2023-10-11 10:50:03,874 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:50:03,877 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 10:50:03,877 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:50:03,877 MultiCorpus: 1085 train + 148 dev + 364 test sentences - NER_HIPE_2022 Corpus: 1085 train + 148 dev + 364 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/sv/with_doc_seperator 2023-10-11 10:50:03,877 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:50:03,877 Train: 1085 sentences 2023-10-11 10:50:03,877 (train_with_dev=False, train_with_test=False) 2023-10-11 10:50:03,877 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:50:03,877 Training Params: 2023-10-11 10:50:03,877 - learning_rate: "0.00015" 2023-10-11 10:50:03,878 - mini_batch_size: "4" 2023-10-11 10:50:03,878 - max_epochs: "10" 2023-10-11 10:50:03,878 - shuffle: "True" 2023-10-11 10:50:03,878 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:50:03,878 Plugins: 2023-10-11 10:50:03,878 - TensorboardLogger 2023-10-11 10:50:03,878 - LinearScheduler | warmup_fraction: '0.1' 2023-10-11 10:50:03,878 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:50:03,878 Final evaluation on model from best epoch (best-model.pt) 2023-10-11 10:50:03,878 - metric: "('micro avg', 'f1-score')" 2023-10-11 10:50:03,878 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:50:03,878 Computation: 2023-10-11 10:50:03,878 - compute on device: cuda:0 2023-10-11 10:50:03,878 - embedding storage: none 2023-10-11 10:50:03,878 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:50:03,878 Model training base path: "hmbench-newseye/sv-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-3" 2023-10-11 10:50:03,879 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:50:03,879 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:50:03,879 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-11 10:50:13,487 epoch 1 - iter 27/272 - loss 2.82659249 - time (sec): 9.61 - samples/sec: 598.94 - lr: 0.000014 - momentum: 0.000000 2023-10-11 10:50:22,796 epoch 1 - iter 54/272 - loss 2.81733222 - time (sec): 18.92 - samples/sec: 592.05 - lr: 0.000029 - momentum: 0.000000 2023-10-11 10:50:32,122 epoch 1 - iter 81/272 - loss 2.79616710 - time (sec): 28.24 - samples/sec: 580.96 - lr: 0.000044 - momentum: 0.000000 2023-10-11 10:50:41,422 epoch 1 - iter 108/272 - loss 2.75078140 - time (sec): 37.54 - samples/sec: 572.55 - lr: 0.000059 - momentum: 0.000000 2023-10-11 10:50:50,767 epoch 1 - iter 135/272 - loss 2.66645771 - time (sec): 46.89 - samples/sec: 571.66 - lr: 0.000074 - momentum: 0.000000 2023-10-11 10:50:59,522 epoch 1 - iter 162/272 - loss 2.58487111 - time (sec): 55.64 - samples/sec: 564.72 - lr: 0.000089 - momentum: 0.000000 2023-10-11 10:51:08,232 epoch 1 - iter 189/272 - loss 2.48987222 - time (sec): 64.35 - samples/sec: 557.49 - lr: 0.000104 - momentum: 0.000000 2023-10-11 10:51:17,966 epoch 1 - iter 216/272 - loss 2.36263029 - time (sec): 74.09 - samples/sec: 561.83 - lr: 0.000119 - momentum: 0.000000 2023-10-11 10:51:26,940 epoch 1 - iter 243/272 - loss 2.24603417 - time (sec): 83.06 - samples/sec: 560.39 - lr: 0.000133 - momentum: 0.000000 2023-10-11 10:51:36,410 epoch 1 - iter 270/272 - loss 2.12867006 - time (sec): 92.53 - samples/sec: 558.34 - lr: 0.000148 - momentum: 0.000000 2023-10-11 10:51:36,944 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:51:36,945 EPOCH 1 done: loss 2.1205 - lr: 0.000148 2023-10-11 10:51:41,788 DEV : loss 0.8181562423706055 - f1-score (micro avg) 0.0 2023-10-11 10:51:41,796 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:51:51,485 epoch 2 - iter 27/272 - loss 0.77588845 - time (sec): 9.69 - samples/sec: 602.16 - lr: 0.000148 - momentum: 0.000000 2023-10-11 10:52:00,658 epoch 2 - iter 54/272 - loss 0.73677202 - time (sec): 18.86 - samples/sec: 597.29 - lr: 0.000147 - momentum: 0.000000 2023-10-11 10:52:09,718 epoch 2 - iter 81/272 - loss 0.70960269 - time (sec): 27.92 - samples/sec: 587.29 - lr: 0.000145 - momentum: 0.000000 2023-10-11 10:52:18,476 epoch 2 - iter 108/272 - loss 0.67231305 - time (sec): 36.68 - samples/sec: 574.11 - lr: 0.000143 - momentum: 0.000000 2023-10-11 10:52:27,575 epoch 2 - iter 135/272 - loss 0.63398133 - time (sec): 45.78 - samples/sec: 571.45 - lr: 0.000142 - momentum: 0.000000 2023-10-11 10:52:36,162 epoch 2 - iter 162/272 - loss 0.60685794 - time (sec): 54.36 - samples/sec: 561.27 - lr: 0.000140 - momentum: 0.000000 2023-10-11 10:52:45,245 epoch 2 - iter 189/272 - loss 0.57321778 - time (sec): 63.45 - samples/sec: 556.34 - lr: 0.000138 - momentum: 0.000000 2023-10-11 10:52:54,524 epoch 2 - iter 216/272 - loss 0.55390885 - time (sec): 72.73 - samples/sec: 554.04 - lr: 0.000137 - momentum: 0.000000 2023-10-11 10:53:04,155 epoch 2 - iter 243/272 - loss 0.53466124 - time (sec): 82.36 - samples/sec: 553.70 - lr: 0.000135 - momentum: 0.000000 2023-10-11 10:53:14,344 epoch 2 - iter 270/272 - loss 0.51513046 - time (sec): 92.55 - samples/sec: 558.92 - lr: 0.000134 - momentum: 0.000000 2023-10-11 10:53:14,819 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:53:14,819 EPOCH 2 done: loss 0.5129 - lr: 0.000134 2023-10-11 10:53:20,319 DEV : loss 0.30094385147094727 - f1-score (micro avg) 0.318 2023-10-11 10:53:20,328 saving best model 2023-10-11 10:53:21,176 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:53:30,341 epoch 3 - iter 27/272 - loss 0.36013238 - time (sec): 9.16 - samples/sec: 557.38 - lr: 0.000132 - momentum: 0.000000 2023-10-11 10:53:39,870 epoch 3 - iter 54/272 - loss 0.33093762 - time (sec): 18.69 - samples/sec: 566.19 - lr: 0.000130 - momentum: 0.000000 2023-10-11 10:53:49,723 epoch 3 - iter 81/272 - loss 0.32749068 - time (sec): 28.54 - samples/sec: 583.92 - lr: 0.000128 - momentum: 0.000000 2023-10-11 10:53:58,383 epoch 3 - iter 108/272 - loss 0.33782747 - time (sec): 37.20 - samples/sec: 565.12 - lr: 0.000127 - momentum: 0.000000 2023-10-11 10:54:07,944 epoch 3 - iter 135/272 - loss 0.33686699 - time (sec): 46.77 - samples/sec: 551.43 - lr: 0.000125 - momentum: 0.000000 2023-10-11 10:54:18,169 epoch 3 - iter 162/272 - loss 0.32243636 - time (sec): 56.99 - samples/sec: 557.39 - lr: 0.000123 - momentum: 0.000000 2023-10-11 10:54:27,286 epoch 3 - iter 189/272 - loss 0.31031933 - time (sec): 66.11 - samples/sec: 558.27 - lr: 0.000122 - momentum: 0.000000 2023-10-11 10:54:36,651 epoch 3 - iter 216/272 - loss 0.29566962 - time (sec): 75.47 - samples/sec: 555.53 - lr: 0.000120 - momentum: 0.000000 2023-10-11 10:54:45,706 epoch 3 - iter 243/272 - loss 0.28469825 - time (sec): 84.53 - samples/sec: 552.48 - lr: 0.000119 - momentum: 0.000000 2023-10-11 10:54:54,928 epoch 3 - iter 270/272 - loss 0.28376485 - time (sec): 93.75 - samples/sec: 552.23 - lr: 0.000117 - momentum: 0.000000 2023-10-11 10:54:55,379 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:54:55,380 EPOCH 3 done: loss 0.2831 - lr: 0.000117 2023-10-11 10:55:00,783 DEV : loss 0.20688828825950623 - f1-score (micro avg) 0.5714 2023-10-11 10:55:00,792 saving best model 2023-10-11 10:55:03,337 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:55:12,297 epoch 4 - iter 27/272 - loss 0.18883592 - time (sec): 8.96 - samples/sec: 538.43 - lr: 0.000115 - momentum: 0.000000 2023-10-11 10:55:20,299 epoch 4 - iter 54/272 - loss 0.21454971 - time (sec): 16.96 - samples/sec: 510.04 - lr: 0.000113 - momentum: 0.000000 2023-10-11 10:55:29,545 epoch 4 - iter 81/272 - loss 0.22342544 - time (sec): 26.20 - samples/sec: 537.13 - lr: 0.000112 - momentum: 0.000000 2023-10-11 10:55:38,397 epoch 4 - iter 108/272 - loss 0.20820804 - time (sec): 35.06 - samples/sec: 541.85 - lr: 0.000110 - momentum: 0.000000 2023-10-11 10:55:47,376 epoch 4 - iter 135/272 - loss 0.20439682 - time (sec): 44.03 - samples/sec: 543.39 - lr: 0.000108 - momentum: 0.000000 2023-10-11 10:55:55,921 epoch 4 - iter 162/272 - loss 0.19740033 - time (sec): 52.58 - samples/sec: 537.66 - lr: 0.000107 - momentum: 0.000000 2023-10-11 10:56:06,557 epoch 4 - iter 189/272 - loss 0.18830402 - time (sec): 63.22 - samples/sec: 555.38 - lr: 0.000105 - momentum: 0.000000 2023-10-11 10:56:15,674 epoch 4 - iter 216/272 - loss 0.18412200 - time (sec): 72.33 - samples/sec: 557.92 - lr: 0.000103 - momentum: 0.000000 2023-10-11 10:56:25,318 epoch 4 - iter 243/272 - loss 0.17948316 - time (sec): 81.98 - samples/sec: 559.34 - lr: 0.000102 - momentum: 0.000000 2023-10-11 10:56:35,122 epoch 4 - iter 270/272 - loss 0.17495580 - time (sec): 91.78 - samples/sec: 562.76 - lr: 0.000100 - momentum: 0.000000 2023-10-11 10:56:35,663 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:56:35,663 EPOCH 4 done: loss 0.1745 - lr: 0.000100 2023-10-11 10:56:41,543 DEV : loss 0.16108796000480652 - f1-score (micro avg) 0.636 2023-10-11 10:56:41,552 saving best model 2023-10-11 10:56:44,080 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:56:54,081 epoch 5 - iter 27/272 - loss 0.11279072 - time (sec): 10.00 - samples/sec: 575.65 - lr: 0.000098 - momentum: 0.000000 2023-10-11 10:57:04,141 epoch 5 - iter 54/272 - loss 0.12313934 - time (sec): 20.06 - samples/sec: 554.27 - lr: 0.000097 - momentum: 0.000000 2023-10-11 10:57:13,844 epoch 5 - iter 81/272 - loss 0.11655880 - time (sec): 29.76 - samples/sec: 552.92 - lr: 0.000095 - momentum: 0.000000 2023-10-11 10:57:22,628 epoch 5 - iter 108/272 - loss 0.11981029 - time (sec): 38.54 - samples/sec: 540.70 - lr: 0.000093 - momentum: 0.000000 2023-10-11 10:57:31,316 epoch 5 - iter 135/272 - loss 0.12033932 - time (sec): 47.23 - samples/sec: 534.06 - lr: 0.000092 - momentum: 0.000000 2023-10-11 10:57:40,567 epoch 5 - iter 162/272 - loss 0.12185608 - time (sec): 56.48 - samples/sec: 536.12 - lr: 0.000090 - momentum: 0.000000 2023-10-11 10:57:49,789 epoch 5 - iter 189/272 - loss 0.11473005 - time (sec): 65.71 - samples/sec: 536.88 - lr: 0.000088 - momentum: 0.000000 2023-10-11 10:57:59,333 epoch 5 - iter 216/272 - loss 0.11661259 - time (sec): 75.25 - samples/sec: 538.49 - lr: 0.000087 - momentum: 0.000000 2023-10-11 10:58:09,636 epoch 5 - iter 243/272 - loss 0.11767849 - time (sec): 85.55 - samples/sec: 547.14 - lr: 0.000085 - momentum: 0.000000 2023-10-11 10:58:18,905 epoch 5 - iter 270/272 - loss 0.11723516 - time (sec): 94.82 - samples/sec: 546.86 - lr: 0.000084 - momentum: 0.000000 2023-10-11 10:58:19,303 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:58:19,303 EPOCH 5 done: loss 0.1171 - lr: 0.000084 2023-10-11 10:58:24,880 DEV : loss 0.13878858089447021 - f1-score (micro avg) 0.7559 2023-10-11 10:58:24,889 saving best model 2023-10-11 10:58:27,426 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:58:36,265 epoch 6 - iter 27/272 - loss 0.07309716 - time (sec): 8.83 - samples/sec: 529.07 - lr: 0.000082 - momentum: 0.000000 2023-10-11 10:58:45,128 epoch 6 - iter 54/272 - loss 0.08693004 - time (sec): 17.70 - samples/sec: 521.43 - lr: 0.000080 - momentum: 0.000000 2023-10-11 10:58:54,405 epoch 6 - iter 81/272 - loss 0.09456387 - time (sec): 26.97 - samples/sec: 523.42 - lr: 0.000078 - momentum: 0.000000 2023-10-11 10:59:04,184 epoch 6 - iter 108/272 - loss 0.08984032 - time (sec): 36.75 - samples/sec: 537.09 - lr: 0.000077 - momentum: 0.000000 2023-10-11 10:59:14,018 epoch 6 - iter 135/272 - loss 0.08508482 - time (sec): 46.59 - samples/sec: 552.64 - lr: 0.000075 - momentum: 0.000000 2023-10-11 10:59:23,202 epoch 6 - iter 162/272 - loss 0.08590144 - time (sec): 55.77 - samples/sec: 541.76 - lr: 0.000073 - momentum: 0.000000 2023-10-11 10:59:32,673 epoch 6 - iter 189/272 - loss 0.08188597 - time (sec): 65.24 - samples/sec: 546.70 - lr: 0.000072 - momentum: 0.000000 2023-10-11 10:59:41,961 epoch 6 - iter 216/272 - loss 0.08590837 - time (sec): 74.53 - samples/sec: 544.08 - lr: 0.000070 - momentum: 0.000000 2023-10-11 10:59:51,615 epoch 6 - iter 243/272 - loss 0.08500069 - time (sec): 84.18 - samples/sec: 546.89 - lr: 0.000069 - momentum: 0.000000 2023-10-11 11:00:01,257 epoch 6 - iter 270/272 - loss 0.08427618 - time (sec): 93.83 - samples/sec: 548.96 - lr: 0.000067 - momentum: 0.000000 2023-10-11 11:00:01,944 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:00:01,944 EPOCH 6 done: loss 0.0842 - lr: 0.000067 2023-10-11 11:00:07,586 DEV : loss 0.13508526980876923 - f1-score (micro avg) 0.7784 2023-10-11 11:00:07,594 saving best model 2023-10-11 11:00:10,077 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:00:19,490 epoch 7 - iter 27/272 - loss 0.06349629 - time (sec): 9.41 - samples/sec: 572.82 - lr: 0.000065 - momentum: 0.000000 2023-10-11 11:00:28,354 epoch 7 - iter 54/272 - loss 0.07168990 - time (sec): 18.27 - samples/sec: 558.79 - lr: 0.000063 - momentum: 0.000000 2023-10-11 11:00:38,273 epoch 7 - iter 81/272 - loss 0.06744160 - time (sec): 28.19 - samples/sec: 569.92 - lr: 0.000062 - momentum: 0.000000 2023-10-11 11:00:47,594 epoch 7 - iter 108/272 - loss 0.06533570 - time (sec): 37.51 - samples/sec: 569.37 - lr: 0.000060 - momentum: 0.000000 2023-10-11 11:00:56,990 epoch 7 - iter 135/272 - loss 0.07004227 - time (sec): 46.91 - samples/sec: 569.69 - lr: 0.000058 - momentum: 0.000000 2023-10-11 11:01:06,041 epoch 7 - iter 162/272 - loss 0.06680624 - time (sec): 55.96 - samples/sec: 564.74 - lr: 0.000057 - momentum: 0.000000 2023-10-11 11:01:15,636 epoch 7 - iter 189/272 - loss 0.07011620 - time (sec): 65.56 - samples/sec: 563.17 - lr: 0.000055 - momentum: 0.000000 2023-10-11 11:01:24,444 epoch 7 - iter 216/272 - loss 0.06882717 - time (sec): 74.36 - samples/sec: 557.55 - lr: 0.000053 - momentum: 0.000000 2023-10-11 11:01:34,079 epoch 7 - iter 243/272 - loss 0.06683075 - time (sec): 84.00 - samples/sec: 557.83 - lr: 0.000052 - momentum: 0.000000 2023-10-11 11:01:43,455 epoch 7 - iter 270/272 - loss 0.06398691 - time (sec): 93.37 - samples/sec: 554.75 - lr: 0.000050 - momentum: 0.000000 2023-10-11 11:01:43,855 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:01:43,855 EPOCH 7 done: loss 0.0638 - lr: 0.000050 2023-10-11 11:01:49,480 DEV : loss 0.13904628157615662 - f1-score (micro avg) 0.7544 2023-10-11 11:01:49,488 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:01:58,810 epoch 8 - iter 27/272 - loss 0.05151019 - time (sec): 9.32 - samples/sec: 555.05 - lr: 0.000048 - momentum: 0.000000 2023-10-11 11:02:07,667 epoch 8 - iter 54/272 - loss 0.04875397 - time (sec): 18.18 - samples/sec: 538.60 - lr: 0.000047 - momentum: 0.000000 2023-10-11 11:02:17,611 epoch 8 - iter 81/272 - loss 0.05284993 - time (sec): 28.12 - samples/sec: 547.95 - lr: 0.000045 - momentum: 0.000000 2023-10-11 11:02:26,809 epoch 8 - iter 108/272 - loss 0.05991550 - time (sec): 37.32 - samples/sec: 544.64 - lr: 0.000043 - momentum: 0.000000 2023-10-11 11:02:36,284 epoch 8 - iter 135/272 - loss 0.05652978 - time (sec): 46.79 - samples/sec: 546.34 - lr: 0.000042 - momentum: 0.000000 2023-10-11 11:02:45,701 epoch 8 - iter 162/272 - loss 0.05626381 - time (sec): 56.21 - samples/sec: 547.48 - lr: 0.000040 - momentum: 0.000000 2023-10-11 11:02:55,335 epoch 8 - iter 189/272 - loss 0.05419822 - time (sec): 65.85 - samples/sec: 549.05 - lr: 0.000038 - momentum: 0.000000 2023-10-11 11:03:05,159 epoch 8 - iter 216/272 - loss 0.05356223 - time (sec): 75.67 - samples/sec: 549.84 - lr: 0.000037 - momentum: 0.000000 2023-10-11 11:03:14,930 epoch 8 - iter 243/272 - loss 0.05043893 - time (sec): 85.44 - samples/sec: 551.54 - lr: 0.000035 - momentum: 0.000000 2023-10-11 11:03:23,724 epoch 8 - iter 270/272 - loss 0.05041046 - time (sec): 94.23 - samples/sec: 548.22 - lr: 0.000034 - momentum: 0.000000 2023-10-11 11:03:24,272 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:03:24,272 EPOCH 8 done: loss 0.0502 - lr: 0.000034 2023-10-11 11:03:30,009 DEV : loss 0.13503918051719666 - f1-score (micro avg) 0.7964 2023-10-11 11:03:30,018 saving best model 2023-10-11 11:03:32,544 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:03:42,316 epoch 9 - iter 27/272 - loss 0.04416837 - time (sec): 9.77 - samples/sec: 597.14 - lr: 0.000032 - momentum: 0.000000 2023-10-11 11:03:51,802 epoch 9 - iter 54/272 - loss 0.04781217 - time (sec): 19.25 - samples/sec: 577.94 - lr: 0.000030 - momentum: 0.000000 2023-10-11 11:04:00,376 epoch 9 - iter 81/272 - loss 0.04608928 - time (sec): 27.83 - samples/sec: 558.18 - lr: 0.000028 - momentum: 0.000000 2023-10-11 11:04:09,553 epoch 9 - iter 108/272 - loss 0.04616787 - time (sec): 37.01 - samples/sec: 556.51 - lr: 0.000027 - momentum: 0.000000 2023-10-11 11:04:19,183 epoch 9 - iter 135/272 - loss 0.04546524 - time (sec): 46.63 - samples/sec: 558.30 - lr: 0.000025 - momentum: 0.000000 2023-10-11 11:04:28,689 epoch 9 - iter 162/272 - loss 0.04381471 - time (sec): 56.14 - samples/sec: 559.12 - lr: 0.000023 - momentum: 0.000000 2023-10-11 11:04:38,019 epoch 9 - iter 189/272 - loss 0.04496020 - time (sec): 65.47 - samples/sec: 554.38 - lr: 0.000022 - momentum: 0.000000 2023-10-11 11:04:47,008 epoch 9 - iter 216/272 - loss 0.04477837 - time (sec): 74.46 - samples/sec: 551.81 - lr: 0.000020 - momentum: 0.000000 2023-10-11 11:04:57,105 epoch 9 - iter 243/272 - loss 0.04362140 - time (sec): 84.56 - samples/sec: 556.23 - lr: 0.000019 - momentum: 0.000000 2023-10-11 11:05:06,298 epoch 9 - iter 270/272 - loss 0.04226270 - time (sec): 93.75 - samples/sec: 553.27 - lr: 0.000017 - momentum: 0.000000 2023-10-11 11:05:06,661 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:05:06,662 EPOCH 9 done: loss 0.0423 - lr: 0.000017 2023-10-11 11:05:12,189 DEV : loss 0.13971544802188873 - f1-score (micro avg) 0.7899 2023-10-11 11:05:12,198 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:05:22,115 epoch 10 - iter 27/272 - loss 0.04362523 - time (sec): 9.92 - samples/sec: 549.26 - lr: 0.000015 - momentum: 0.000000 2023-10-11 11:05:30,817 epoch 10 - iter 54/272 - loss 0.04805838 - time (sec): 18.62 - samples/sec: 523.99 - lr: 0.000013 - momentum: 0.000000 2023-10-11 11:05:41,534 epoch 10 - iter 81/272 - loss 0.04944586 - time (sec): 29.33 - samples/sec: 557.68 - lr: 0.000012 - momentum: 0.000000 2023-10-11 11:05:51,898 epoch 10 - iter 108/272 - loss 0.04975532 - time (sec): 39.70 - samples/sec: 565.25 - lr: 0.000010 - momentum: 0.000000 2023-10-11 11:06:01,700 epoch 10 - iter 135/272 - loss 0.04678516 - time (sec): 49.50 - samples/sec: 561.01 - lr: 0.000008 - momentum: 0.000000 2023-10-11 11:06:10,214 epoch 10 - iter 162/272 - loss 0.04446697 - time (sec): 58.01 - samples/sec: 549.99 - lr: 0.000007 - momentum: 0.000000 2023-10-11 11:06:20,152 epoch 10 - iter 189/272 - loss 0.04288849 - time (sec): 67.95 - samples/sec: 550.37 - lr: 0.000005 - momentum: 0.000000 2023-10-11 11:06:29,363 epoch 10 - iter 216/272 - loss 0.04070236 - time (sec): 77.16 - samples/sec: 543.35 - lr: 0.000003 - momentum: 0.000000 2023-10-11 11:06:38,730 epoch 10 - iter 243/272 - loss 0.03991694 - time (sec): 86.53 - samples/sec: 543.33 - lr: 0.000002 - momentum: 0.000000 2023-10-11 11:06:47,930 epoch 10 - iter 270/272 - loss 0.03815397 - time (sec): 95.73 - samples/sec: 540.55 - lr: 0.000000 - momentum: 0.000000 2023-10-11 11:06:48,386 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:06:48,387 EPOCH 10 done: loss 0.0382 - lr: 0.000000 2023-10-11 11:06:54,102 DEV : loss 0.1388687789440155 - f1-score (micro avg) 0.7985 2023-10-11 11:06:54,111 saving best model 2023-10-11 11:06:57,461 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:06:57,463 Loading model from best epoch ... 2023-10-11 11:07:01,068 SequenceTagger predicts: Dictionary with 17 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd, S-ORG, B-ORG, E-ORG, I-ORG 2023-10-11 11:07:13,976 Results: - F-score (micro) 0.7448 - F-score (macro) 0.6813 - Accuracy 0.6137 By class: precision recall f1-score support LOC 0.7652 0.8462 0.8037 312 PER 0.6357 0.8558 0.7295 208 ORG 0.4706 0.4364 0.4528 55 HumanProd 0.7083 0.7727 0.7391 22 micro avg 0.6900 0.8090 0.7448 597 macro avg 0.6450 0.7278 0.6813 597 weighted avg 0.6909 0.8090 0.7431 597 2023-10-11 11:07:13,976 ----------------------------------------------------------------------------------------------------