2023-10-11 09:57:56,378 ---------------------------------------------------------------------------------------------------- 2023-10-11 09:57:56,380 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 09:57:56,380 ---------------------------------------------------------------------------------------------------- 2023-10-11 09:57:56,381 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 09:57:56,381 ---------------------------------------------------------------------------------------------------- 2023-10-11 09:57:56,381 Train: 1085 sentences 2023-10-11 09:57:56,381 (train_with_dev=False, train_with_test=False) 2023-10-11 09:57:56,381 ---------------------------------------------------------------------------------------------------- 2023-10-11 09:57:56,381 Training Params: 2023-10-11 09:57:56,381 - learning_rate: "0.00016" 2023-10-11 09:57:56,381 - mini_batch_size: "4" 2023-10-11 09:57:56,381 - max_epochs: "10" 2023-10-11 09:57:56,381 - shuffle: "True" 2023-10-11 09:57:56,381 ---------------------------------------------------------------------------------------------------- 2023-10-11 09:57:56,381 Plugins: 2023-10-11 09:57:56,381 - TensorboardLogger 2023-10-11 09:57:56,382 - LinearScheduler | warmup_fraction: '0.1' 2023-10-11 09:57:56,382 ---------------------------------------------------------------------------------------------------- 2023-10-11 09:57:56,382 Final evaluation on model from best epoch (best-model.pt) 2023-10-11 09:57:56,382 - metric: "('micro avg', 'f1-score')" 2023-10-11 09:57:56,382 ---------------------------------------------------------------------------------------------------- 2023-10-11 09:57:56,382 Computation: 2023-10-11 09:57:56,382 - compute on device: cuda:0 2023-10-11 09:57:56,382 - embedding storage: none 2023-10-11 09:57:56,382 ---------------------------------------------------------------------------------------------------- 2023-10-11 09:57:56,382 Model training base path: "hmbench-newseye/sv-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-2" 2023-10-11 09:57:56,382 ---------------------------------------------------------------------------------------------------- 2023-10-11 09:57:56,382 ---------------------------------------------------------------------------------------------------- 2023-10-11 09:57:56,382 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-11 09:58:05,782 epoch 1 - iter 27/272 - loss 2.84961659 - time (sec): 9.40 - samples/sec: 576.00 - lr: 0.000015 - momentum: 0.000000 2023-10-11 09:58:14,469 epoch 1 - iter 54/272 - loss 2.83915292 - time (sec): 18.08 - samples/sec: 545.61 - lr: 0.000031 - momentum: 0.000000 2023-10-11 09:58:23,736 epoch 1 - iter 81/272 - loss 2.81803749 - time (sec): 27.35 - samples/sec: 553.45 - lr: 0.000047 - momentum: 0.000000 2023-10-11 09:58:33,589 epoch 1 - iter 108/272 - loss 2.74804522 - time (sec): 37.20 - samples/sec: 564.68 - lr: 0.000063 - momentum: 0.000000 2023-10-11 09:58:42,886 epoch 1 - iter 135/272 - loss 2.65294364 - time (sec): 46.50 - samples/sec: 566.48 - lr: 0.000079 - momentum: 0.000000 2023-10-11 09:58:51,421 epoch 1 - iter 162/272 - loss 2.56578574 - time (sec): 55.04 - samples/sec: 556.46 - lr: 0.000095 - momentum: 0.000000 2023-10-11 09:59:00,707 epoch 1 - iter 189/272 - loss 2.45465098 - time (sec): 64.32 - samples/sec: 554.14 - lr: 0.000111 - momentum: 0.000000 2023-10-11 09:59:09,883 epoch 1 - iter 216/272 - loss 2.33901060 - time (sec): 73.50 - samples/sec: 554.24 - lr: 0.000126 - momentum: 0.000000 2023-10-11 09:59:19,912 epoch 1 - iter 243/272 - loss 2.19029882 - time (sec): 83.53 - samples/sec: 558.64 - lr: 0.000142 - momentum: 0.000000 2023-10-11 09:59:29,206 epoch 1 - iter 270/272 - loss 2.07205638 - time (sec): 92.82 - samples/sec: 559.30 - lr: 0.000158 - momentum: 0.000000 2023-10-11 09:59:29,534 ---------------------------------------------------------------------------------------------------- 2023-10-11 09:59:29,535 EPOCH 1 done: loss 2.0705 - lr: 0.000158 2023-10-11 09:59:34,650 DEV : loss 0.7345565557479858 - f1-score (micro avg) 0.0 2023-10-11 09:59:34,658 ---------------------------------------------------------------------------------------------------- 2023-10-11 09:59:44,295 epoch 2 - iter 27/272 - loss 0.70414994 - time (sec): 9.63 - samples/sec: 601.26 - lr: 0.000158 - momentum: 0.000000 2023-10-11 09:59:53,165 epoch 2 - iter 54/272 - loss 0.63540563 - time (sec): 18.50 - samples/sec: 581.59 - lr: 0.000157 - momentum: 0.000000 2023-10-11 10:00:02,825 epoch 2 - iter 81/272 - loss 0.63286981 - time (sec): 28.16 - samples/sec: 590.61 - lr: 0.000155 - momentum: 0.000000 2023-10-11 10:00:12,024 epoch 2 - iter 108/272 - loss 0.60068837 - time (sec): 37.36 - samples/sec: 583.88 - lr: 0.000153 - momentum: 0.000000 2023-10-11 10:00:21,275 epoch 2 - iter 135/272 - loss 0.58169819 - time (sec): 46.61 - samples/sec: 578.50 - lr: 0.000151 - momentum: 0.000000 2023-10-11 10:00:29,801 epoch 2 - iter 162/272 - loss 0.55542542 - time (sec): 55.14 - samples/sec: 569.54 - lr: 0.000149 - momentum: 0.000000 2023-10-11 10:00:38,726 epoch 2 - iter 189/272 - loss 0.54212044 - time (sec): 64.07 - samples/sec: 563.12 - lr: 0.000148 - momentum: 0.000000 2023-10-11 10:00:48,281 epoch 2 - iter 216/272 - loss 0.51858578 - time (sec): 73.62 - samples/sec: 561.94 - lr: 0.000146 - momentum: 0.000000 2023-10-11 10:00:57,669 epoch 2 - iter 243/272 - loss 0.50379454 - time (sec): 83.01 - samples/sec: 557.51 - lr: 0.000144 - momentum: 0.000000 2023-10-11 10:01:08,182 epoch 2 - iter 270/272 - loss 0.49383828 - time (sec): 93.52 - samples/sec: 554.15 - lr: 0.000142 - momentum: 0.000000 2023-10-11 10:01:08,590 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:01:08,590 EPOCH 2 done: loss 0.4930 - lr: 0.000142 2023-10-11 10:01:14,578 DEV : loss 0.3049907982349396 - f1-score (micro avg) 0.2867 2023-10-11 10:01:14,589 saving best model 2023-10-11 10:01:15,564 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:01:26,175 epoch 3 - iter 27/272 - loss 0.36825189 - time (sec): 10.61 - samples/sec: 495.68 - lr: 0.000141 - momentum: 0.000000 2023-10-11 10:01:36,526 epoch 3 - iter 54/272 - loss 0.35828698 - time (sec): 20.96 - samples/sec: 489.94 - lr: 0.000139 - momentum: 0.000000 2023-10-11 10:01:46,586 epoch 3 - iter 81/272 - loss 0.33371894 - time (sec): 31.02 - samples/sec: 489.93 - lr: 0.000137 - momentum: 0.000000 2023-10-11 10:01:56,178 epoch 3 - iter 108/272 - loss 0.32842770 - time (sec): 40.61 - samples/sec: 500.40 - lr: 0.000135 - momentum: 0.000000 2023-10-11 10:02:06,182 epoch 3 - iter 135/272 - loss 0.32494522 - time (sec): 50.62 - samples/sec: 513.93 - lr: 0.000133 - momentum: 0.000000 2023-10-11 10:02:15,811 epoch 3 - iter 162/272 - loss 0.31495256 - time (sec): 60.24 - samples/sec: 514.72 - lr: 0.000132 - momentum: 0.000000 2023-10-11 10:02:26,550 epoch 3 - iter 189/272 - loss 0.31307865 - time (sec): 70.98 - samples/sec: 525.09 - lr: 0.000130 - momentum: 0.000000 2023-10-11 10:02:36,602 epoch 3 - iter 216/272 - loss 0.30261968 - time (sec): 81.04 - samples/sec: 527.34 - lr: 0.000128 - momentum: 0.000000 2023-10-11 10:02:45,602 epoch 3 - iter 243/272 - loss 0.30154536 - time (sec): 90.04 - samples/sec: 520.58 - lr: 0.000126 - momentum: 0.000000 2023-10-11 10:02:55,043 epoch 3 - iter 270/272 - loss 0.29990213 - time (sec): 99.48 - samples/sec: 519.97 - lr: 0.000125 - momentum: 0.000000 2023-10-11 10:02:55,531 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:02:55,531 EPOCH 3 done: loss 0.3001 - lr: 0.000125 2023-10-11 10:03:01,411 DEV : loss 0.23118416965007782 - f1-score (micro avg) 0.4514 2023-10-11 10:03:01,419 saving best model 2023-10-11 10:03:03,968 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:03:13,204 epoch 4 - iter 27/272 - loss 0.26389527 - time (sec): 9.23 - samples/sec: 543.60 - lr: 0.000123 - momentum: 0.000000 2023-10-11 10:03:22,197 epoch 4 - iter 54/272 - loss 0.23946997 - time (sec): 18.22 - samples/sec: 524.90 - lr: 0.000121 - momentum: 0.000000 2023-10-11 10:03:32,348 epoch 4 - iter 81/272 - loss 0.22494046 - time (sec): 28.37 - samples/sec: 549.85 - lr: 0.000119 - momentum: 0.000000 2023-10-11 10:03:42,207 epoch 4 - iter 108/272 - loss 0.22128238 - time (sec): 38.23 - samples/sec: 551.86 - lr: 0.000117 - momentum: 0.000000 2023-10-11 10:03:51,557 epoch 4 - iter 135/272 - loss 0.21790054 - time (sec): 47.58 - samples/sec: 546.55 - lr: 0.000116 - momentum: 0.000000 2023-10-11 10:04:01,550 epoch 4 - iter 162/272 - loss 0.21355483 - time (sec): 57.58 - samples/sec: 549.19 - lr: 0.000114 - momentum: 0.000000 2023-10-11 10:04:10,733 epoch 4 - iter 189/272 - loss 0.21461498 - time (sec): 66.76 - samples/sec: 543.62 - lr: 0.000112 - momentum: 0.000000 2023-10-11 10:04:20,210 epoch 4 - iter 216/272 - loss 0.21161711 - time (sec): 76.24 - samples/sec: 543.55 - lr: 0.000110 - momentum: 0.000000 2023-10-11 10:04:29,699 epoch 4 - iter 243/272 - loss 0.21439789 - time (sec): 85.72 - samples/sec: 544.83 - lr: 0.000109 - momentum: 0.000000 2023-10-11 10:04:38,989 epoch 4 - iter 270/272 - loss 0.21064388 - time (sec): 95.01 - samples/sec: 545.07 - lr: 0.000107 - momentum: 0.000000 2023-10-11 10:04:39,417 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:04:39,418 EPOCH 4 done: loss 0.2106 - lr: 0.000107 2023-10-11 10:04:45,299 DEV : loss 0.17590400576591492 - f1-score (micro avg) 0.5839 2023-10-11 10:04:45,309 saving best model 2023-10-11 10:04:47,847 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:04:56,859 epoch 5 - iter 27/272 - loss 0.16095320 - time (sec): 9.01 - samples/sec: 518.32 - lr: 0.000105 - momentum: 0.000000 2023-10-11 10:05:06,266 epoch 5 - iter 54/272 - loss 0.15303063 - time (sec): 18.41 - samples/sec: 545.17 - lr: 0.000103 - momentum: 0.000000 2023-10-11 10:05:15,581 epoch 5 - iter 81/272 - loss 0.14715890 - time (sec): 27.73 - samples/sec: 553.13 - lr: 0.000101 - momentum: 0.000000 2023-10-11 10:05:24,566 epoch 5 - iter 108/272 - loss 0.14859354 - time (sec): 36.71 - samples/sec: 551.36 - lr: 0.000100 - momentum: 0.000000 2023-10-11 10:05:33,383 epoch 5 - iter 135/272 - loss 0.13938521 - time (sec): 45.53 - samples/sec: 549.22 - lr: 0.000098 - momentum: 0.000000 2023-10-11 10:05:43,089 epoch 5 - iter 162/272 - loss 0.13872093 - time (sec): 55.24 - samples/sec: 559.36 - lr: 0.000096 - momentum: 0.000000 2023-10-11 10:05:52,177 epoch 5 - iter 189/272 - loss 0.14314448 - time (sec): 64.33 - samples/sec: 558.08 - lr: 0.000094 - momentum: 0.000000 2023-10-11 10:06:01,633 epoch 5 - iter 216/272 - loss 0.14519265 - time (sec): 73.78 - samples/sec: 559.16 - lr: 0.000093 - momentum: 0.000000 2023-10-11 10:06:11,241 epoch 5 - iter 243/272 - loss 0.14574570 - time (sec): 83.39 - samples/sec: 559.96 - lr: 0.000091 - momentum: 0.000000 2023-10-11 10:06:20,699 epoch 5 - iter 270/272 - loss 0.14307060 - time (sec): 92.85 - samples/sec: 556.93 - lr: 0.000089 - momentum: 0.000000 2023-10-11 10:06:21,219 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:06:21,219 EPOCH 5 done: loss 0.1430 - lr: 0.000089 2023-10-11 10:06:26,814 DEV : loss 0.15896683931350708 - f1-score (micro avg) 0.6123 2023-10-11 10:06:26,824 saving best model 2023-10-11 10:06:29,363 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:06:39,090 epoch 6 - iter 27/272 - loss 0.12861859 - time (sec): 9.72 - samples/sec: 563.75 - lr: 0.000087 - momentum: 0.000000 2023-10-11 10:06:48,027 epoch 6 - iter 54/272 - loss 0.12625714 - time (sec): 18.66 - samples/sec: 546.90 - lr: 0.000085 - momentum: 0.000000 2023-10-11 10:06:57,647 epoch 6 - iter 81/272 - loss 0.11812378 - time (sec): 28.28 - samples/sec: 551.60 - lr: 0.000084 - momentum: 0.000000 2023-10-11 10:07:07,496 epoch 6 - iter 108/272 - loss 0.11590135 - time (sec): 38.13 - samples/sec: 563.80 - lr: 0.000082 - momentum: 0.000000 2023-10-11 10:07:16,590 epoch 6 - iter 135/272 - loss 0.11257005 - time (sec): 47.22 - samples/sec: 548.24 - lr: 0.000080 - momentum: 0.000000 2023-10-11 10:07:26,339 epoch 6 - iter 162/272 - loss 0.10607560 - time (sec): 56.97 - samples/sec: 553.59 - lr: 0.000078 - momentum: 0.000000 2023-10-11 10:07:35,893 epoch 6 - iter 189/272 - loss 0.10194287 - time (sec): 66.53 - samples/sec: 550.51 - lr: 0.000077 - momentum: 0.000000 2023-10-11 10:07:45,318 epoch 6 - iter 216/272 - loss 0.10614524 - time (sec): 75.95 - samples/sec: 547.98 - lr: 0.000075 - momentum: 0.000000 2023-10-11 10:07:54,921 epoch 6 - iter 243/272 - loss 0.10334886 - time (sec): 85.55 - samples/sec: 547.68 - lr: 0.000073 - momentum: 0.000000 2023-10-11 10:08:04,092 epoch 6 - iter 270/272 - loss 0.10347814 - time (sec): 94.72 - samples/sec: 546.02 - lr: 0.000071 - momentum: 0.000000 2023-10-11 10:08:04,598 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:08:04,599 EPOCH 6 done: loss 0.1033 - lr: 0.000071 2023-10-11 10:08:10,290 DEV : loss 0.1453726589679718 - f1-score (micro avg) 0.6964 2023-10-11 10:08:10,299 saving best model 2023-10-11 10:08:12,837 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:08:22,005 epoch 7 - iter 27/272 - loss 0.08200013 - time (sec): 9.16 - samples/sec: 526.74 - lr: 0.000069 - momentum: 0.000000 2023-10-11 10:08:31,942 epoch 7 - iter 54/272 - loss 0.07169928 - time (sec): 19.10 - samples/sec: 549.62 - lr: 0.000068 - momentum: 0.000000 2023-10-11 10:08:41,790 epoch 7 - iter 81/272 - loss 0.06750559 - time (sec): 28.95 - samples/sec: 552.18 - lr: 0.000066 - momentum: 0.000000 2023-10-11 10:08:51,214 epoch 7 - iter 108/272 - loss 0.06876153 - time (sec): 38.37 - samples/sec: 549.77 - lr: 0.000064 - momentum: 0.000000 2023-10-11 10:09:01,123 epoch 7 - iter 135/272 - loss 0.07382666 - time (sec): 48.28 - samples/sec: 549.32 - lr: 0.000062 - momentum: 0.000000 2023-10-11 10:09:10,687 epoch 7 - iter 162/272 - loss 0.07376310 - time (sec): 57.85 - samples/sec: 542.87 - lr: 0.000061 - momentum: 0.000000 2023-10-11 10:09:19,909 epoch 7 - iter 189/272 - loss 0.07298171 - time (sec): 67.07 - samples/sec: 533.52 - lr: 0.000059 - momentum: 0.000000 2023-10-11 10:09:30,231 epoch 7 - iter 216/272 - loss 0.07154073 - time (sec): 77.39 - samples/sec: 535.81 - lr: 0.000057 - momentum: 0.000000 2023-10-11 10:09:39,463 epoch 7 - iter 243/272 - loss 0.07659368 - time (sec): 86.62 - samples/sec: 535.04 - lr: 0.000055 - momentum: 0.000000 2023-10-11 10:09:49,375 epoch 7 - iter 270/272 - loss 0.07547255 - time (sec): 96.53 - samples/sec: 536.22 - lr: 0.000054 - momentum: 0.000000 2023-10-11 10:09:49,841 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:09:49,841 EPOCH 7 done: loss 0.0753 - lr: 0.000054 2023-10-11 10:09:55,608 DEV : loss 0.1472298800945282 - f1-score (micro avg) 0.7478 2023-10-11 10:09:55,617 saving best model 2023-10-11 10:09:58,161 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:10:07,364 epoch 8 - iter 27/272 - loss 0.07327268 - time (sec): 9.20 - samples/sec: 499.40 - lr: 0.000052 - momentum: 0.000000 2023-10-11 10:10:16,567 epoch 8 - iter 54/272 - loss 0.05912708 - time (sec): 18.40 - samples/sec: 510.50 - lr: 0.000050 - momentum: 0.000000 2023-10-11 10:10:26,716 epoch 8 - iter 81/272 - loss 0.06315843 - time (sec): 28.55 - samples/sec: 538.72 - lr: 0.000048 - momentum: 0.000000 2023-10-11 10:10:36,403 epoch 8 - iter 108/272 - loss 0.06440136 - time (sec): 38.24 - samples/sec: 540.82 - lr: 0.000046 - momentum: 0.000000 2023-10-11 10:10:45,979 epoch 8 - iter 135/272 - loss 0.06543166 - time (sec): 47.81 - samples/sec: 541.26 - lr: 0.000045 - momentum: 0.000000 2023-10-11 10:10:54,660 epoch 8 - iter 162/272 - loss 0.06830101 - time (sec): 56.49 - samples/sec: 531.16 - lr: 0.000043 - momentum: 0.000000 2023-10-11 10:11:04,624 epoch 8 - iter 189/272 - loss 0.06687904 - time (sec): 66.46 - samples/sec: 535.98 - lr: 0.000041 - momentum: 0.000000 2023-10-11 10:11:15,233 epoch 8 - iter 216/272 - loss 0.06417230 - time (sec): 77.07 - samples/sec: 545.82 - lr: 0.000039 - momentum: 0.000000 2023-10-11 10:11:24,573 epoch 8 - iter 243/272 - loss 0.06203206 - time (sec): 86.41 - samples/sec: 541.39 - lr: 0.000038 - momentum: 0.000000 2023-10-11 10:11:34,240 epoch 8 - iter 270/272 - loss 0.05987987 - time (sec): 96.07 - samples/sec: 539.08 - lr: 0.000036 - momentum: 0.000000 2023-10-11 10:11:34,685 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:11:34,686 EPOCH 8 done: loss 0.0599 - lr: 0.000036 2023-10-11 10:11:40,788 DEV : loss 0.1424403041601181 - f1-score (micro avg) 0.7653 2023-10-11 10:11:40,796 saving best model 2023-10-11 10:11:43,341 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:11:51,856 epoch 9 - iter 27/272 - loss 0.07666643 - time (sec): 8.51 - samples/sec: 504.34 - lr: 0.000034 - momentum: 0.000000 2023-10-11 10:12:01,071 epoch 9 - iter 54/272 - loss 0.07280220 - time (sec): 17.73 - samples/sec: 533.59 - lr: 0.000032 - momentum: 0.000000 2023-10-11 10:12:10,786 epoch 9 - iter 81/272 - loss 0.05972387 - time (sec): 27.44 - samples/sec: 527.72 - lr: 0.000030 - momentum: 0.000000 2023-10-11 10:12:20,137 epoch 9 - iter 108/272 - loss 0.06294634 - time (sec): 36.79 - samples/sec: 531.16 - lr: 0.000029 - momentum: 0.000000 2023-10-11 10:12:29,121 epoch 9 - iter 135/272 - loss 0.06234610 - time (sec): 45.78 - samples/sec: 522.90 - lr: 0.000027 - momentum: 0.000000 2023-10-11 10:12:39,294 epoch 9 - iter 162/272 - loss 0.05914625 - time (sec): 55.95 - samples/sec: 534.53 - lr: 0.000025 - momentum: 0.000000 2023-10-11 10:12:48,743 epoch 9 - iter 189/272 - loss 0.05774082 - time (sec): 65.40 - samples/sec: 535.18 - lr: 0.000023 - momentum: 0.000000 2023-10-11 10:12:58,443 epoch 9 - iter 216/272 - loss 0.05539923 - time (sec): 75.10 - samples/sec: 537.09 - lr: 0.000022 - momentum: 0.000000 2023-10-11 10:13:08,510 epoch 9 - iter 243/272 - loss 0.05258815 - time (sec): 85.16 - samples/sec: 543.20 - lr: 0.000020 - momentum: 0.000000 2023-10-11 10:13:18,255 epoch 9 - iter 270/272 - loss 0.05055787 - time (sec): 94.91 - samples/sec: 545.35 - lr: 0.000018 - momentum: 0.000000 2023-10-11 10:13:18,703 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:13:18,703 EPOCH 9 done: loss 0.0506 - lr: 0.000018 2023-10-11 10:13:24,318 DEV : loss 0.14669708907604218 - f1-score (micro avg) 0.7576 2023-10-11 10:13:24,327 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:13:34,114 epoch 10 - iter 27/272 - loss 0.05704288 - time (sec): 9.79 - samples/sec: 547.25 - lr: 0.000016 - momentum: 0.000000 2023-10-11 10:13:43,165 epoch 10 - iter 54/272 - loss 0.06282741 - time (sec): 18.84 - samples/sec: 532.27 - lr: 0.000014 - momentum: 0.000000 2023-10-11 10:13:52,252 epoch 10 - iter 81/272 - loss 0.05470713 - time (sec): 27.92 - samples/sec: 535.00 - lr: 0.000013 - momentum: 0.000000 2023-10-11 10:14:01,156 epoch 10 - iter 108/272 - loss 0.05095954 - time (sec): 36.83 - samples/sec: 534.87 - lr: 0.000011 - momentum: 0.000000 2023-10-11 10:14:11,443 epoch 10 - iter 135/272 - loss 0.04981593 - time (sec): 47.11 - samples/sec: 550.36 - lr: 0.000009 - momentum: 0.000000 2023-10-11 10:14:20,845 epoch 10 - iter 162/272 - loss 0.04680562 - time (sec): 56.52 - samples/sec: 545.26 - lr: 0.000007 - momentum: 0.000000 2023-10-11 10:14:30,321 epoch 10 - iter 189/272 - loss 0.04579138 - time (sec): 65.99 - samples/sec: 545.81 - lr: 0.000005 - momentum: 0.000000 2023-10-11 10:14:39,821 epoch 10 - iter 216/272 - loss 0.04474162 - time (sec): 75.49 - samples/sec: 548.68 - lr: 0.000004 - momentum: 0.000000 2023-10-11 10:14:49,718 epoch 10 - iter 243/272 - loss 0.04341634 - time (sec): 85.39 - samples/sec: 549.72 - lr: 0.000002 - momentum: 0.000000 2023-10-11 10:14:58,779 epoch 10 - iter 270/272 - loss 0.04483462 - time (sec): 94.45 - samples/sec: 547.28 - lr: 0.000000 - momentum: 0.000000 2023-10-11 10:14:59,278 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:14:59,278 EPOCH 10 done: loss 0.0447 - lr: 0.000000 2023-10-11 10:15:04,833 DEV : loss 0.14579297602176666 - f1-score (micro avg) 0.7607 2023-10-11 10:15:05,711 ---------------------------------------------------------------------------------------------------- 2023-10-11 10:15:05,713 Loading model from best epoch ... 2023-10-11 10:15:09,559 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 10:15:22,453 Results: - F-score (micro) 0.7359 - F-score (macro) 0.6371 - Accuracy 0.6043 By class: precision recall f1-score support LOC 0.7330 0.8622 0.7923 312 PER 0.6784 0.8317 0.7473 208 ORG 0.3415 0.2545 0.2917 55 HumanProd 0.6129 0.8636 0.7170 22 micro avg 0.6844 0.7956 0.7359 597 macro avg 0.5914 0.7030 0.6371 597 weighted avg 0.6735 0.7956 0.7277 597 2023-10-11 10:15:22,454 ----------------------------------------------------------------------------------------------------