2023-10-17 13:22:43,668 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:22:43,670 Model: "SequenceTagger( (embeddings): TransformerWordEmbeddings( (model): ElectraModel( (embeddings): ElectraEmbeddings( (word_embeddings): Embedding(32001, 768) (position_embeddings): Embedding(512, 768) (token_type_embeddings): Embedding(2, 768) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) (encoder): ElectraEncoder( (layer): ModuleList( (0-11): 12 x ElectraLayer( (attention): ElectraAttention( (self): ElectraSelfAttention( (query): Linear(in_features=768, out_features=768, bias=True) (key): Linear(in_features=768, out_features=768, bias=True) (value): Linear(in_features=768, out_features=768, bias=True) (dropout): Dropout(p=0.1, inplace=False) ) (output): ElectraSelfOutput( (dense): Linear(in_features=768, out_features=768, bias=True) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) (intermediate): ElectraIntermediate( (dense): Linear(in_features=768, out_features=3072, bias=True) (intermediate_act_fn): GELUActivation() ) (output): ElectraOutput( (dense): Linear(in_features=3072, out_features=768, bias=True) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) ) ) ) (locked_dropout): LockedDropout(p=0.5) (linear): Linear(in_features=768, out_features=13, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-17 13:22:43,670 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:22:43,670 MultiCorpus: 6183 train + 680 dev + 2113 test sentences - NER_HIPE_2022 Corpus: 6183 train + 680 dev + 2113 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/topres19th/en/with_doc_seperator 2023-10-17 13:22:43,670 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:22:43,670 Train: 6183 sentences 2023-10-17 13:22:43,670 (train_with_dev=False, train_with_test=False) 2023-10-17 13:22:43,670 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:22:43,670 Training Params: 2023-10-17 13:22:43,670 - learning_rate: "5e-05" 2023-10-17 13:22:43,671 - mini_batch_size: "8" 2023-10-17 13:22:43,671 - max_epochs: "10" 2023-10-17 13:22:43,671 - shuffle: "True" 2023-10-17 13:22:43,671 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:22:43,671 Plugins: 2023-10-17 13:22:43,671 - TensorboardLogger 2023-10-17 13:22:43,671 - LinearScheduler | warmup_fraction: '0.1' 2023-10-17 13:22:43,671 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:22:43,671 Final evaluation on model from best epoch (best-model.pt) 2023-10-17 13:22:43,671 - metric: "('micro avg', 'f1-score')" 2023-10-17 13:22:43,671 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:22:43,671 Computation: 2023-10-17 13:22:43,671 - compute on device: cuda:0 2023-10-17 13:22:43,671 - embedding storage: none 2023-10-17 13:22:43,671 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:22:43,671 Model training base path: "hmbench-topres19th/en-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4" 2023-10-17 13:22:43,672 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:22:43,672 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:22:43,672 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-17 13:22:50,667 epoch 1 - iter 77/773 - loss 2.65276957 - time (sec): 6.99 - samples/sec: 1714.39 - lr: 0.000005 - momentum: 0.000000 2023-10-17 13:22:58,785 epoch 1 - iter 154/773 - loss 1.50483396 - time (sec): 15.11 - samples/sec: 1552.80 - lr: 0.000010 - momentum: 0.000000 2023-10-17 13:23:07,010 epoch 1 - iter 231/773 - loss 1.04258462 - time (sec): 23.34 - samples/sec: 1532.79 - lr: 0.000015 - momentum: 0.000000 2023-10-17 13:23:14,617 epoch 1 - iter 308/773 - loss 0.81592386 - time (sec): 30.94 - samples/sec: 1544.05 - lr: 0.000020 - momentum: 0.000000 2023-10-17 13:23:21,831 epoch 1 - iter 385/773 - loss 0.66498214 - time (sec): 38.16 - samples/sec: 1600.85 - lr: 0.000025 - momentum: 0.000000 2023-10-17 13:23:29,252 epoch 1 - iter 462/773 - loss 0.57446633 - time (sec): 45.58 - samples/sec: 1615.99 - lr: 0.000030 - momentum: 0.000000 2023-10-17 13:23:36,947 epoch 1 - iter 539/773 - loss 0.51014535 - time (sec): 53.27 - samples/sec: 1611.30 - lr: 0.000035 - momentum: 0.000000 2023-10-17 13:23:44,588 epoch 1 - iter 616/773 - loss 0.45806067 - time (sec): 60.91 - samples/sec: 1616.25 - lr: 0.000040 - momentum: 0.000000 2023-10-17 13:23:52,202 epoch 1 - iter 693/773 - loss 0.41752980 - time (sec): 68.53 - samples/sec: 1619.91 - lr: 0.000045 - momentum: 0.000000 2023-10-17 13:23:59,910 epoch 1 - iter 770/773 - loss 0.38405207 - time (sec): 76.24 - samples/sec: 1623.09 - lr: 0.000050 - momentum: 0.000000 2023-10-17 13:24:00,248 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:24:00,248 EPOCH 1 done: loss 0.3827 - lr: 0.000050 2023-10-17 13:24:03,117 DEV : loss 0.07178231328725815 - f1-score (micro avg) 0.7388 2023-10-17 13:24:03,153 saving best model 2023-10-17 13:24:03,704 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:24:11,320 epoch 2 - iter 77/773 - loss 0.10166111 - time (sec): 7.61 - samples/sec: 1610.43 - lr: 0.000049 - momentum: 0.000000 2023-10-17 13:24:19,934 epoch 2 - iter 154/773 - loss 0.09050508 - time (sec): 16.23 - samples/sec: 1543.22 - lr: 0.000049 - momentum: 0.000000 2023-10-17 13:24:28,525 epoch 2 - iter 231/773 - loss 0.07988076 - time (sec): 24.82 - samples/sec: 1489.76 - lr: 0.000048 - momentum: 0.000000 2023-10-17 13:24:36,766 epoch 2 - iter 308/773 - loss 0.08428363 - time (sec): 33.06 - samples/sec: 1488.34 - lr: 0.000048 - momentum: 0.000000 2023-10-17 13:24:44,604 epoch 2 - iter 385/773 - loss 0.08289357 - time (sec): 40.90 - samples/sec: 1503.67 - lr: 0.000047 - momentum: 0.000000 2023-10-17 13:24:52,380 epoch 2 - iter 462/773 - loss 0.08196435 - time (sec): 48.67 - samples/sec: 1511.52 - lr: 0.000047 - momentum: 0.000000 2023-10-17 13:25:00,140 epoch 2 - iter 539/773 - loss 0.08262791 - time (sec): 56.43 - samples/sec: 1528.55 - lr: 0.000046 - momentum: 0.000000 2023-10-17 13:25:07,168 epoch 2 - iter 616/773 - loss 0.08205644 - time (sec): 63.46 - samples/sec: 1544.80 - lr: 0.000046 - momentum: 0.000000 2023-10-17 13:25:14,562 epoch 2 - iter 693/773 - loss 0.08432683 - time (sec): 70.86 - samples/sec: 1567.04 - lr: 0.000045 - momentum: 0.000000 2023-10-17 13:25:22,493 epoch 2 - iter 770/773 - loss 0.08280984 - time (sec): 78.79 - samples/sec: 1573.85 - lr: 0.000044 - momentum: 0.000000 2023-10-17 13:25:22,786 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:25:22,787 EPOCH 2 done: loss 0.0828 - lr: 0.000044 2023-10-17 13:25:26,108 DEV : loss 0.06471428275108337 - f1-score (micro avg) 0.7308 2023-10-17 13:25:26,148 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:25:33,242 epoch 3 - iter 77/773 - loss 0.04734008 - time (sec): 7.09 - samples/sec: 1914.14 - lr: 0.000044 - momentum: 0.000000 2023-10-17 13:25:40,706 epoch 3 - iter 154/773 - loss 0.04641381 - time (sec): 14.56 - samples/sec: 1745.71 - lr: 0.000043 - momentum: 0.000000 2023-10-17 13:25:48,038 epoch 3 - iter 231/773 - loss 0.04809967 - time (sec): 21.89 - samples/sec: 1733.65 - lr: 0.000043 - momentum: 0.000000 2023-10-17 13:25:55,854 epoch 3 - iter 308/773 - loss 0.04990812 - time (sec): 29.70 - samples/sec: 1677.29 - lr: 0.000042 - momentum: 0.000000 2023-10-17 13:26:03,686 epoch 3 - iter 385/773 - loss 0.05048319 - time (sec): 37.54 - samples/sec: 1699.50 - lr: 0.000042 - momentum: 0.000000 2023-10-17 13:26:11,254 epoch 3 - iter 462/773 - loss 0.05159978 - time (sec): 45.10 - samples/sec: 1672.23 - lr: 0.000041 - momentum: 0.000000 2023-10-17 13:26:19,097 epoch 3 - iter 539/773 - loss 0.05384271 - time (sec): 52.95 - samples/sec: 1658.86 - lr: 0.000041 - momentum: 0.000000 2023-10-17 13:26:27,220 epoch 3 - iter 616/773 - loss 0.05339688 - time (sec): 61.07 - samples/sec: 1636.21 - lr: 0.000040 - momentum: 0.000000 2023-10-17 13:26:34,983 epoch 3 - iter 693/773 - loss 0.05430285 - time (sec): 68.83 - samples/sec: 1628.32 - lr: 0.000039 - momentum: 0.000000 2023-10-17 13:26:42,445 epoch 3 - iter 770/773 - loss 0.05408300 - time (sec): 76.29 - samples/sec: 1623.99 - lr: 0.000039 - momentum: 0.000000 2023-10-17 13:26:42,713 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:26:42,713 EPOCH 3 done: loss 0.0539 - lr: 0.000039 2023-10-17 13:26:45,984 DEV : loss 0.07156790047883987 - f1-score (micro avg) 0.7604 2023-10-17 13:26:46,017 saving best model 2023-10-17 13:26:46,701 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:26:54,707 epoch 4 - iter 77/773 - loss 0.03412300 - time (sec): 8.00 - samples/sec: 1555.61 - lr: 0.000038 - momentum: 0.000000 2023-10-17 13:27:02,413 epoch 4 - iter 154/773 - loss 0.03849885 - time (sec): 15.71 - samples/sec: 1595.13 - lr: 0.000038 - momentum: 0.000000 2023-10-17 13:27:10,142 epoch 4 - iter 231/773 - loss 0.03711970 - time (sec): 23.44 - samples/sec: 1565.17 - lr: 0.000037 - momentum: 0.000000 2023-10-17 13:27:17,871 epoch 4 - iter 308/773 - loss 0.03516844 - time (sec): 31.17 - samples/sec: 1585.88 - lr: 0.000037 - momentum: 0.000000 2023-10-17 13:27:25,414 epoch 4 - iter 385/773 - loss 0.03362153 - time (sec): 38.71 - samples/sec: 1605.42 - lr: 0.000036 - momentum: 0.000000 2023-10-17 13:27:33,440 epoch 4 - iter 462/773 - loss 0.03692177 - time (sec): 46.74 - samples/sec: 1601.94 - lr: 0.000036 - momentum: 0.000000 2023-10-17 13:27:40,563 epoch 4 - iter 539/773 - loss 0.03726784 - time (sec): 53.86 - samples/sec: 1622.04 - lr: 0.000035 - momentum: 0.000000 2023-10-17 13:27:48,205 epoch 4 - iter 616/773 - loss 0.04542885 - time (sec): 61.50 - samples/sec: 1615.02 - lr: 0.000034 - momentum: 0.000000 2023-10-17 13:27:56,324 epoch 4 - iter 693/773 - loss 0.04596823 - time (sec): 69.62 - samples/sec: 1598.90 - lr: 0.000034 - momentum: 0.000000 2023-10-17 13:28:03,402 epoch 4 - iter 770/773 - loss 0.04522777 - time (sec): 76.70 - samples/sec: 1614.88 - lr: 0.000033 - momentum: 0.000000 2023-10-17 13:28:03,672 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:28:03,673 EPOCH 4 done: loss 0.0453 - lr: 0.000033 2023-10-17 13:28:06,586 DEV : loss 0.07239558547735214 - f1-score (micro avg) 0.7909 2023-10-17 13:28:06,616 saving best model 2023-10-17 13:28:08,029 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:28:15,411 epoch 5 - iter 77/773 - loss 0.02843471 - time (sec): 7.38 - samples/sec: 1753.59 - lr: 0.000033 - momentum: 0.000000 2023-10-17 13:28:22,936 epoch 5 - iter 154/773 - loss 0.02395437 - time (sec): 14.90 - samples/sec: 1673.42 - lr: 0.000032 - momentum: 0.000000 2023-10-17 13:28:30,854 epoch 5 - iter 231/773 - loss 0.02645457 - time (sec): 22.82 - samples/sec: 1620.38 - lr: 0.000032 - momentum: 0.000000 2023-10-17 13:28:38,767 epoch 5 - iter 308/773 - loss 0.02834241 - time (sec): 30.73 - samples/sec: 1609.37 - lr: 0.000031 - momentum: 0.000000 2023-10-17 13:28:46,449 epoch 5 - iter 385/773 - loss 0.02915922 - time (sec): 38.42 - samples/sec: 1588.49 - lr: 0.000031 - momentum: 0.000000 2023-10-17 13:28:54,069 epoch 5 - iter 462/773 - loss 0.02818305 - time (sec): 46.04 - samples/sec: 1595.37 - lr: 0.000030 - momentum: 0.000000 2023-10-17 13:29:01,800 epoch 5 - iter 539/773 - loss 0.02922204 - time (sec): 53.77 - samples/sec: 1581.52 - lr: 0.000029 - momentum: 0.000000 2023-10-17 13:29:09,568 epoch 5 - iter 616/773 - loss 0.02941636 - time (sec): 61.53 - samples/sec: 1587.22 - lr: 0.000029 - momentum: 0.000000 2023-10-17 13:29:17,758 epoch 5 - iter 693/773 - loss 0.02871886 - time (sec): 69.72 - samples/sec: 1593.88 - lr: 0.000028 - momentum: 0.000000 2023-10-17 13:29:25,953 epoch 5 - iter 770/773 - loss 0.02810066 - time (sec): 77.92 - samples/sec: 1588.80 - lr: 0.000028 - momentum: 0.000000 2023-10-17 13:29:26,265 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:29:26,265 EPOCH 5 done: loss 0.0285 - lr: 0.000028 2023-10-17 13:29:29,547 DEV : loss 0.09678769111633301 - f1-score (micro avg) 0.7579 2023-10-17 13:29:29,578 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:29:37,484 epoch 6 - iter 77/773 - loss 0.02911176 - time (sec): 7.90 - samples/sec: 1494.95 - lr: 0.000027 - momentum: 0.000000 2023-10-17 13:29:44,833 epoch 6 - iter 154/773 - loss 0.02314913 - time (sec): 15.25 - samples/sec: 1620.87 - lr: 0.000027 - momentum: 0.000000 2023-10-17 13:29:52,208 epoch 6 - iter 231/773 - loss 0.02211935 - time (sec): 22.63 - samples/sec: 1654.38 - lr: 0.000026 - momentum: 0.000000 2023-10-17 13:29:59,974 epoch 6 - iter 308/773 - loss 0.01982596 - time (sec): 30.39 - samples/sec: 1657.08 - lr: 0.000026 - momentum: 0.000000 2023-10-17 13:30:07,513 epoch 6 - iter 385/773 - loss 0.01939127 - time (sec): 37.93 - samples/sec: 1633.05 - lr: 0.000025 - momentum: 0.000000 2023-10-17 13:30:14,978 epoch 6 - iter 462/773 - loss 0.02020479 - time (sec): 45.40 - samples/sec: 1624.45 - lr: 0.000024 - momentum: 0.000000 2023-10-17 13:30:22,704 epoch 6 - iter 539/773 - loss 0.01996966 - time (sec): 53.12 - samples/sec: 1610.55 - lr: 0.000024 - momentum: 0.000000 2023-10-17 13:30:30,362 epoch 6 - iter 616/773 - loss 0.01907375 - time (sec): 60.78 - samples/sec: 1610.86 - lr: 0.000023 - momentum: 0.000000 2023-10-17 13:30:38,021 epoch 6 - iter 693/773 - loss 0.01872125 - time (sec): 68.44 - samples/sec: 1621.15 - lr: 0.000023 - momentum: 0.000000 2023-10-17 13:30:45,781 epoch 6 - iter 770/773 - loss 0.01930353 - time (sec): 76.20 - samples/sec: 1624.68 - lr: 0.000022 - momentum: 0.000000 2023-10-17 13:30:46,080 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:30:46,080 EPOCH 6 done: loss 0.0192 - lr: 0.000022 2023-10-17 13:30:49,469 DEV : loss 0.10031934082508087 - f1-score (micro avg) 0.7698 2023-10-17 13:30:49,503 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:30:57,507 epoch 7 - iter 77/773 - loss 0.01926813 - time (sec): 8.00 - samples/sec: 1545.66 - lr: 0.000022 - momentum: 0.000000 2023-10-17 13:31:05,453 epoch 7 - iter 154/773 - loss 0.01404153 - time (sec): 15.95 - samples/sec: 1536.40 - lr: 0.000021 - momentum: 0.000000 2023-10-17 13:31:13,251 epoch 7 - iter 231/773 - loss 0.01790692 - time (sec): 23.74 - samples/sec: 1561.47 - lr: 0.000021 - momentum: 0.000000 2023-10-17 13:31:20,527 epoch 7 - iter 308/773 - loss 0.01596954 - time (sec): 31.02 - samples/sec: 1581.49 - lr: 0.000020 - momentum: 0.000000 2023-10-17 13:31:27,656 epoch 7 - iter 385/773 - loss 0.01481654 - time (sec): 38.15 - samples/sec: 1616.68 - lr: 0.000019 - momentum: 0.000000 2023-10-17 13:31:34,613 epoch 7 - iter 462/773 - loss 0.01463562 - time (sec): 45.11 - samples/sec: 1619.89 - lr: 0.000019 - momentum: 0.000000 2023-10-17 13:31:42,114 epoch 7 - iter 539/773 - loss 0.01492694 - time (sec): 52.61 - samples/sec: 1638.03 - lr: 0.000018 - momentum: 0.000000 2023-10-17 13:31:49,500 epoch 7 - iter 616/773 - loss 0.01420990 - time (sec): 59.99 - samples/sec: 1653.83 - lr: 0.000018 - momentum: 0.000000 2023-10-17 13:31:56,721 epoch 7 - iter 693/773 - loss 0.01390834 - time (sec): 67.21 - samples/sec: 1661.42 - lr: 0.000017 - momentum: 0.000000 2023-10-17 13:32:03,143 epoch 7 - iter 770/773 - loss 0.01406545 - time (sec): 73.64 - samples/sec: 1681.21 - lr: 0.000017 - momentum: 0.000000 2023-10-17 13:32:03,388 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:32:03,389 EPOCH 7 done: loss 0.0140 - lr: 0.000017 2023-10-17 13:32:06,276 DEV : loss 0.11071208119392395 - f1-score (micro avg) 0.7683 2023-10-17 13:32:06,305 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:32:13,130 epoch 8 - iter 77/773 - loss 0.00843154 - time (sec): 6.82 - samples/sec: 1807.24 - lr: 0.000016 - momentum: 0.000000 2023-10-17 13:32:20,080 epoch 8 - iter 154/773 - loss 0.00677113 - time (sec): 13.77 - samples/sec: 1766.95 - lr: 0.000016 - momentum: 0.000000 2023-10-17 13:32:27,329 epoch 8 - iter 231/773 - loss 0.00836479 - time (sec): 21.02 - samples/sec: 1728.27 - lr: 0.000015 - momentum: 0.000000 2023-10-17 13:32:34,265 epoch 8 - iter 308/773 - loss 0.00790711 - time (sec): 27.96 - samples/sec: 1729.60 - lr: 0.000014 - momentum: 0.000000 2023-10-17 13:32:41,366 epoch 8 - iter 385/773 - loss 0.00790440 - time (sec): 35.06 - samples/sec: 1735.15 - lr: 0.000014 - momentum: 0.000000 2023-10-17 13:32:48,613 epoch 8 - iter 462/773 - loss 0.00842461 - time (sec): 42.31 - samples/sec: 1731.46 - lr: 0.000013 - momentum: 0.000000 2023-10-17 13:32:56,661 epoch 8 - iter 539/773 - loss 0.00918393 - time (sec): 50.35 - samples/sec: 1726.83 - lr: 0.000013 - momentum: 0.000000 2023-10-17 13:33:04,126 epoch 8 - iter 616/773 - loss 0.00943817 - time (sec): 57.82 - samples/sec: 1724.76 - lr: 0.000012 - momentum: 0.000000 2023-10-17 13:33:11,859 epoch 8 - iter 693/773 - loss 0.00992381 - time (sec): 65.55 - samples/sec: 1705.82 - lr: 0.000012 - momentum: 0.000000 2023-10-17 13:33:19,708 epoch 8 - iter 770/773 - loss 0.01030405 - time (sec): 73.40 - samples/sec: 1688.12 - lr: 0.000011 - momentum: 0.000000 2023-10-17 13:33:20,005 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:33:20,005 EPOCH 8 done: loss 0.0104 - lr: 0.000011 2023-10-17 13:33:23,104 DEV : loss 0.112131267786026 - f1-score (micro avg) 0.7789 2023-10-17 13:33:23,141 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:33:30,765 epoch 9 - iter 77/773 - loss 0.00874412 - time (sec): 7.62 - samples/sec: 1611.31 - lr: 0.000011 - momentum: 0.000000 2023-10-17 13:33:38,513 epoch 9 - iter 154/773 - loss 0.00785734 - time (sec): 15.37 - samples/sec: 1621.77 - lr: 0.000010 - momentum: 0.000000 2023-10-17 13:33:46,462 epoch 9 - iter 231/773 - loss 0.00708825 - time (sec): 23.32 - samples/sec: 1644.66 - lr: 0.000009 - momentum: 0.000000 2023-10-17 13:33:54,388 epoch 9 - iter 308/773 - loss 0.00617174 - time (sec): 31.24 - samples/sec: 1655.71 - lr: 0.000009 - momentum: 0.000000 2023-10-17 13:34:02,210 epoch 9 - iter 385/773 - loss 0.00736950 - time (sec): 39.07 - samples/sec: 1594.91 - lr: 0.000008 - momentum: 0.000000 2023-10-17 13:34:09,935 epoch 9 - iter 462/773 - loss 0.00704324 - time (sec): 46.79 - samples/sec: 1584.76 - lr: 0.000008 - momentum: 0.000000 2023-10-17 13:34:17,149 epoch 9 - iter 539/773 - loss 0.00678870 - time (sec): 54.01 - samples/sec: 1599.87 - lr: 0.000007 - momentum: 0.000000 2023-10-17 13:34:24,344 epoch 9 - iter 616/773 - loss 0.00708850 - time (sec): 61.20 - samples/sec: 1608.61 - lr: 0.000007 - momentum: 0.000000 2023-10-17 13:34:31,470 epoch 9 - iter 693/773 - loss 0.00671587 - time (sec): 68.33 - samples/sec: 1616.86 - lr: 0.000006 - momentum: 0.000000 2023-10-17 13:34:39,709 epoch 9 - iter 770/773 - loss 0.00624570 - time (sec): 76.57 - samples/sec: 1615.20 - lr: 0.000006 - momentum: 0.000000 2023-10-17 13:34:40,011 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:34:40,012 EPOCH 9 done: loss 0.0062 - lr: 0.000006 2023-10-17 13:34:43,192 DEV : loss 0.1130354106426239 - f1-score (micro avg) 0.804 2023-10-17 13:34:43,224 saving best model 2023-10-17 13:34:44,723 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:34:52,444 epoch 10 - iter 77/773 - loss 0.00330740 - time (sec): 7.71 - samples/sec: 1678.74 - lr: 0.000005 - momentum: 0.000000 2023-10-17 13:35:00,074 epoch 10 - iter 154/773 - loss 0.00380506 - time (sec): 15.34 - samples/sec: 1652.24 - lr: 0.000005 - momentum: 0.000000 2023-10-17 13:35:07,781 epoch 10 - iter 231/773 - loss 0.00357509 - time (sec): 23.05 - samples/sec: 1681.37 - lr: 0.000004 - momentum: 0.000000 2023-10-17 13:35:15,374 epoch 10 - iter 308/773 - loss 0.00323943 - time (sec): 30.64 - samples/sec: 1665.77 - lr: 0.000003 - momentum: 0.000000 2023-10-17 13:35:23,448 epoch 10 - iter 385/773 - loss 0.00377474 - time (sec): 38.71 - samples/sec: 1632.78 - lr: 0.000003 - momentum: 0.000000 2023-10-17 13:35:31,104 epoch 10 - iter 462/773 - loss 0.00358345 - time (sec): 46.37 - samples/sec: 1635.38 - lr: 0.000002 - momentum: 0.000000 2023-10-17 13:35:39,051 epoch 10 - iter 539/773 - loss 0.00328378 - time (sec): 54.32 - samples/sec: 1615.54 - lr: 0.000002 - momentum: 0.000000 2023-10-17 13:35:46,804 epoch 10 - iter 616/773 - loss 0.00342164 - time (sec): 62.07 - samples/sec: 1596.22 - lr: 0.000001 - momentum: 0.000000 2023-10-17 13:35:54,614 epoch 10 - iter 693/773 - loss 0.00318027 - time (sec): 69.88 - samples/sec: 1588.59 - lr: 0.000001 - momentum: 0.000000 2023-10-17 13:36:02,805 epoch 10 - iter 770/773 - loss 0.00357272 - time (sec): 78.07 - samples/sec: 1585.49 - lr: 0.000000 - momentum: 0.000000 2023-10-17 13:36:03,113 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:36:03,113 EPOCH 10 done: loss 0.0036 - lr: 0.000000 2023-10-17 13:36:06,294 DEV : loss 0.11559101939201355 - f1-score (micro avg) 0.8032 2023-10-17 13:36:06,911 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:36:06,913 Loading model from best epoch ... 2023-10-17 13:36:09,226 SequenceTagger predicts: Dictionary with 13 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-BUILDING, B-BUILDING, E-BUILDING, I-BUILDING, S-STREET, B-STREET, E-STREET, I-STREET 2023-10-17 13:36:17,554 Results: - F-score (micro) 0.8142 - F-score (macro) 0.7437 - Accuracy 0.7039 By class: precision recall f1-score support LOC 0.8480 0.8668 0.8573 946 BUILDING 0.5990 0.6216 0.6101 185 STREET 0.7778 0.7500 0.7636 56 micro avg 0.8054 0.8231 0.8142 1187 macro avg 0.7416 0.7461 0.7437 1187 weighted avg 0.8059 0.8231 0.8143 1187 2023-10-17 13:36:17,554 ----------------------------------------------------------------------------------------------------