| 2023-10-15 19:15:51,162 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:15:51,163 Model: "SequenceTagger( | |
| (embeddings): TransformerWordEmbeddings( | |
| (model): BertModel( | |
| (embeddings): BertEmbeddings( | |
| (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): 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() | |
| )" | |
| 2023-10-15 19:15:51,163 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:15:51,163 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 | |
| 2023-10-15 19:15:51,163 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:15:51,163 Train: 20847 sentences | |
| 2023-10-15 19:15:51,163 (train_with_dev=False, train_with_test=False) | |
| 2023-10-15 19:15:51,163 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:15:51,163 Training Params: | |
| 2023-10-15 19:15:51,163 - learning_rate: "3e-05" | |
| 2023-10-15 19:15:51,163 - mini_batch_size: "8" | |
| 2023-10-15 19:15:51,163 - max_epochs: "10" | |
| 2023-10-15 19:15:51,163 - shuffle: "True" | |
| 2023-10-15 19:15:51,163 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:15:51,163 Plugins: | |
| 2023-10-15 19:15:51,163 - LinearScheduler | warmup_fraction: '0.1' | |
| 2023-10-15 19:15:51,163 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:15:51,163 Final evaluation on model from best epoch (best-model.pt) | |
| 2023-10-15 19:15:51,163 - metric: "('micro avg', 'f1-score')" | |
| 2023-10-15 19:15:51,164 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:15:51,164 Computation: | |
| 2023-10-15 19:15:51,164 - compute on device: cuda:0 | |
| 2023-10-15 19:15:51,164 - embedding storage: none | |
| 2023-10-15 19:15:51,164 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:15:51,164 Model training base path: "hmbench-newseye/de-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4" | |
| 2023-10-15 19:15:51,164 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:15:51,164 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:16:09,360 epoch 1 - iter 260/2606 - loss 1.92391321 - time (sec): 18.20 - samples/sec: 1919.99 - lr: 0.000003 - momentum: 0.000000 | |
| 2023-10-15 19:16:28,930 epoch 1 - iter 520/2606 - loss 1.17346399 - time (sec): 37.77 - samples/sec: 1945.08 - lr: 0.000006 - momentum: 0.000000 | |
| 2023-10-15 19:16:48,664 epoch 1 - iter 780/2606 - loss 0.89934868 - time (sec): 57.50 - samples/sec: 1891.11 - lr: 0.000009 - momentum: 0.000000 | |
| 2023-10-15 19:17:08,284 epoch 1 - iter 1040/2606 - loss 0.74351492 - time (sec): 77.12 - samples/sec: 1876.60 - lr: 0.000012 - momentum: 0.000000 | |
| 2023-10-15 19:17:28,718 epoch 1 - iter 1300/2606 - loss 0.64214251 - time (sec): 97.55 - samples/sec: 1879.58 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-15 19:17:47,276 epoch 1 - iter 1560/2606 - loss 0.57443185 - time (sec): 116.11 - samples/sec: 1881.20 - lr: 0.000018 - momentum: 0.000000 | |
| 2023-10-15 19:18:05,939 epoch 1 - iter 1820/2606 - loss 0.52041728 - time (sec): 134.77 - samples/sec: 1888.99 - lr: 0.000021 - momentum: 0.000000 | |
| 2023-10-15 19:18:25,928 epoch 1 - iter 2080/2606 - loss 0.47595359 - time (sec): 154.76 - samples/sec: 1883.05 - lr: 0.000024 - momentum: 0.000000 | |
| 2023-10-15 19:18:44,711 epoch 1 - iter 2340/2606 - loss 0.44512887 - time (sec): 173.55 - samples/sec: 1888.47 - lr: 0.000027 - momentum: 0.000000 | |
| 2023-10-15 19:19:04,604 epoch 1 - iter 2600/2606 - loss 0.41482047 - time (sec): 193.44 - samples/sec: 1895.69 - lr: 0.000030 - momentum: 0.000000 | |
| 2023-10-15 19:19:05,010 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:19:05,010 EPOCH 1 done: loss 0.4143 - lr: 0.000030 | |
| 2023-10-15 19:19:11,106 DEV : loss 0.10955972969532013 - f1-score (micro avg) 0.3005 | |
| 2023-10-15 19:19:11,131 saving best model | |
| 2023-10-15 19:19:11,522 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:19:30,718 epoch 2 - iter 260/2606 - loss 0.15574514 - time (sec): 19.19 - samples/sec: 1975.51 - lr: 0.000030 - momentum: 0.000000 | |
| 2023-10-15 19:19:50,545 epoch 2 - iter 520/2606 - loss 0.14461461 - time (sec): 39.02 - samples/sec: 1951.19 - lr: 0.000029 - momentum: 0.000000 | |
| 2023-10-15 19:20:10,592 epoch 2 - iter 780/2606 - loss 0.14719216 - time (sec): 59.07 - samples/sec: 1919.66 - lr: 0.000029 - momentum: 0.000000 | |
| 2023-10-15 19:20:29,932 epoch 2 - iter 1040/2606 - loss 0.14978370 - time (sec): 78.41 - samples/sec: 1886.12 - lr: 0.000029 - momentum: 0.000000 | |
| 2023-10-15 19:20:48,736 epoch 2 - iter 1300/2606 - loss 0.14928595 - time (sec): 97.21 - samples/sec: 1905.10 - lr: 0.000028 - momentum: 0.000000 | |
| 2023-10-15 19:21:07,741 epoch 2 - iter 1560/2606 - loss 0.14599580 - time (sec): 116.22 - samples/sec: 1900.04 - lr: 0.000028 - momentum: 0.000000 | |
| 2023-10-15 19:21:27,819 epoch 2 - iter 1820/2606 - loss 0.14527716 - time (sec): 136.30 - samples/sec: 1903.92 - lr: 0.000028 - momentum: 0.000000 | |
| 2023-10-15 19:21:45,993 epoch 2 - iter 2080/2606 - loss 0.14710466 - time (sec): 154.47 - samples/sec: 1906.17 - lr: 0.000027 - momentum: 0.000000 | |
| 2023-10-15 19:22:05,723 epoch 2 - iter 2340/2606 - loss 0.14702214 - time (sec): 174.20 - samples/sec: 1910.48 - lr: 0.000027 - momentum: 0.000000 | |
| 2023-10-15 19:22:23,853 epoch 2 - iter 2600/2606 - loss 0.14636417 - time (sec): 192.33 - samples/sec: 1908.65 - lr: 0.000027 - momentum: 0.000000 | |
| 2023-10-15 19:22:24,162 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:22:24,162 EPOCH 2 done: loss 0.1463 - lr: 0.000027 | |
| 2023-10-15 19:22:33,192 DEV : loss 0.1457057148218155 - f1-score (micro avg) 0.3846 | |
| 2023-10-15 19:22:33,219 saving best model | |
| 2023-10-15 19:22:33,803 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:22:52,494 epoch 3 - iter 260/2606 - loss 0.11346962 - time (sec): 18.69 - samples/sec: 1951.28 - lr: 0.000026 - momentum: 0.000000 | |
| 2023-10-15 19:23:11,261 epoch 3 - iter 520/2606 - loss 0.10225830 - time (sec): 37.46 - samples/sec: 1950.98 - lr: 0.000026 - momentum: 0.000000 | |
| 2023-10-15 19:23:30,606 epoch 3 - iter 780/2606 - loss 0.10240697 - time (sec): 56.80 - samples/sec: 1941.86 - lr: 0.000026 - momentum: 0.000000 | |
| 2023-10-15 19:23:50,015 epoch 3 - iter 1040/2606 - loss 0.10533913 - time (sec): 76.21 - samples/sec: 1946.08 - lr: 0.000025 - momentum: 0.000000 | |
| 2023-10-15 19:24:09,002 epoch 3 - iter 1300/2606 - loss 0.10238580 - time (sec): 95.20 - samples/sec: 1939.49 - lr: 0.000025 - momentum: 0.000000 | |
| 2023-10-15 19:24:27,332 epoch 3 - iter 1560/2606 - loss 0.10065496 - time (sec): 113.53 - samples/sec: 1939.49 - lr: 0.000025 - momentum: 0.000000 | |
| 2023-10-15 19:24:45,939 epoch 3 - iter 1820/2606 - loss 0.10154642 - time (sec): 132.13 - samples/sec: 1939.77 - lr: 0.000024 - momentum: 0.000000 | |
| 2023-10-15 19:25:05,809 epoch 3 - iter 2080/2606 - loss 0.10010787 - time (sec): 152.00 - samples/sec: 1942.38 - lr: 0.000024 - momentum: 0.000000 | |
| 2023-10-15 19:25:24,693 epoch 3 - iter 2340/2606 - loss 0.09902059 - time (sec): 170.89 - samples/sec: 1942.54 - lr: 0.000024 - momentum: 0.000000 | |
| 2023-10-15 19:25:43,245 epoch 3 - iter 2600/2606 - loss 0.09893031 - time (sec): 189.44 - samples/sec: 1935.23 - lr: 0.000023 - momentum: 0.000000 | |
| 2023-10-15 19:25:43,628 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:25:43,628 EPOCH 3 done: loss 0.0991 - lr: 0.000023 | |
| 2023-10-15 19:25:53,849 DEV : loss 0.15512152016162872 - f1-score (micro avg) 0.3475 | |
| 2023-10-15 19:25:53,878 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:26:13,527 epoch 4 - iter 260/2606 - loss 0.06739338 - time (sec): 19.65 - samples/sec: 1865.51 - lr: 0.000023 - momentum: 0.000000 | |
| 2023-10-15 19:26:31,967 epoch 4 - iter 520/2606 - loss 0.06849993 - time (sec): 38.09 - samples/sec: 1867.65 - lr: 0.000023 - momentum: 0.000000 | |
| 2023-10-15 19:26:50,197 epoch 4 - iter 780/2606 - loss 0.06889900 - time (sec): 56.32 - samples/sec: 1905.56 - lr: 0.000022 - momentum: 0.000000 | |
| 2023-10-15 19:27:09,613 epoch 4 - iter 1040/2606 - loss 0.06729654 - time (sec): 75.73 - samples/sec: 1904.96 - lr: 0.000022 - momentum: 0.000000 | |
| 2023-10-15 19:27:27,854 epoch 4 - iter 1300/2606 - loss 0.06648601 - time (sec): 93.97 - samples/sec: 1913.28 - lr: 0.000022 - momentum: 0.000000 | |
| 2023-10-15 19:27:46,186 epoch 4 - iter 1560/2606 - loss 0.06828801 - time (sec): 112.31 - samples/sec: 1919.17 - lr: 0.000021 - momentum: 0.000000 | |
| 2023-10-15 19:28:05,703 epoch 4 - iter 1820/2606 - loss 0.06782280 - time (sec): 131.82 - samples/sec: 1925.69 - lr: 0.000021 - momentum: 0.000000 | |
| 2023-10-15 19:28:24,290 epoch 4 - iter 2080/2606 - loss 0.06771520 - time (sec): 150.41 - samples/sec: 1921.25 - lr: 0.000021 - momentum: 0.000000 | |
| 2023-10-15 19:28:43,604 epoch 4 - iter 2340/2606 - loss 0.06777618 - time (sec): 169.72 - samples/sec: 1930.49 - lr: 0.000020 - momentum: 0.000000 | |
| 2023-10-15 19:29:04,044 epoch 4 - iter 2600/2606 - loss 0.06706676 - time (sec): 190.16 - samples/sec: 1927.32 - lr: 0.000020 - momentum: 0.000000 | |
| 2023-10-15 19:29:04,568 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:29:04,568 EPOCH 4 done: loss 0.0670 - lr: 0.000020 | |
| 2023-10-15 19:29:14,511 DEV : loss 0.2712627947330475 - f1-score (micro avg) 0.3677 | |
| 2023-10-15 19:29:14,540 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:29:33,612 epoch 5 - iter 260/2606 - loss 0.04149693 - time (sec): 19.07 - samples/sec: 1798.17 - lr: 0.000020 - momentum: 0.000000 | |
| 2023-10-15 19:29:52,714 epoch 5 - iter 520/2606 - loss 0.04775383 - time (sec): 38.17 - samples/sec: 1819.62 - lr: 0.000019 - momentum: 0.000000 | |
| 2023-10-15 19:30:11,638 epoch 5 - iter 780/2606 - loss 0.04929622 - time (sec): 57.10 - samples/sec: 1838.47 - lr: 0.000019 - momentum: 0.000000 | |
| 2023-10-15 19:30:30,777 epoch 5 - iter 1040/2606 - loss 0.05052347 - time (sec): 76.24 - samples/sec: 1864.80 - lr: 0.000019 - momentum: 0.000000 | |
| 2023-10-15 19:30:50,503 epoch 5 - iter 1300/2606 - loss 0.04823138 - time (sec): 95.96 - samples/sec: 1873.78 - lr: 0.000018 - momentum: 0.000000 | |
| 2023-10-15 19:31:09,329 epoch 5 - iter 1560/2606 - loss 0.04790977 - time (sec): 114.79 - samples/sec: 1874.31 - lr: 0.000018 - momentum: 0.000000 | |
| 2023-10-15 19:31:28,373 epoch 5 - iter 1820/2606 - loss 0.04792462 - time (sec): 133.83 - samples/sec: 1888.43 - lr: 0.000018 - momentum: 0.000000 | |
| 2023-10-15 19:31:48,692 epoch 5 - iter 2080/2606 - loss 0.04694655 - time (sec): 154.15 - samples/sec: 1888.54 - lr: 0.000017 - momentum: 0.000000 | |
| 2023-10-15 19:32:08,442 epoch 5 - iter 2340/2606 - loss 0.04649203 - time (sec): 173.90 - samples/sec: 1890.63 - lr: 0.000017 - momentum: 0.000000 | |
| 2023-10-15 19:32:28,539 epoch 5 - iter 2600/2606 - loss 0.04707216 - time (sec): 194.00 - samples/sec: 1889.42 - lr: 0.000017 - momentum: 0.000000 | |
| 2023-10-15 19:32:28,985 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:32:28,985 EPOCH 5 done: loss 0.0471 - lr: 0.000017 | |
| 2023-10-15 19:32:37,543 DEV : loss 0.3082272708415985 - f1-score (micro avg) 0.3575 | |
| 2023-10-15 19:32:37,573 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:32:58,397 epoch 6 - iter 260/2606 - loss 0.04629757 - time (sec): 20.82 - samples/sec: 1800.01 - lr: 0.000016 - momentum: 0.000000 | |
| 2023-10-15 19:33:18,141 epoch 6 - iter 520/2606 - loss 0.04081543 - time (sec): 40.57 - samples/sec: 1862.63 - lr: 0.000016 - momentum: 0.000000 | |
| 2023-10-15 19:33:37,203 epoch 6 - iter 780/2606 - loss 0.03989296 - time (sec): 59.63 - samples/sec: 1867.02 - lr: 0.000016 - momentum: 0.000000 | |
| 2023-10-15 19:33:57,457 epoch 6 - iter 1040/2606 - loss 0.03670245 - time (sec): 79.88 - samples/sec: 1884.93 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-15 19:34:16,388 epoch 6 - iter 1300/2606 - loss 0.03648850 - time (sec): 98.81 - samples/sec: 1889.84 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-15 19:34:35,186 epoch 6 - iter 1560/2606 - loss 0.03581498 - time (sec): 117.61 - samples/sec: 1894.20 - lr: 0.000015 - momentum: 0.000000 | |
| 2023-10-15 19:34:53,273 epoch 6 - iter 1820/2606 - loss 0.03655022 - time (sec): 135.70 - samples/sec: 1898.08 - lr: 0.000014 - momentum: 0.000000 | |
| 2023-10-15 19:35:12,704 epoch 6 - iter 2080/2606 - loss 0.03723489 - time (sec): 155.13 - samples/sec: 1888.04 - lr: 0.000014 - momentum: 0.000000 | |
| 2023-10-15 19:35:31,353 epoch 6 - iter 2340/2606 - loss 0.03749024 - time (sec): 173.78 - samples/sec: 1885.47 - lr: 0.000014 - momentum: 0.000000 | |
| 2023-10-15 19:35:51,127 epoch 6 - iter 2600/2606 - loss 0.03764428 - time (sec): 193.55 - samples/sec: 1891.36 - lr: 0.000013 - momentum: 0.000000 | |
| 2023-10-15 19:35:51,754 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:35:51,755 EPOCH 6 done: loss 0.0375 - lr: 0.000013 | |
| 2023-10-15 19:36:00,179 DEV : loss 0.3629515767097473 - f1-score (micro avg) 0.3784 | |
| 2023-10-15 19:36:00,211 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:36:19,967 epoch 7 - iter 260/2606 - loss 0.03047340 - time (sec): 19.75 - samples/sec: 1898.67 - lr: 0.000013 - momentum: 0.000000 | |
| 2023-10-15 19:36:40,131 epoch 7 - iter 520/2606 - loss 0.02575510 - time (sec): 39.92 - samples/sec: 1896.61 - lr: 0.000013 - momentum: 0.000000 | |
| 2023-10-15 19:36:59,611 epoch 7 - iter 780/2606 - loss 0.02827658 - time (sec): 59.40 - samples/sec: 1870.85 - lr: 0.000012 - momentum: 0.000000 | |
| 2023-10-15 19:37:19,040 epoch 7 - iter 1040/2606 - loss 0.02756663 - time (sec): 78.83 - samples/sec: 1842.08 - lr: 0.000012 - momentum: 0.000000 | |
| 2023-10-15 19:37:38,095 epoch 7 - iter 1300/2606 - loss 0.02790894 - time (sec): 97.88 - samples/sec: 1871.36 - lr: 0.000012 - momentum: 0.000000 | |
| 2023-10-15 19:37:58,031 epoch 7 - iter 1560/2606 - loss 0.02774421 - time (sec): 117.82 - samples/sec: 1852.99 - lr: 0.000011 - momentum: 0.000000 | |
| 2023-10-15 19:38:18,116 epoch 7 - iter 1820/2606 - loss 0.02704856 - time (sec): 137.90 - samples/sec: 1867.48 - lr: 0.000011 - momentum: 0.000000 | |
| 2023-10-15 19:38:36,158 epoch 7 - iter 2080/2606 - loss 0.02668550 - time (sec): 155.95 - samples/sec: 1871.85 - lr: 0.000011 - momentum: 0.000000 | |
| 2023-10-15 19:38:56,151 epoch 7 - iter 2340/2606 - loss 0.02647415 - time (sec): 175.94 - samples/sec: 1876.91 - lr: 0.000010 - momentum: 0.000000 | |
| 2023-10-15 19:39:14,749 epoch 7 - iter 2600/2606 - loss 0.02598872 - time (sec): 194.54 - samples/sec: 1883.74 - lr: 0.000010 - momentum: 0.000000 | |
| 2023-10-15 19:39:15,214 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:39:15,215 EPOCH 7 done: loss 0.0259 - lr: 0.000010 | |
| 2023-10-15 19:39:23,729 DEV : loss 0.4032082259654999 - f1-score (micro avg) 0.3755 | |
| 2023-10-15 19:39:23,763 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:39:42,293 epoch 8 - iter 260/2606 - loss 0.02130861 - time (sec): 18.53 - samples/sec: 1838.66 - lr: 0.000010 - momentum: 0.000000 | |
| 2023-10-15 19:40:02,040 epoch 8 - iter 520/2606 - loss 0.02419061 - time (sec): 38.28 - samples/sec: 1901.32 - lr: 0.000009 - momentum: 0.000000 | |
| 2023-10-15 19:40:22,033 epoch 8 - iter 780/2606 - loss 0.02294569 - time (sec): 58.27 - samples/sec: 1865.33 - lr: 0.000009 - momentum: 0.000000 | |
| 2023-10-15 19:40:42,622 epoch 8 - iter 1040/2606 - loss 0.02293506 - time (sec): 78.86 - samples/sec: 1835.03 - lr: 0.000009 - momentum: 0.000000 | |
| 2023-10-15 19:41:03,396 epoch 8 - iter 1300/2606 - loss 0.02188562 - time (sec): 99.63 - samples/sec: 1833.62 - lr: 0.000008 - momentum: 0.000000 | |
| 2023-10-15 19:41:24,998 epoch 8 - iter 1560/2606 - loss 0.02081983 - time (sec): 121.23 - samples/sec: 1818.18 - lr: 0.000008 - momentum: 0.000000 | |
| 2023-10-15 19:41:44,286 epoch 8 - iter 1820/2606 - loss 0.02030741 - time (sec): 140.52 - samples/sec: 1831.38 - lr: 0.000008 - momentum: 0.000000 | |
| 2023-10-15 19:42:04,008 epoch 8 - iter 2080/2606 - loss 0.02048971 - time (sec): 160.24 - samples/sec: 1830.46 - lr: 0.000007 - momentum: 0.000000 | |
| 2023-10-15 19:42:23,884 epoch 8 - iter 2340/2606 - loss 0.02009509 - time (sec): 180.12 - samples/sec: 1834.39 - lr: 0.000007 - momentum: 0.000000 | |
| 2023-10-15 19:42:42,745 epoch 8 - iter 2600/2606 - loss 0.02025086 - time (sec): 198.98 - samples/sec: 1842.00 - lr: 0.000007 - momentum: 0.000000 | |
| 2023-10-15 19:42:43,215 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:42:43,215 EPOCH 8 done: loss 0.0202 - lr: 0.000007 | |
| 2023-10-15 19:42:51,734 DEV : loss 0.43708279728889465 - f1-score (micro avg) 0.388 | |
| 2023-10-15 19:42:51,783 saving best model | |
| 2023-10-15 19:42:52,338 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:43:13,088 epoch 9 - iter 260/2606 - loss 0.00830062 - time (sec): 20.74 - samples/sec: 1913.67 - lr: 0.000006 - momentum: 0.000000 | |
| 2023-10-15 19:43:33,517 epoch 9 - iter 520/2606 - loss 0.01417015 - time (sec): 41.17 - samples/sec: 1893.78 - lr: 0.000006 - momentum: 0.000000 | |
| 2023-10-15 19:43:53,965 epoch 9 - iter 780/2606 - loss 0.01392241 - time (sec): 61.62 - samples/sec: 1872.93 - lr: 0.000006 - momentum: 0.000000 | |
| 2023-10-15 19:44:13,095 epoch 9 - iter 1040/2606 - loss 0.01304909 - time (sec): 80.75 - samples/sec: 1874.94 - lr: 0.000005 - momentum: 0.000000 | |
| 2023-10-15 19:44:32,003 epoch 9 - iter 1300/2606 - loss 0.01361230 - time (sec): 99.66 - samples/sec: 1885.50 - lr: 0.000005 - momentum: 0.000000 | |
| 2023-10-15 19:44:50,113 epoch 9 - iter 1560/2606 - loss 0.01364023 - time (sec): 117.77 - samples/sec: 1873.26 - lr: 0.000005 - momentum: 0.000000 | |
| 2023-10-15 19:45:09,412 epoch 9 - iter 1820/2606 - loss 0.01401743 - time (sec): 137.07 - samples/sec: 1879.93 - lr: 0.000004 - momentum: 0.000000 | |
| 2023-10-15 19:45:28,173 epoch 9 - iter 2080/2606 - loss 0.01378355 - time (sec): 155.83 - samples/sec: 1887.95 - lr: 0.000004 - momentum: 0.000000 | |
| 2023-10-15 19:45:47,780 epoch 9 - iter 2340/2606 - loss 0.01384268 - time (sec): 175.44 - samples/sec: 1883.86 - lr: 0.000004 - momentum: 0.000000 | |
| 2023-10-15 19:46:06,792 epoch 9 - iter 2600/2606 - loss 0.01416667 - time (sec): 194.45 - samples/sec: 1886.86 - lr: 0.000003 - momentum: 0.000000 | |
| 2023-10-15 19:46:07,137 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:46:07,137 EPOCH 9 done: loss 0.0142 - lr: 0.000003 | |
| 2023-10-15 19:46:16,060 DEV : loss 0.4454714357852936 - f1-score (micro avg) 0.3916 | |
| 2023-10-15 19:46:16,121 saving best model | |
| 2023-10-15 19:46:16,679 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:46:34,814 epoch 10 - iter 260/2606 - loss 0.01013364 - time (sec): 18.13 - samples/sec: 1956.91 - lr: 0.000003 - momentum: 0.000000 | |
| 2023-10-15 19:46:52,895 epoch 10 - iter 520/2606 - loss 0.01020078 - time (sec): 36.21 - samples/sec: 1912.88 - lr: 0.000003 - momentum: 0.000000 | |
| 2023-10-15 19:47:13,028 epoch 10 - iter 780/2606 - loss 0.00988399 - time (sec): 56.35 - samples/sec: 1889.53 - lr: 0.000002 - momentum: 0.000000 | |
| 2023-10-15 19:47:32,306 epoch 10 - iter 1040/2606 - loss 0.00932846 - time (sec): 75.63 - samples/sec: 1883.30 - lr: 0.000002 - momentum: 0.000000 | |
| 2023-10-15 19:47:51,475 epoch 10 - iter 1300/2606 - loss 0.00902175 - time (sec): 94.79 - samples/sec: 1883.35 - lr: 0.000002 - momentum: 0.000000 | |
| 2023-10-15 19:48:11,350 epoch 10 - iter 1560/2606 - loss 0.00978682 - time (sec): 114.67 - samples/sec: 1875.36 - lr: 0.000001 - momentum: 0.000000 | |
| 2023-10-15 19:48:31,274 epoch 10 - iter 1820/2606 - loss 0.00990480 - time (sec): 134.59 - samples/sec: 1873.23 - lr: 0.000001 - momentum: 0.000000 | |
| 2023-10-15 19:48:52,189 epoch 10 - iter 2080/2606 - loss 0.00957063 - time (sec): 155.51 - samples/sec: 1872.54 - lr: 0.000001 - momentum: 0.000000 | |
| 2023-10-15 19:49:12,022 epoch 10 - iter 2340/2606 - loss 0.00971093 - time (sec): 175.34 - samples/sec: 1864.56 - lr: 0.000000 - momentum: 0.000000 | |
| 2023-10-15 19:49:33,011 epoch 10 - iter 2600/2606 - loss 0.00974048 - time (sec): 196.33 - samples/sec: 1867.06 - lr: 0.000000 - momentum: 0.000000 | |
| 2023-10-15 19:49:33,453 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:49:33,453 EPOCH 10 done: loss 0.0097 - lr: 0.000000 | |
| 2023-10-15 19:49:43,629 DEV : loss 0.4776557981967926 - f1-score (micro avg) 0.3873 | |
| 2023-10-15 19:49:44,046 ---------------------------------------------------------------------------------------------------- | |
| 2023-10-15 19:49:44,047 Loading model from best epoch ... | |
| 2023-10-15 19:49:45,671 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 | |
| 2023-10-15 19:50:02,289 | |
| Results: | |
| - F-score (micro) 0.483 | |
| - F-score (macro) 0.3375 | |
| - Accuracy 0.3225 | |
| By class: | |
| precision recall f1-score support | |
| LOC 0.5216 0.5766 0.5477 1214 | |
| PER 0.4303 0.4851 0.4561 808 | |
| ORG 0.3243 0.3711 0.3461 353 | |
| HumanProd 0.0000 0.0000 0.0000 15 | |
| micro avg 0.4574 0.5117 0.4830 2390 | |
| macro avg 0.3190 0.3582 0.3375 2390 | |
| weighted avg 0.4583 0.5117 0.4835 2390 | |
| 2023-10-15 19:50:02,289 ---------------------------------------------------------------------------------------------------- | |