2023-10-11 12:17:47,458 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:17:47,460 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 12:17:47,460 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:17:47,461 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 12:17:47,461 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:17:47,461 Train: 1085 sentences 2023-10-11 12:17:47,461 (train_with_dev=False, train_with_test=False) 2023-10-11 12:17:47,461 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:17:47,461 Training Params: 2023-10-11 12:17:47,461 - learning_rate: "0.00016" 2023-10-11 12:17:47,461 - mini_batch_size: "4" 2023-10-11 12:17:47,461 - max_epochs: "10" 2023-10-11 12:17:47,461 - shuffle: "True" 2023-10-11 12:17:47,461 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:17:47,461 Plugins: 2023-10-11 12:17:47,461 - TensorboardLogger 2023-10-11 12:17:47,462 - LinearScheduler | warmup_fraction: '0.1' 2023-10-11 12:17:47,462 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:17:47,462 Final evaluation on model from best epoch (best-model.pt) 2023-10-11 12:17:47,462 - metric: "('micro avg', 'f1-score')" 2023-10-11 12:17:47,462 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:17:47,462 Computation: 2023-10-11 12:17:47,462 - compute on device: cuda:0 2023-10-11 12:17:47,462 - embedding storage: none 2023-10-11 12:17:47,462 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:17:47,462 Model training base path: "hmbench-newseye/sv-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-4" 2023-10-11 12:17:47,462 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:17:47,462 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:17:47,462 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-11 12:17:57,185 epoch 1 - iter 27/272 - loss 2.83784856 - time (sec): 9.72 - samples/sec: 573.53 - lr: 0.000015 - momentum: 0.000000 2023-10-11 12:18:07,237 epoch 1 - iter 54/272 - loss 2.83077250 - time (sec): 19.77 - samples/sec: 575.99 - lr: 0.000031 - momentum: 0.000000 2023-10-11 12:18:16,704 epoch 1 - iter 81/272 - loss 2.80770322 - time (sec): 29.24 - samples/sec: 570.04 - lr: 0.000047 - momentum: 0.000000 2023-10-11 12:18:25,982 epoch 1 - iter 108/272 - loss 2.75243362 - time (sec): 38.52 - samples/sec: 565.04 - lr: 0.000063 - momentum: 0.000000 2023-10-11 12:18:34,945 epoch 1 - iter 135/272 - loss 2.67455773 - time (sec): 47.48 - samples/sec: 554.80 - lr: 0.000079 - momentum: 0.000000 2023-10-11 12:18:43,921 epoch 1 - iter 162/272 - loss 2.57823913 - time (sec): 56.46 - samples/sec: 550.87 - lr: 0.000095 - momentum: 0.000000 2023-10-11 12:18:52,759 epoch 1 - iter 189/272 - loss 2.47103511 - time (sec): 65.29 - samples/sec: 548.01 - lr: 0.000111 - momentum: 0.000000 2023-10-11 12:19:02,746 epoch 1 - iter 216/272 - loss 2.34565635 - time (sec): 75.28 - samples/sec: 551.99 - lr: 0.000126 - momentum: 0.000000 2023-10-11 12:19:11,733 epoch 1 - iter 243/272 - loss 2.22212358 - time (sec): 84.27 - samples/sec: 552.05 - lr: 0.000142 - momentum: 0.000000 2023-10-11 12:19:20,836 epoch 1 - iter 270/272 - loss 2.09514760 - time (sec): 93.37 - samples/sec: 553.28 - lr: 0.000158 - momentum: 0.000000 2023-10-11 12:19:21,365 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:19:21,365 EPOCH 1 done: loss 2.0875 - lr: 0.000158 2023-10-11 12:19:26,097 DEV : loss 0.7258709073066711 - f1-score (micro avg) 0.0 2023-10-11 12:19:26,106 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:19:35,428 epoch 2 - iter 27/272 - loss 0.70983607 - time (sec): 9.32 - samples/sec: 573.84 - lr: 0.000158 - momentum: 0.000000 2023-10-11 12:19:44,149 epoch 2 - iter 54/272 - loss 0.68848670 - time (sec): 18.04 - samples/sec: 560.06 - lr: 0.000157 - momentum: 0.000000 2023-10-11 12:19:52,270 epoch 2 - iter 81/272 - loss 0.66075679 - time (sec): 26.16 - samples/sec: 534.87 - lr: 0.000155 - momentum: 0.000000 2023-10-11 12:20:02,359 epoch 2 - iter 108/272 - loss 0.60795335 - time (sec): 36.25 - samples/sec: 557.87 - lr: 0.000153 - momentum: 0.000000 2023-10-11 12:20:11,747 epoch 2 - iter 135/272 - loss 0.58284167 - time (sec): 45.64 - samples/sec: 552.89 - lr: 0.000151 - momentum: 0.000000 2023-10-11 12:20:21,743 epoch 2 - iter 162/272 - loss 0.53477008 - time (sec): 55.63 - samples/sec: 558.84 - lr: 0.000149 - momentum: 0.000000 2023-10-11 12:20:30,560 epoch 2 - iter 189/272 - loss 0.51131648 - time (sec): 64.45 - samples/sec: 549.06 - lr: 0.000148 - momentum: 0.000000 2023-10-11 12:20:39,933 epoch 2 - iter 216/272 - loss 0.48449409 - time (sec): 73.83 - samples/sec: 547.01 - lr: 0.000146 - momentum: 0.000000 2023-10-11 12:20:49,031 epoch 2 - iter 243/272 - loss 0.47029300 - time (sec): 82.92 - samples/sec: 544.63 - lr: 0.000144 - momentum: 0.000000 2023-10-11 12:20:59,571 epoch 2 - iter 270/272 - loss 0.45240623 - time (sec): 93.46 - samples/sec: 554.23 - lr: 0.000142 - momentum: 0.000000 2023-10-11 12:20:59,996 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:20:59,996 EPOCH 2 done: loss 0.4521 - lr: 0.000142 2023-10-11 12:21:05,525 DEV : loss 0.27614670991897583 - f1-score (micro avg) 0.3235 2023-10-11 12:21:05,534 saving best model 2023-10-11 12:21:06,411 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:21:15,715 epoch 3 - iter 27/272 - loss 0.23309089 - time (sec): 9.30 - samples/sec: 548.35 - lr: 0.000141 - momentum: 0.000000 2023-10-11 12:21:25,472 epoch 3 - iter 54/272 - loss 0.27485890 - time (sec): 19.06 - samples/sec: 567.71 - lr: 0.000139 - momentum: 0.000000 2023-10-11 12:21:34,737 epoch 3 - iter 81/272 - loss 0.27012296 - time (sec): 28.32 - samples/sec: 569.76 - lr: 0.000137 - momentum: 0.000000 2023-10-11 12:21:43,893 epoch 3 - iter 108/272 - loss 0.26806557 - time (sec): 37.48 - samples/sec: 560.35 - lr: 0.000135 - momentum: 0.000000 2023-10-11 12:21:53,166 epoch 3 - iter 135/272 - loss 0.26644863 - time (sec): 46.75 - samples/sec: 560.30 - lr: 0.000133 - momentum: 0.000000 2023-10-11 12:22:02,417 epoch 3 - iter 162/272 - loss 0.27294415 - time (sec): 56.00 - samples/sec: 559.74 - lr: 0.000132 - momentum: 0.000000 2023-10-11 12:22:12,036 epoch 3 - iter 189/272 - loss 0.26862119 - time (sec): 65.62 - samples/sec: 563.37 - lr: 0.000130 - momentum: 0.000000 2023-10-11 12:22:21,344 epoch 3 - iter 216/272 - loss 0.26282089 - time (sec): 74.93 - samples/sec: 559.23 - lr: 0.000128 - momentum: 0.000000 2023-10-11 12:22:30,546 epoch 3 - iter 243/272 - loss 0.26662864 - time (sec): 84.13 - samples/sec: 556.68 - lr: 0.000126 - momentum: 0.000000 2023-10-11 12:22:39,525 epoch 3 - iter 270/272 - loss 0.26058992 - time (sec): 93.11 - samples/sec: 554.85 - lr: 0.000125 - momentum: 0.000000 2023-10-11 12:22:40,069 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:22:40,069 EPOCH 3 done: loss 0.2592 - lr: 0.000125 2023-10-11 12:22:45,596 DEV : loss 0.19185124337673187 - f1-score (micro avg) 0.5766 2023-10-11 12:22:45,604 saving best model 2023-10-11 12:22:48,091 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:22:57,047 epoch 4 - iter 27/272 - loss 0.17559727 - time (sec): 8.95 - samples/sec: 544.28 - lr: 0.000123 - momentum: 0.000000 2023-10-11 12:23:06,445 epoch 4 - iter 54/272 - loss 0.14763057 - time (sec): 18.35 - samples/sec: 564.09 - lr: 0.000121 - momentum: 0.000000 2023-10-11 12:23:16,156 epoch 4 - iter 81/272 - loss 0.16759112 - time (sec): 28.06 - samples/sec: 578.45 - lr: 0.000119 - momentum: 0.000000 2023-10-11 12:23:25,501 epoch 4 - iter 108/272 - loss 0.16355731 - time (sec): 37.41 - samples/sec: 574.32 - lr: 0.000117 - momentum: 0.000000 2023-10-11 12:23:35,043 epoch 4 - iter 135/272 - loss 0.15538460 - time (sec): 46.95 - samples/sec: 575.87 - lr: 0.000116 - momentum: 0.000000 2023-10-11 12:23:43,666 epoch 4 - iter 162/272 - loss 0.15731211 - time (sec): 55.57 - samples/sec: 567.27 - lr: 0.000114 - momentum: 0.000000 2023-10-11 12:23:53,322 epoch 4 - iter 189/272 - loss 0.15895911 - time (sec): 65.23 - samples/sec: 569.26 - lr: 0.000112 - momentum: 0.000000 2023-10-11 12:24:02,161 epoch 4 - iter 216/272 - loss 0.15623754 - time (sec): 74.07 - samples/sec: 564.59 - lr: 0.000110 - momentum: 0.000000 2023-10-11 12:24:10,962 epoch 4 - iter 243/272 - loss 0.15352344 - time (sec): 82.87 - samples/sec: 560.85 - lr: 0.000109 - momentum: 0.000000 2023-10-11 12:24:20,498 epoch 4 - iter 270/272 - loss 0.15512637 - time (sec): 92.40 - samples/sec: 560.35 - lr: 0.000107 - momentum: 0.000000 2023-10-11 12:24:20,943 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:24:20,943 EPOCH 4 done: loss 0.1547 - lr: 0.000107 2023-10-11 12:24:26,384 DEV : loss 0.15080419182777405 - f1-score (micro avg) 0.6617 2023-10-11 12:24:26,392 saving best model 2023-10-11 12:24:28,921 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:24:38,560 epoch 5 - iter 27/272 - loss 0.11298860 - time (sec): 9.63 - samples/sec: 581.63 - lr: 0.000105 - momentum: 0.000000 2023-10-11 12:24:47,611 epoch 5 - iter 54/272 - loss 0.11828460 - time (sec): 18.69 - samples/sec: 562.67 - lr: 0.000103 - momentum: 0.000000 2023-10-11 12:24:57,149 epoch 5 - iter 81/272 - loss 0.12324494 - time (sec): 28.22 - samples/sec: 576.89 - lr: 0.000101 - momentum: 0.000000 2023-10-11 12:25:06,969 epoch 5 - iter 108/272 - loss 0.11225919 - time (sec): 38.04 - samples/sec: 581.49 - lr: 0.000100 - momentum: 0.000000 2023-10-11 12:25:15,898 epoch 5 - iter 135/272 - loss 0.10889944 - time (sec): 46.97 - samples/sec: 577.04 - lr: 0.000098 - momentum: 0.000000 2023-10-11 12:25:24,717 epoch 5 - iter 162/272 - loss 0.10759412 - time (sec): 55.79 - samples/sec: 569.39 - lr: 0.000096 - momentum: 0.000000 2023-10-11 12:25:33,490 epoch 5 - iter 189/272 - loss 0.10450328 - time (sec): 64.56 - samples/sec: 563.78 - lr: 0.000094 - momentum: 0.000000 2023-10-11 12:25:42,652 epoch 5 - iter 216/272 - loss 0.10251942 - time (sec): 73.73 - samples/sec: 564.22 - lr: 0.000093 - momentum: 0.000000 2023-10-11 12:25:51,924 epoch 5 - iter 243/272 - loss 0.10536311 - time (sec): 83.00 - samples/sec: 564.83 - lr: 0.000091 - momentum: 0.000000 2023-10-11 12:26:00,818 epoch 5 - iter 270/272 - loss 0.10201182 - time (sec): 91.89 - samples/sec: 561.89 - lr: 0.000089 - momentum: 0.000000 2023-10-11 12:26:01,388 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:26:01,388 EPOCH 5 done: loss 0.1023 - lr: 0.000089 2023-10-11 12:26:06,848 DEV : loss 0.1377904713153839 - f1-score (micro avg) 0.6462 2023-10-11 12:26:06,856 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:26:16,012 epoch 6 - iter 27/272 - loss 0.06858667 - time (sec): 9.15 - samples/sec: 560.28 - lr: 0.000087 - momentum: 0.000000 2023-10-11 12:26:25,347 epoch 6 - iter 54/272 - loss 0.07270870 - time (sec): 18.49 - samples/sec: 552.38 - lr: 0.000085 - momentum: 0.000000 2023-10-11 12:26:35,819 epoch 6 - iter 81/272 - loss 0.07138524 - time (sec): 28.96 - samples/sec: 571.95 - lr: 0.000084 - momentum: 0.000000 2023-10-11 12:26:44,751 epoch 6 - iter 108/272 - loss 0.07082171 - time (sec): 37.89 - samples/sec: 558.81 - lr: 0.000082 - momentum: 0.000000 2023-10-11 12:26:53,646 epoch 6 - iter 135/272 - loss 0.06942375 - time (sec): 46.79 - samples/sec: 554.10 - lr: 0.000080 - momentum: 0.000000 2023-10-11 12:27:02,878 epoch 6 - iter 162/272 - loss 0.06931374 - time (sec): 56.02 - samples/sec: 555.19 - lr: 0.000078 - momentum: 0.000000 2023-10-11 12:27:11,743 epoch 6 - iter 189/272 - loss 0.07471804 - time (sec): 64.89 - samples/sec: 551.02 - lr: 0.000077 - momentum: 0.000000 2023-10-11 12:27:21,147 epoch 6 - iter 216/272 - loss 0.07358616 - time (sec): 74.29 - samples/sec: 551.36 - lr: 0.000075 - momentum: 0.000000 2023-10-11 12:27:31,007 epoch 6 - iter 243/272 - loss 0.07385375 - time (sec): 84.15 - samples/sec: 554.53 - lr: 0.000073 - momentum: 0.000000 2023-10-11 12:27:40,226 epoch 6 - iter 270/272 - loss 0.07393521 - time (sec): 93.37 - samples/sec: 554.54 - lr: 0.000071 - momentum: 0.000000 2023-10-11 12:27:40,629 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:27:40,630 EPOCH 6 done: loss 0.0737 - lr: 0.000071 2023-10-11 12:27:46,135 DEV : loss 0.13831757009029388 - f1-score (micro avg) 0.7681 2023-10-11 12:27:46,142 saving best model 2023-10-11 12:27:48,649 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:27:57,510 epoch 7 - iter 27/272 - loss 0.05129014 - time (sec): 8.86 - samples/sec: 549.17 - lr: 0.000069 - momentum: 0.000000 2023-10-11 12:28:06,049 epoch 7 - iter 54/272 - loss 0.06171732 - time (sec): 17.40 - samples/sec: 535.80 - lr: 0.000068 - momentum: 0.000000 2023-10-11 12:28:14,620 epoch 7 - iter 81/272 - loss 0.05770185 - time (sec): 25.97 - samples/sec: 533.83 - lr: 0.000066 - momentum: 0.000000 2023-10-11 12:28:23,980 epoch 7 - iter 108/272 - loss 0.05578072 - time (sec): 35.33 - samples/sec: 540.64 - lr: 0.000064 - momentum: 0.000000 2023-10-11 12:28:33,686 epoch 7 - iter 135/272 - loss 0.05183090 - time (sec): 45.03 - samples/sec: 548.95 - lr: 0.000062 - momentum: 0.000000 2023-10-11 12:28:43,514 epoch 7 - iter 162/272 - loss 0.05034293 - time (sec): 54.86 - samples/sec: 556.20 - lr: 0.000061 - momentum: 0.000000 2023-10-11 12:28:52,842 epoch 7 - iter 189/272 - loss 0.05522158 - time (sec): 64.19 - samples/sec: 553.57 - lr: 0.000059 - momentum: 0.000000 2023-10-11 12:29:01,292 epoch 7 - iter 216/272 - loss 0.05317533 - time (sec): 72.64 - samples/sec: 547.22 - lr: 0.000057 - momentum: 0.000000 2023-10-11 12:29:11,075 epoch 7 - iter 243/272 - loss 0.05369366 - time (sec): 82.42 - samples/sec: 554.67 - lr: 0.000055 - momentum: 0.000000 2023-10-11 12:29:21,164 epoch 7 - iter 270/272 - loss 0.05345304 - time (sec): 92.51 - samples/sec: 560.18 - lr: 0.000054 - momentum: 0.000000 2023-10-11 12:29:21,547 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:29:21,548 EPOCH 7 done: loss 0.0535 - lr: 0.000054 2023-10-11 12:29:27,021 DEV : loss 0.14701178669929504 - f1-score (micro avg) 0.7956 2023-10-11 12:29:27,029 saving best model 2023-10-11 12:29:29,508 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:29:38,586 epoch 8 - iter 27/272 - loss 0.02662217 - time (sec): 9.07 - samples/sec: 567.16 - lr: 0.000052 - momentum: 0.000000 2023-10-11 12:29:48,267 epoch 8 - iter 54/272 - loss 0.04238572 - time (sec): 18.76 - samples/sec: 584.96 - lr: 0.000050 - momentum: 0.000000 2023-10-11 12:29:57,482 epoch 8 - iter 81/272 - loss 0.04066595 - time (sec): 27.97 - samples/sec: 578.41 - lr: 0.000048 - momentum: 0.000000 2023-10-11 12:30:06,552 epoch 8 - iter 108/272 - loss 0.03956566 - time (sec): 37.04 - samples/sec: 565.28 - lr: 0.000046 - momentum: 0.000000 2023-10-11 12:30:16,114 epoch 8 - iter 135/272 - loss 0.03767651 - time (sec): 46.60 - samples/sec: 567.35 - lr: 0.000045 - momentum: 0.000000 2023-10-11 12:30:26,166 epoch 8 - iter 162/272 - loss 0.04051798 - time (sec): 56.65 - samples/sec: 577.03 - lr: 0.000043 - momentum: 0.000000 2023-10-11 12:30:34,685 epoch 8 - iter 189/272 - loss 0.04409632 - time (sec): 65.17 - samples/sec: 565.84 - lr: 0.000041 - momentum: 0.000000 2023-10-11 12:30:44,286 epoch 8 - iter 216/272 - loss 0.04433015 - time (sec): 74.77 - samples/sec: 566.86 - lr: 0.000039 - momentum: 0.000000 2023-10-11 12:30:53,159 epoch 8 - iter 243/272 - loss 0.04347717 - time (sec): 83.65 - samples/sec: 563.86 - lr: 0.000038 - momentum: 0.000000 2023-10-11 12:31:02,010 epoch 8 - iter 270/272 - loss 0.04284592 - time (sec): 92.50 - samples/sec: 559.27 - lr: 0.000036 - momentum: 0.000000 2023-10-11 12:31:02,455 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:31:02,455 EPOCH 8 done: loss 0.0426 - lr: 0.000036 2023-10-11 12:31:07,906 DEV : loss 0.14679642021656036 - f1-score (micro avg) 0.8015 2023-10-11 12:31:07,914 saving best model 2023-10-11 12:31:10,409 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:31:19,316 epoch 9 - iter 27/272 - loss 0.03044480 - time (sec): 8.90 - samples/sec: 549.94 - lr: 0.000034 - momentum: 0.000000 2023-10-11 12:31:27,617 epoch 9 - iter 54/272 - loss 0.04315447 - time (sec): 17.20 - samples/sec: 538.76 - lr: 0.000032 - momentum: 0.000000 2023-10-11 12:31:36,320 epoch 9 - iter 81/272 - loss 0.03877733 - time (sec): 25.91 - samples/sec: 540.85 - lr: 0.000030 - momentum: 0.000000 2023-10-11 12:31:45,762 epoch 9 - iter 108/272 - loss 0.03541826 - time (sec): 35.35 - samples/sec: 553.43 - lr: 0.000029 - momentum: 0.000000 2023-10-11 12:31:55,099 epoch 9 - iter 135/272 - loss 0.03420976 - time (sec): 44.69 - samples/sec: 559.43 - lr: 0.000027 - momentum: 0.000000 2023-10-11 12:32:04,049 epoch 9 - iter 162/272 - loss 0.03216315 - time (sec): 53.64 - samples/sec: 558.43 - lr: 0.000025 - momentum: 0.000000 2023-10-11 12:32:13,330 epoch 9 - iter 189/272 - loss 0.03081178 - time (sec): 62.92 - samples/sec: 558.94 - lr: 0.000023 - momentum: 0.000000 2023-10-11 12:32:23,043 epoch 9 - iter 216/272 - loss 0.03134856 - time (sec): 72.63 - samples/sec: 564.70 - lr: 0.000022 - momentum: 0.000000 2023-10-11 12:32:32,028 epoch 9 - iter 243/272 - loss 0.03443496 - time (sec): 81.61 - samples/sec: 562.97 - lr: 0.000020 - momentum: 0.000000 2023-10-11 12:32:41,604 epoch 9 - iter 270/272 - loss 0.03386938 - time (sec): 91.19 - samples/sec: 567.00 - lr: 0.000018 - momentum: 0.000000 2023-10-11 12:32:42,083 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:32:42,083 EPOCH 9 done: loss 0.0337 - lr: 0.000018 2023-10-11 12:32:47,749 DEV : loss 0.149958074092865 - f1-score (micro avg) 0.8096 2023-10-11 12:32:47,757 saving best model 2023-10-11 12:32:50,259 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:32:59,526 epoch 10 - iter 27/272 - loss 0.03830851 - time (sec): 9.26 - samples/sec: 576.23 - lr: 0.000016 - momentum: 0.000000 2023-10-11 12:33:08,503 epoch 10 - iter 54/272 - loss 0.03546184 - time (sec): 18.24 - samples/sec: 571.99 - lr: 0.000014 - momentum: 0.000000 2023-10-11 12:33:17,103 epoch 10 - iter 81/272 - loss 0.03145260 - time (sec): 26.84 - samples/sec: 562.54 - lr: 0.000013 - momentum: 0.000000 2023-10-11 12:33:26,206 epoch 10 - iter 108/272 - loss 0.03259123 - time (sec): 35.94 - samples/sec: 557.04 - lr: 0.000011 - momentum: 0.000000 2023-10-11 12:33:36,463 epoch 10 - iter 135/272 - loss 0.02974927 - time (sec): 46.20 - samples/sec: 570.11 - lr: 0.000009 - momentum: 0.000000 2023-10-11 12:33:45,171 epoch 10 - iter 162/272 - loss 0.02889802 - time (sec): 54.91 - samples/sec: 558.86 - lr: 0.000007 - momentum: 0.000000 2023-10-11 12:33:54,983 epoch 10 - iter 189/272 - loss 0.02806859 - time (sec): 64.72 - samples/sec: 559.46 - lr: 0.000005 - momentum: 0.000000 2023-10-11 12:34:04,346 epoch 10 - iter 216/272 - loss 0.02816137 - time (sec): 74.08 - samples/sec: 557.21 - lr: 0.000004 - momentum: 0.000000 2023-10-11 12:34:13,684 epoch 10 - iter 243/272 - loss 0.02889036 - time (sec): 83.42 - samples/sec: 558.00 - lr: 0.000002 - momentum: 0.000000 2023-10-11 12:34:23,246 epoch 10 - iter 270/272 - loss 0.02974897 - time (sec): 92.98 - samples/sec: 556.61 - lr: 0.000000 - momentum: 0.000000 2023-10-11 12:34:23,713 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:34:23,713 EPOCH 10 done: loss 0.0297 - lr: 0.000000 2023-10-11 12:34:29,409 DEV : loss 0.15252262353897095 - f1-score (micro avg) 0.8051 2023-10-11 12:34:30,254 ---------------------------------------------------------------------------------------------------- 2023-10-11 12:34:30,256 Loading model from best epoch ... 2023-10-11 12:34:34,051 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 12:34:46,395 Results: - F-score (micro) 0.7562 - F-score (macro) 0.6778 - Accuracy 0.6276 By class: precision recall f1-score support LOC 0.7437 0.8558 0.7958 312 PER 0.7287 0.8654 0.7912 208 ORG 0.3793 0.4000 0.3894 55 HumanProd 0.6667 0.8182 0.7347 22 micro avg 0.7048 0.8157 0.7562 597 macro avg 0.6296 0.7348 0.6778 597 weighted avg 0.7021 0.8157 0.7545 597 2023-10-11 12:34:46,395 ----------------------------------------------------------------------------------------------------