| 2023-10-25 11:00:55,891 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:00:55,892 Model: "SequenceTagger( | |
| (embeddings): TransformerWordEmbeddings( | |
| (model): BertModel( | |
| (embeddings): BertEmbeddings( | |
| (word_embeddings): Embedding(64001, 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): BertEncoder( | |
| (layer): ModuleList( | |
| (0-11): 12 x BertLayer( | |
| (attention): BertAttention( | |
| (self): BertSelfAttention( | |
| (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): BertSelfOutput( | |
| (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): BertIntermediate( | |
| (dense): Linear(in_features=768, out_features=3072, bias=True) | |
| (intermediate_act_fn): GELUActivation() | |
| ) | |
| (output): BertOutput( | |
| (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) | |
| ) | |
| ) | |
| ) | |
| ) | |
| (pooler): BertPooler( | |
| (dense): Linear(in_features=768, out_features=768, bias=True) | |
| (activation): Tanh() | |
| ) | |
| ) | |
| ) | |
| (locked_dropout): LockedDropout(p=0.5) | |
| (linear): Linear(in_features=768, out_features=17, bias=True) | |
| (loss_function): CrossEntropyLoss() | |
| )" | |
| 2023-10-25 11:00:55,892 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:00:55,892 MultiCorpus: 20847 train + 1123 dev + 3350 test sentences | |
| - NER_HIPE_2022 Corpus: 20847 train + 1123 dev + 3350 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/de/with_doc_seperator | |
| 2023-10-25 11:00:55,892 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:00:55,892 Train: 20847 sentences | |
| 2023-10-25 11:00:55,892 (train_with_dev=False, train_with_test=False) | |
| 2023-10-25 11:00:55,892 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:00:55,892 Training Params: | |
| 2023-10-25 11:00:55,892 - learning_rate: "5e-05" | |
| 2023-10-25 11:00:55,892 - mini_batch_size: "4" | |
| 2023-10-25 11:00:55,892 - max_epochs: "10" | |
| 2023-10-25 11:00:55,892 - shuffle: "True" | |
| 2023-10-25 11:00:55,892 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:00:55,892 Plugins: | |
| 2023-10-25 11:00:55,892 - TensorboardLogger | |
| 2023-10-25 11:00:55,892 - LinearScheduler | warmup_fraction: '0.1' | |
| 2023-10-25 11:00:55,892 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:00:55,892 Final evaluation on model from best epoch (best-model.pt) | |
| 2023-10-25 11:00:55,892 - metric: "('micro avg', 'f1-score')" | |
| 2023-10-25 11:00:55,892 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:00:55,892 Computation: | |
| 2023-10-25 11:00:55,892 - compute on device: cuda:0 | |
| 2023-10-25 11:00:55,892 - embedding storage: none | |
| 2023-10-25 11:00:55,893 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:00:55,893 Model training base path: "hmbench-newseye/de-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1" | |
| 2023-10-25 11:00:55,893 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:00:55,893 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:00:55,893 Logging anything other than scalars to TensorBoard is currently not supported. | |
| 2023-10-25 11:01:18,136 epoch 1 - iter 521/5212 - loss 1.37149992 - time (sec): 22.24 - samples/sec: 1721.34 - lr: 0.000005 - momentum: 0.000000 | |
| 2023-10-25 11:01:40,217 epoch 1 - iter 1042/5212 - loss 0.87935906 - time (sec): 44.32 - samples/sec: 1665.09 - lr: 0.000010 - momentum: 0.000000 | |
| 2023-10-25 11:02:01,925 epoch 1 - iter 1563/5212 - loss 0.68503997 - time (sec): 66.03 - samples/sec: 1643.34 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-25 11:02:23,692 epoch 1 - iter 2084/5212 - loss 0.58285105 - time (sec): 87.80 - samples/sec: 1629.58 - lr: 0.000020 - momentum: 0.000000 | |
| 2023-10-25 11:02:45,198 epoch 1 - iter 2605/5212 - loss 0.50956019 - time (sec): 109.30 - samples/sec: 1641.78 - lr: 0.000025 - momentum: 0.000000 | |
| 2023-10-25 11:03:07,147 epoch 1 - iter 3126/5212 - loss 0.46284092 - time (sec): 131.25 - samples/sec: 1652.16 - lr: 0.000030 - momentum: 0.000000 | |
| 2023-10-25 11:03:29,327 epoch 1 - iter 3647/5212 - loss 0.42868172 - time (sec): 153.43 - samples/sec: 1660.62 - lr: 0.000035 - momentum: 0.000000 | |
| 2023-10-25 11:03:51,454 epoch 1 - iter 4168/5212 - loss 0.40148351 - time (sec): 175.56 - samples/sec: 1666.12 - lr: 0.000040 - momentum: 0.000000 | |
| 2023-10-25 11:04:13,887 epoch 1 - iter 4689/5212 - loss 0.38468469 - time (sec): 197.99 - samples/sec: 1665.16 - lr: 0.000045 - momentum: 0.000000 | |
| 2023-10-25 11:04:36,107 epoch 1 - iter 5210/5212 - loss 0.36578248 - time (sec): 220.21 - samples/sec: 1668.00 - lr: 0.000050 - momentum: 0.000000 | |
| 2023-10-25 11:04:36,182 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:04:36,183 EPOCH 1 done: loss 0.3660 - lr: 0.000050 | |
| 2023-10-25 11:04:39,923 DEV : loss 0.13538858294487 - f1-score (micro avg) 0.3034 | |
| 2023-10-25 11:04:39,948 saving best model | |
| 2023-10-25 11:04:40,317 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:05:03,229 epoch 2 - iter 521/5212 - loss 0.20469904 - time (sec): 22.91 - samples/sec: 1536.72 - lr: 0.000049 - momentum: 0.000000 | |
| 2023-10-25 11:05:25,136 epoch 2 - iter 1042/5212 - loss 0.23803288 - time (sec): 44.82 - samples/sec: 1661.73 - lr: 0.000049 - momentum: 0.000000 | |
| 2023-10-25 11:05:47,474 epoch 2 - iter 1563/5212 - loss 0.23244269 - time (sec): 67.16 - samples/sec: 1670.90 - lr: 0.000048 - momentum: 0.000000 | |
| 2023-10-25 11:06:09,399 epoch 2 - iter 2084/5212 - loss 0.22959571 - time (sec): 89.08 - samples/sec: 1658.83 - lr: 0.000048 - momentum: 0.000000 | |
| 2023-10-25 11:06:31,093 epoch 2 - iter 2605/5212 - loss 0.22320080 - time (sec): 110.78 - samples/sec: 1645.05 - lr: 0.000047 - momentum: 0.000000 | |
| 2023-10-25 11:06:53,231 epoch 2 - iter 3126/5212 - loss 0.22319310 - time (sec): 132.91 - samples/sec: 1642.13 - lr: 0.000047 - momentum: 0.000000 | |
| 2023-10-25 11:07:15,062 epoch 2 - iter 3647/5212 - loss 0.21867469 - time (sec): 154.74 - samples/sec: 1640.78 - lr: 0.000046 - momentum: 0.000000 | |
| 2023-10-25 11:07:37,150 epoch 2 - iter 4168/5212 - loss 0.21838122 - time (sec): 176.83 - samples/sec: 1642.21 - lr: 0.000046 - momentum: 0.000000 | |
| 2023-10-25 11:07:59,564 epoch 2 - iter 4689/5212 - loss 0.22961072 - time (sec): 199.25 - samples/sec: 1652.99 - lr: 0.000045 - momentum: 0.000000 | |
| 2023-10-25 11:08:21,823 epoch 2 - iter 5210/5212 - loss 0.22659583 - time (sec): 221.50 - samples/sec: 1658.66 - lr: 0.000044 - momentum: 0.000000 | |
| 2023-10-25 11:08:21,903 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:08:21,903 EPOCH 2 done: loss 0.2266 - lr: 0.000044 | |
| 2023-10-25 11:08:28,788 DEV : loss 0.14999593794345856 - f1-score (micro avg) 0.317 | |
| 2023-10-25 11:08:28,813 saving best model | |
| 2023-10-25 11:08:29,311 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:08:51,082 epoch 3 - iter 521/5212 - loss 0.39731821 - time (sec): 21.77 - samples/sec: 1643.10 - lr: 0.000044 - momentum: 0.000000 | |
| 2023-10-25 11:09:13,112 epoch 3 - iter 1042/5212 - loss 0.33525516 - time (sec): 43.80 - samples/sec: 1660.21 - lr: 0.000043 - momentum: 0.000000 | |
| 2023-10-25 11:09:35,064 epoch 3 - iter 1563/5212 - loss 0.28290022 - time (sec): 65.75 - samples/sec: 1636.73 - lr: 0.000043 - momentum: 0.000000 | |
| 2023-10-25 11:09:56,744 epoch 3 - iter 2084/5212 - loss 0.25099625 - time (sec): 87.43 - samples/sec: 1654.98 - lr: 0.000042 - momentum: 0.000000 | |
| 2023-10-25 11:10:19,059 epoch 3 - iter 2605/5212 - loss 0.22758921 - time (sec): 109.74 - samples/sec: 1659.03 - lr: 0.000042 - momentum: 0.000000 | |
| 2023-10-25 11:10:40,793 epoch 3 - iter 3126/5212 - loss 0.22506705 - time (sec): 131.48 - samples/sec: 1660.35 - lr: 0.000041 - momentum: 0.000000 | |
| 2023-10-25 11:11:02,558 epoch 3 - iter 3647/5212 - loss 0.21926003 - time (sec): 153.24 - samples/sec: 1684.17 - lr: 0.000041 - momentum: 0.000000 | |
| 2023-10-25 11:11:24,327 epoch 3 - iter 4168/5212 - loss 0.21800940 - time (sec): 175.01 - samples/sec: 1674.58 - lr: 0.000040 - momentum: 0.000000 | |
| 2023-10-25 11:11:46,643 epoch 3 - iter 4689/5212 - loss 0.21324164 - time (sec): 197.33 - samples/sec: 1684.30 - lr: 0.000039 - momentum: 0.000000 | |
| 2023-10-25 11:12:08,682 epoch 3 - iter 5210/5212 - loss 0.21112728 - time (sec): 219.37 - samples/sec: 1674.21 - lr: 0.000039 - momentum: 0.000000 | |
| 2023-10-25 11:12:08,759 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:12:08,759 EPOCH 3 done: loss 0.2111 - lr: 0.000039 | |
| 2023-10-25 11:12:15,629 DEV : loss 0.1838080883026123 - f1-score (micro avg) 0.2752 | |
| 2023-10-25 11:12:15,654 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:12:37,555 epoch 4 - iter 521/5212 - loss 0.14551985 - time (sec): 21.90 - samples/sec: 1637.84 - lr: 0.000038 - momentum: 0.000000 | |
| 2023-10-25 11:12:59,454 epoch 4 - iter 1042/5212 - loss 0.14054748 - time (sec): 43.80 - samples/sec: 1676.75 - lr: 0.000038 - momentum: 0.000000 | |
| 2023-10-25 11:13:21,366 epoch 4 - iter 1563/5212 - loss 0.14476054 - time (sec): 65.71 - samples/sec: 1676.82 - lr: 0.000037 - momentum: 0.000000 | |
| 2023-10-25 11:13:43,227 epoch 4 - iter 2084/5212 - loss 0.14805005 - time (sec): 87.57 - samples/sec: 1685.25 - lr: 0.000037 - momentum: 0.000000 | |
| 2023-10-25 11:14:04,835 epoch 4 - iter 2605/5212 - loss 0.15829476 - time (sec): 109.18 - samples/sec: 1676.45 - lr: 0.000036 - momentum: 0.000000 | |
| 2023-10-25 11:14:26,663 epoch 4 - iter 3126/5212 - loss 0.15472574 - time (sec): 131.01 - samples/sec: 1686.89 - lr: 0.000036 - momentum: 0.000000 | |
| 2023-10-25 11:14:48,463 epoch 4 - iter 3647/5212 - loss 0.14772175 - time (sec): 152.81 - samples/sec: 1688.63 - lr: 0.000035 - momentum: 0.000000 | |
| 2023-10-25 11:15:10,375 epoch 4 - iter 4168/5212 - loss 0.14632759 - time (sec): 174.72 - samples/sec: 1676.75 - lr: 0.000034 - momentum: 0.000000 | |
| 2023-10-25 11:15:32,499 epoch 4 - iter 4689/5212 - loss 0.14404764 - time (sec): 196.84 - samples/sec: 1675.46 - lr: 0.000034 - momentum: 0.000000 | |
| 2023-10-25 11:15:54,778 epoch 4 - iter 5210/5212 - loss 0.14268477 - time (sec): 219.12 - samples/sec: 1676.29 - lr: 0.000033 - momentum: 0.000000 | |
| 2023-10-25 11:15:54,863 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:15:54,863 EPOCH 4 done: loss 0.1427 - lr: 0.000033 | |
| 2023-10-25 11:16:01,737 DEV : loss 0.22028455138206482 - f1-score (micro avg) 0.3096 | |
| 2023-10-25 11:16:01,763 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:16:23,704 epoch 5 - iter 521/5212 - loss 0.10320092 - time (sec): 21.94 - samples/sec: 1709.74 - lr: 0.000033 - momentum: 0.000000 | |
| 2023-10-25 11:16:45,397 epoch 5 - iter 1042/5212 - loss 0.10355668 - time (sec): 43.63 - samples/sec: 1662.39 - lr: 0.000032 - momentum: 0.000000 | |
| 2023-10-25 11:17:07,635 epoch 5 - iter 1563/5212 - loss 0.10540687 - time (sec): 65.87 - samples/sec: 1685.07 - lr: 0.000032 - momentum: 0.000000 | |
| 2023-10-25 11:17:29,500 epoch 5 - iter 2084/5212 - loss 0.10136807 - time (sec): 87.74 - samples/sec: 1702.49 - lr: 0.000031 - momentum: 0.000000 | |
| 2023-10-25 11:17:51,526 epoch 5 - iter 2605/5212 - loss 0.10353459 - time (sec): 109.76 - samples/sec: 1712.76 - lr: 0.000031 - momentum: 0.000000 | |
| 2023-10-25 11:18:13,450 epoch 5 - iter 3126/5212 - loss 0.10253579 - time (sec): 131.69 - samples/sec: 1700.34 - lr: 0.000030 - momentum: 0.000000 | |
| 2023-10-25 11:18:35,496 epoch 5 - iter 3647/5212 - loss 0.10477648 - time (sec): 153.73 - samples/sec: 1688.45 - lr: 0.000029 - momentum: 0.000000 | |
| 2023-10-25 11:18:57,342 epoch 5 - iter 4168/5212 - loss 0.10361175 - time (sec): 175.58 - samples/sec: 1697.83 - lr: 0.000029 - momentum: 0.000000 | |
| 2023-10-25 11:19:19,350 epoch 5 - iter 4689/5212 - loss 0.10359200 - time (sec): 197.59 - samples/sec: 1680.29 - lr: 0.000028 - momentum: 0.000000 | |
| 2023-10-25 11:19:41,597 epoch 5 - iter 5210/5212 - loss 0.10503616 - time (sec): 219.83 - samples/sec: 1671.08 - lr: 0.000028 - momentum: 0.000000 | |
| 2023-10-25 11:19:41,677 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:19:41,677 EPOCH 5 done: loss 0.1050 - lr: 0.000028 | |
| 2023-10-25 11:19:47,864 DEV : loss 0.22237569093704224 - f1-score (micro avg) 0.3813 | |
| 2023-10-25 11:19:47,891 saving best model | |
| 2023-10-25 11:19:48,294 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:20:10,627 epoch 6 - iter 521/5212 - loss 0.08012409 - time (sec): 22.33 - samples/sec: 1606.65 - lr: 0.000027 - momentum: 0.000000 | |
| 2023-10-25 11:20:33,460 epoch 6 - iter 1042/5212 - loss 0.07475817 - time (sec): 45.16 - samples/sec: 1652.26 - lr: 0.000027 - momentum: 0.000000 | |
| 2023-10-25 11:20:54,638 epoch 6 - iter 1563/5212 - loss 0.07771787 - time (sec): 66.34 - samples/sec: 1656.96 - lr: 0.000026 - momentum: 0.000000 | |
| 2023-10-25 11:21:16,836 epoch 6 - iter 2084/5212 - loss 0.07455376 - time (sec): 88.54 - samples/sec: 1675.11 - lr: 0.000026 - momentum: 0.000000 | |
| 2023-10-25 11:21:38,932 epoch 6 - iter 2605/5212 - loss 0.07609439 - time (sec): 110.64 - samples/sec: 1681.84 - lr: 0.000025 - momentum: 0.000000 | |
| 2023-10-25 11:22:01,190 epoch 6 - iter 3126/5212 - loss 0.07530603 - time (sec): 132.89 - samples/sec: 1673.73 - lr: 0.000024 - momentum: 0.000000 | |
| 2023-10-25 11:22:23,343 epoch 6 - iter 3647/5212 - loss 0.07426848 - time (sec): 155.05 - samples/sec: 1691.85 - lr: 0.000024 - momentum: 0.000000 | |
| 2023-10-25 11:22:45,050 epoch 6 - iter 4168/5212 - loss 0.07447383 - time (sec): 176.75 - samples/sec: 1694.20 - lr: 0.000023 - momentum: 0.000000 | |
| 2023-10-25 11:23:06,759 epoch 6 - iter 4689/5212 - loss 0.07623639 - time (sec): 198.46 - samples/sec: 1675.58 - lr: 0.000023 - momentum: 0.000000 | |
| 2023-10-25 11:23:28,556 epoch 6 - iter 5210/5212 - loss 0.07631310 - time (sec): 220.26 - samples/sec: 1667.59 - lr: 0.000022 - momentum: 0.000000 | |
| 2023-10-25 11:23:28,640 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:23:28,640 EPOCH 6 done: loss 0.0763 - lr: 0.000022 | |
| 2023-10-25 11:23:34,855 DEV : loss 0.32064488530158997 - f1-score (micro avg) 0.3633 | |
| 2023-10-25 11:23:34,881 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:23:56,935 epoch 7 - iter 521/5212 - loss 0.05039789 - time (sec): 22.05 - samples/sec: 1871.83 - lr: 0.000022 - momentum: 0.000000 | |
| 2023-10-25 11:24:19,042 epoch 7 - iter 1042/5212 - loss 0.05944987 - time (sec): 44.16 - samples/sec: 1698.53 - lr: 0.000021 - momentum: 0.000000 | |
| 2023-10-25 11:24:40,988 epoch 7 - iter 1563/5212 - loss 0.06349842 - time (sec): 66.11 - samples/sec: 1651.93 - lr: 0.000021 - momentum: 0.000000 | |
| 2023-10-25 11:25:03,426 epoch 7 - iter 2084/5212 - loss 0.06688293 - time (sec): 88.54 - samples/sec: 1648.17 - lr: 0.000020 - momentum: 0.000000 | |
| 2023-10-25 11:25:24,892 epoch 7 - iter 2605/5212 - loss 0.06760611 - time (sec): 110.01 - samples/sec: 1677.77 - lr: 0.000019 - momentum: 0.000000 | |
| 2023-10-25 11:25:46,879 epoch 7 - iter 3126/5212 - loss 0.06775305 - time (sec): 132.00 - samples/sec: 1672.32 - lr: 0.000019 - momentum: 0.000000 | |
| 2023-10-25 11:26:08,888 epoch 7 - iter 3647/5212 - loss 0.06912261 - time (sec): 154.01 - samples/sec: 1664.05 - lr: 0.000018 - momentum: 0.000000 | |
| 2023-10-25 11:26:31,107 epoch 7 - iter 4168/5212 - loss 0.06908616 - time (sec): 176.22 - samples/sec: 1655.61 - lr: 0.000018 - momentum: 0.000000 | |
| 2023-10-25 11:26:53,084 epoch 7 - iter 4689/5212 - loss 0.07148833 - time (sec): 198.20 - samples/sec: 1663.87 - lr: 0.000017 - momentum: 0.000000 | |
| 2023-10-25 11:27:15,291 epoch 7 - iter 5210/5212 - loss 0.06952026 - time (sec): 220.41 - samples/sec: 1663.04 - lr: 0.000017 - momentum: 0.000000 | |
| 2023-10-25 11:27:15,436 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:27:15,437 EPOCH 7 done: loss 0.0694 - lr: 0.000017 | |
| 2023-10-25 11:27:21,672 DEV : loss 0.3756372332572937 - f1-score (micro avg) 0.3489 | |
| 2023-10-25 11:27:21,697 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:27:43,233 epoch 8 - iter 521/5212 - loss 0.05879330 - time (sec): 21.53 - samples/sec: 1859.19 - lr: 0.000016 - momentum: 0.000000 | |
| 2023-10-25 11:28:04,959 epoch 8 - iter 1042/5212 - loss 0.05539014 - time (sec): 43.26 - samples/sec: 1815.53 - lr: 0.000016 - momentum: 0.000000 | |
| 2023-10-25 11:28:27,149 epoch 8 - iter 1563/5212 - loss 0.05646244 - time (sec): 65.45 - samples/sec: 1747.62 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-25 11:28:49,551 epoch 8 - iter 2084/5212 - loss 0.05287024 - time (sec): 87.85 - samples/sec: 1742.21 - lr: 0.000014 - momentum: 0.000000 | |
| 2023-10-25 11:29:11,741 epoch 8 - iter 2605/5212 - loss 0.05325825 - time (sec): 110.04 - samples/sec: 1728.37 - lr: 0.000014 - momentum: 0.000000 | |
| 2023-10-25 11:29:33,499 epoch 8 - iter 3126/5212 - loss 0.05326024 - time (sec): 131.80 - samples/sec: 1708.19 - lr: 0.000013 - momentum: 0.000000 | |
| 2023-10-25 11:29:55,509 epoch 8 - iter 3647/5212 - loss 0.05476219 - time (sec): 153.81 - samples/sec: 1685.49 - lr: 0.000013 - momentum: 0.000000 | |
| 2023-10-25 11:30:18,302 epoch 8 - iter 4168/5212 - loss 0.05479079 - time (sec): 176.60 - samples/sec: 1679.91 - lr: 0.000012 - momentum: 0.000000 | |
| 2023-10-25 11:30:40,251 epoch 8 - iter 4689/5212 - loss 0.05530734 - time (sec): 198.55 - samples/sec: 1675.48 - lr: 0.000012 - momentum: 0.000000 | |
| 2023-10-25 11:31:02,189 epoch 8 - iter 5210/5212 - loss 0.05608379 - time (sec): 220.49 - samples/sec: 1666.00 - lr: 0.000011 - momentum: 0.000000 | |
| 2023-10-25 11:31:02,275 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:31:02,275 EPOCH 8 done: loss 0.0561 - lr: 0.000011 | |
| 2023-10-25 11:31:08,474 DEV : loss 0.3516038954257965 - f1-score (micro avg) 0.3603 | |
| 2023-10-25 11:31:08,500 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:31:30,381 epoch 9 - iter 521/5212 - loss 0.03414719 - time (sec): 21.88 - samples/sec: 1677.12 - lr: 0.000011 - momentum: 0.000000 | |
| 2023-10-25 11:31:51,575 epoch 9 - iter 1042/5212 - loss 0.03394267 - time (sec): 43.07 - samples/sec: 1687.17 - lr: 0.000010 - momentum: 0.000000 | |
| 2023-10-25 11:32:12,995 epoch 9 - iter 1563/5212 - loss 0.03753554 - time (sec): 64.49 - samples/sec: 1713.27 - lr: 0.000009 - momentum: 0.000000 | |
| 2023-10-25 11:32:34,589 epoch 9 - iter 2084/5212 - loss 0.03899856 - time (sec): 86.09 - samples/sec: 1732.76 - lr: 0.000009 - momentum: 0.000000 | |
| 2023-10-25 11:32:56,245 epoch 9 - iter 2605/5212 - loss 0.03908002 - time (sec): 107.74 - samples/sec: 1716.66 - lr: 0.000008 - momentum: 0.000000 | |
| 2023-10-25 11:33:19,017 epoch 9 - iter 3126/5212 - loss 0.03898247 - time (sec): 130.52 - samples/sec: 1697.62 - lr: 0.000008 - momentum: 0.000000 | |
| 2023-10-25 11:33:40,952 epoch 9 - iter 3647/5212 - loss 0.03853540 - time (sec): 152.45 - samples/sec: 1688.78 - lr: 0.000007 - momentum: 0.000000 | |
| 2023-10-25 11:34:02,644 epoch 9 - iter 4168/5212 - loss 0.03902025 - time (sec): 174.14 - samples/sec: 1686.38 - lr: 0.000007 - momentum: 0.000000 | |
| 2023-10-25 11:34:24,348 epoch 9 - iter 4689/5212 - loss 0.03912183 - time (sec): 195.85 - samples/sec: 1685.83 - lr: 0.000006 - momentum: 0.000000 | |
| 2023-10-25 11:34:46,370 epoch 9 - iter 5210/5212 - loss 0.03907876 - time (sec): 217.87 - samples/sec: 1686.24 - lr: 0.000006 - momentum: 0.000000 | |
| 2023-10-25 11:34:46,460 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:34:46,460 EPOCH 9 done: loss 0.0391 - lr: 0.000006 | |
| 2023-10-25 11:34:53,392 DEV : loss 0.405927836894989 - f1-score (micro avg) 0.3634 | |
| 2023-10-25 11:34:53,418 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:35:14,984 epoch 10 - iter 521/5212 - loss 0.03118015 - time (sec): 21.56 - samples/sec: 1775.07 - lr: 0.000005 - momentum: 0.000000 | |
| 2023-10-25 11:35:37,344 epoch 10 - iter 1042/5212 - loss 0.02868530 - time (sec): 43.93 - samples/sec: 1717.36 - lr: 0.000004 - momentum: 0.000000 | |
| 2023-10-25 11:35:59,555 epoch 10 - iter 1563/5212 - loss 0.02897899 - time (sec): 66.14 - samples/sec: 1698.08 - lr: 0.000004 - momentum: 0.000000 | |
| 2023-10-25 11:36:21,701 epoch 10 - iter 2084/5212 - loss 0.02993813 - time (sec): 88.28 - samples/sec: 1686.02 - lr: 0.000003 - momentum: 0.000000 | |
| 2023-10-25 11:36:43,697 epoch 10 - iter 2605/5212 - loss 0.03083416 - time (sec): 110.28 - samples/sec: 1686.24 - lr: 0.000003 - momentum: 0.000000 | |
| 2023-10-25 11:37:05,594 epoch 10 - iter 3126/5212 - loss 0.03010510 - time (sec): 132.18 - samples/sec: 1688.94 - lr: 0.000002 - momentum: 0.000000 | |
| 2023-10-25 11:37:27,487 epoch 10 - iter 3647/5212 - loss 0.02911948 - time (sec): 154.07 - samples/sec: 1675.51 - lr: 0.000002 - momentum: 0.000000 | |
| 2023-10-25 11:37:49,663 epoch 10 - iter 4168/5212 - loss 0.02885642 - time (sec): 176.24 - samples/sec: 1661.67 - lr: 0.000001 - momentum: 0.000000 | |
| 2023-10-25 11:38:11,733 epoch 10 - iter 4689/5212 - loss 0.02769243 - time (sec): 198.31 - samples/sec: 1660.60 - lr: 0.000001 - momentum: 0.000000 | |
| 2023-10-25 11:38:33,789 epoch 10 - iter 5210/5212 - loss 0.02725114 - time (sec): 220.37 - samples/sec: 1667.00 - lr: 0.000000 - momentum: 0.000000 | |
| 2023-10-25 11:38:33,876 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:38:33,876 EPOCH 10 done: loss 0.0272 - lr: 0.000000 | |
| 2023-10-25 11:38:40,758 DEV : loss 0.41593435406684875 - f1-score (micro avg) 0.368 | |
| 2023-10-25 11:38:41,135 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-25 11:38:41,136 Loading model from best epoch ... | |
| 2023-10-25 11:38:42,646 SequenceTagger predicts: Dictionary with 17 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd | |
| 2023-10-25 11:38:52,346 | |
| Results: | |
| - F-score (micro) 0.3457 | |
| - F-score (macro) 0.225 | |
| - Accuracy 0.211 | |
| By class: | |
| precision recall f1-score support | |
| LOC 0.5061 0.3443 0.4098 1214 | |
| PER 0.3730 0.3106 0.3390 808 | |
| ORG 0.1860 0.1275 0.1513 353 | |
| HumanProd 0.0000 0.0000 0.0000 15 | |
| micro avg 0.4101 0.2987 0.3457 2390 | |
| macro avg 0.2662 0.1956 0.2250 2390 | |
| weighted avg 0.4106 0.2987 0.3451 2390 | |
| 2023-10-25 11:38:52,346 ---------------------------------------------------------------------------------------------------- | |