epoch 7 - iter 112/146 - loss 0.18999195 - time (sec): 68.33 - samples/sec: 491.39 - lr: 0.000058 - momentum: 0.000000 2023-10-10 23:29:01,255 epoch 7 - iter 126/146 - loss 0.19723376 - time (sec): 77.42 - samples/sec: 488.30 - lr: 0.000056 - momentum: 0.000000 2023-10-10 23:29:11,917 epoch 7 - iter 140/146 - loss 0.19244750 - time (sec): 88.08 - samples/sec: 485.24 - lr: 0.000055 - momentum: 0.000000 2023-10-10 23:29:16,231 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:29:16,232 EPOCH 7 done: loss 0.1924 - lr: 0.000055 2023-10-10 23:29:22,237 DEV : loss 0.17689262330532074 - f1-score (micro avg) 0.5087 2023-10-10 23:29:22,248 saving best model 2023-10-10 23:29:30,065 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:29:40,474 epoch 8 - iter 14/146 - loss 0.16935796 - time (sec): 10.40 - samples/sec: 406.27 - lr: 0.000052 - momentum: 0.000000 2023-10-10 23:29:50,068 epoch 8 - iter 28/146 - loss 0.18290286 - time (sec): 20.00 - samples/sec: 424.49 - lr: 0.000051 - momentum: 0.000000 2023-10-10 23:29:59,390 epoch 8 - iter 42/146 - loss 0.16793125 - time (sec): 29.32 - samples/sec: 434.61 - lr: 0.000049 - momentum: 0.000000 2023-10-10 23:30:07,662 epoch 8 - iter 56/146 - loss 0.17320389 - time (sec): 37.59 - samples/sec: 431.94 - lr: 0.000047 - momentum: 0.000000 2023-10-10 23:30:16,446 epoch 8 - iter 70/146 - loss 0.18227748 - time (sec): 46.38 - samples/sec: 452.40 - lr: 0.000045 - momentum: 0.000000 2023-10-10 23:30:24,610 epoch 8 - iter 84/146 - loss 0.18108910 - time (sec): 54.54 - samples/sec: 454.08 - lr: 0.000044 - momentum: 0.000000 2023-10-10 23:30:33,732 epoch 8 - iter 98/146 - loss 0.17092159 - time (sec): 63.66 - samples/sec: 464.89 - lr: 0.000042 - momentum: 0.000000 2023-10-10 23:30:41,860 epoch 8 - iter 112/146 - loss 0.16964323 - time (sec): 71.79 - samples/sec: 465.62 - lr: 0.000040 - momentum: 0.000000 2023-10-10 23:30:51,139 epoch 8 - iter 126/146 - loss 0.16593660 - time (sec): 81.07 - samples/sec: 471.56 - lr: 0.000039 - momentum: 0.000000 2023-10-10 23:31:00,500 epoch 8 - iter 140/146 - loss 0.16434625 - time (sec): 90.43 - samples/sec: 477.80 - lr: 0.000037 - momentum: 0.000000 2023-10-10 23:31:03,477 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:31:03,477 EPOCH 8 done: loss 0.1624 - lr: 0.000037 2023-10-10 23:31:09,101 DEV : loss 0.16704627871513367 - f1-score (micro avg) 0.5245 2023-10-10 23:31:09,111 saving best model 2023-10-10 23:31:16,953 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:31:25,648 epoch 9 - iter 14/146 - loss 0.15024636 - time (sec): 8.69 - samples/sec: 495.48 - lr: 0.000035 - momentum: 0.000000 2023-10-10 23:31:33,764 epoch 9 - iter 28/146 - loss 0.14496549 - time (sec): 16.81 - samples/sec: 490.96 - lr: 0.000033 - momentum: 0.000000 2023-10-10 23:31:41,355 epoch 9 - iter 42/146 - loss 0.16254726 - time (sec): 24.40 - samples/sec: 478.09 - lr: 0.000031 - momentum: 0.000000 2023-10-10 23:31:49,685 epoch 9 - iter 56/146 - loss 0.15667159 - time (sec): 32.73 - samples/sec: 485.05 - lr: 0.000029 - momentum: 0.000000 2023-10-10 23:31:59,645 epoch 9 - iter 70/146 - loss 0.16060893 - time (sec): 42.69 - samples/sec: 507.84 - lr: 0.000028 - momentum: 0.000000 2023-10-10 23:32:07,172 epoch 9 - iter 84/146 - loss 0.15235930 - time (sec): 50.21 - samples/sec: 496.36 - lr: 0.000026 - momentum: 0.000000 2023-10-10 23:32:16,069 epoch 9 - iter 98/146 - loss 0.15236847 - time (sec): 59.11 - samples/sec: 502.54 - lr: 0.000024 - momentum: 0.000000 2023-10-10 23:32:24,523 epoch 9 - iter 112/146 - loss 0.15005669 - time (sec): 67.57 - samples/sec: 502.42 - lr: 0.000023 - momentum: 0.000000 2023-10-10 23:32:32,980 epoch 9 - iter 126/146 - loss 0.14889199 - time (sec): 76.02 - samples/sec: 501.27 - lr: 0.000021 - momentum: 0.000000 2023-10-10 23:32:42,174 epoch 9 - iter 140/146 - loss 0.14689391 - time (sec): 85.22 - samples/sec: 504.07 - lr: 0.000019 - momentum: 0.000000 2023-10-10 23:32:45,520 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:32:45,520 EPOCH 9 done: loss 0.1449 - lr: 0.000019 2023-10-10 23:32:51,520 DEV : loss 0.16200371086597443 - f1-score (micro avg) 0.5683 2023-10-10 23:32:51,529 saving best model 2023-10-10 23:32:55,320 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:33:03,708 epoch 10 - iter 14/146 - loss 0.15671761 - time (sec): 8.38 - samples/sec: 484.05 - lr: 0.000017 - momentum: 0.000000 2023-10-10 23:33:12,911 epoch 10 - iter 28/146 - loss 0.14834250 - time (sec): 17.59 - samples/sec: 499.26 - lr: 0.000015 - momentum: 0.000000 2023-10-10 23:33:21,096 epoch 10 - iter 42/146 - loss 0.14012051 - time (sec): 25.77 - samples/sec: 483.92 - lr: 0.000013 - momentum: 0.000000 2023-10-10 23:33:29,718 epoch 10 - iter 56/146 - loss 0.13354190 - time (sec): 34.39 - samples/sec: 485.93 - lr: 0.000012 - momentum: 0.000000 2023-10-10 23:33:38,610 epoch 10 - iter 70/146 - loss 0.12759908 - time (sec): 43.29 - samples/sec: 486.33 - lr: 0.000010 - momentum: 0.000000 2023-10-10 23:33:46,661 epoch 10 - iter 84/146 - loss 0.12647792 - time (sec): 51.34 - samples/sec: 483.20 - lr: 0.000008 - momentum: 0.000000 2023-10-10 23:33:56,176 epoch 10 - iter 98/146 - loss 0.12934689 - time (sec): 60.85 - samples/sec: 487.90 - lr: 0.000007 - momentum: 0.000000 2023-10-10 23:34:05,223 epoch 10 - iter 112/146 - loss 0.13516861 - time (sec): 69.90 - samples/sec: 490.17 - lr: 0.000005 - momentum: 0.000000 2023-10-10 23:34:14,184 epoch 10 - iter 126/146 - loss 0.13301443 - time (sec): 78.86 - samples/sec: 486.70 - lr: 0.000003 - momentum: 0.000000 2023-10-10 23:34:23,644 epoch 10 - iter 140/146 - loss 0.13565286 - time (sec): 88.32 - samples/sec: 486.78 - lr: 0.000002 - momentum: 0.000000 2023-10-10 23:34:27,061 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:34:27,061 EPOCH 10 done: loss 0.1343 - lr: 0.000002 2023-10-10 23:34:33,022 DEV : loss 0.16090121865272522 - f1-score (micro avg) 0.5875 2023-10-10 23:34:33,032 saving best model 2023-10-10 23:34:41,145 ---------------------------------------------------------------------------------------------------- 2023-10-10 23:34:41,147 Loading model from best epoch ... 2023-10-10 23:34:44,867 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 23:34:58,663 Results: - F-score (micro) 0.6529 - F-score (macro) 0.4086 - Accuracy 0.5355 By class: precision recall f1-score support PER 0.7455 0.7069 0.7257 348 LOC 0.6192 0.7663 0.6849 261 ORG 0.1918 0.2692 0.2240 52 HumanProd 0.0000 0.0000 0.0000 22 micro avg 0.6336 0.6735 0.6529 683 macro avg 0.3891 0.4356 0.4086 683 weighted avg 0.6310 0.6735 0.6485 683 2023-10-10 23:34:58,663 ----------------------------------------------------------------------------------------------------