2023-10-16 21:44:09,762 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:44:09,763 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=13, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-16 21:44:09,763 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:44:09,764 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 2023-10-16 21:44:09,764 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:44:09,764 Train: 6183 sentences 2023-10-16 21:44:09,764 (train_with_dev=False, train_with_test=False) 2023-10-16 21:44:09,764 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:44:09,764 Training Params: 2023-10-16 21:44:09,764 - learning_rate: "5e-05" 2023-10-16 21:44:09,764 - mini_batch_size: "4" 2023-10-16 21:44:09,764 - max_epochs: "10" 2023-10-16 21:44:09,764 - shuffle: "True" 2023-10-16 21:44:09,764 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:44:09,764 Plugins: 2023-10-16 21:44:09,764 - LinearScheduler | warmup_fraction: '0.1' 2023-10-16 21:44:09,764 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:44:09,764 Final evaluation on model from best epoch (best-model.pt) 2023-10-16 21:44:09,764 - metric: "('micro avg', 'f1-score')" 2023-10-16 21:44:09,764 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:44:09,764 Computation: 2023-10-16 21:44:09,764 - compute on device: cuda:0 2023-10-16 21:44:09,764 - embedding storage: none 2023-10-16 21:44:09,764 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:44:09,764 Model training base path: "hmbench-topres19th/en-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2" 2023-10-16 21:44:09,764 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:44:09,764 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:44:16,668 epoch 1 - iter 154/1546 - loss 1.65007029 - time (sec): 6.90 - samples/sec: 1715.81 - lr: 0.000005 - momentum: 0.000000 2023-10-16 21:44:23,611 epoch 1 - iter 308/1546 - loss 0.94129662 - time (sec): 13.85 - samples/sec: 1714.44 - lr: 0.000010 - momentum: 0.000000 2023-10-16 21:44:30,625 epoch 1 - iter 462/1546 - loss 0.64763640 - time (sec): 20.86 - samples/sec: 1777.67 - lr: 0.000015 - momentum: 0.000000 2023-10-16 21:44:37,494 epoch 1 - iter 616/1546 - loss 0.52517050 - time (sec): 27.73 - samples/sec: 1772.61 - lr: 0.000020 - momentum: 0.000000 2023-10-16 21:44:44,367 epoch 1 - iter 770/1546 - loss 0.44714948 - time (sec): 34.60 - samples/sec: 1761.60 - lr: 0.000025 - momentum: 0.000000 2023-10-16 21:44:51,242 epoch 1 - iter 924/1546 - loss 0.39140356 - time (sec): 41.48 - samples/sec: 1762.58 - lr: 0.000030 - momentum: 0.000000 2023-10-16 21:44:58,147 epoch 1 - iter 1078/1546 - loss 0.35053117 - time (sec): 48.38 - samples/sec: 1776.97 - lr: 0.000035 - momentum: 0.000000 2023-10-16 21:45:05,162 epoch 1 - iter 1232/1546 - loss 0.32273432 - time (sec): 55.40 - samples/sec: 1793.05 - lr: 0.000040 - momentum: 0.000000 2023-10-16 21:45:12,237 epoch 1 - iter 1386/1546 - loss 0.29966657 - time (sec): 62.47 - samples/sec: 1779.42 - lr: 0.000045 - momentum: 0.000000 2023-10-16 21:45:19,081 epoch 1 - iter 1540/1546 - loss 0.28039720 - time (sec): 69.32 - samples/sec: 1787.73 - lr: 0.000050 - momentum: 0.000000 2023-10-16 21:45:19,336 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:45:19,336 EPOCH 1 done: loss 0.2799 - lr: 0.000050 2023-10-16 21:45:21,072 DEV : loss 0.0895804762840271 - f1-score (micro avg) 0.637 2023-10-16 21:45:21,095 saving best model 2023-10-16 21:45:21,441 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:45:28,278 epoch 2 - iter 154/1546 - loss 0.09243588 - time (sec): 6.84 - samples/sec: 1772.05 - lr: 0.000049 - momentum: 0.000000 2023-10-16 21:45:35,087 epoch 2 - iter 308/1546 - loss 0.10563878 - time (sec): 13.65 - samples/sec: 1755.06 - lr: 0.000049 - momentum: 0.000000 2023-10-16 21:45:42,042 epoch 2 - iter 462/1546 - loss 0.10498928 - time (sec): 20.60 - samples/sec: 1786.65 - lr: 0.000048 - momentum: 0.000000 2023-10-16 21:45:48,905 epoch 2 - iter 616/1546 - loss 0.10253956 - time (sec): 27.46 - samples/sec: 1779.24 - lr: 0.000048 - momentum: 0.000000 2023-10-16 21:45:55,770 epoch 2 - iter 770/1546 - loss 0.10307917 - time (sec): 34.33 - samples/sec: 1793.99 - lr: 0.000047 - momentum: 0.000000 2023-10-16 21:46:02,644 epoch 2 - iter 924/1546 - loss 0.10366383 - time (sec): 41.20 - samples/sec: 1775.94 - lr: 0.000047 - momentum: 0.000000 2023-10-16 21:46:09,503 epoch 2 - iter 1078/1546 - loss 0.10275984 - time (sec): 48.06 - samples/sec: 1780.26 - lr: 0.000046 - momentum: 0.000000 2023-10-16 21:46:16,416 epoch 2 - iter 1232/1546 - loss 0.10145230 - time (sec): 54.97 - samples/sec: 1779.20 - lr: 0.000046 - momentum: 0.000000 2023-10-16 21:46:23,230 epoch 2 - iter 1386/1546 - loss 0.09850958 - time (sec): 61.79 - samples/sec: 1784.50 - lr: 0.000045 - momentum: 0.000000 2023-10-16 21:46:30,555 epoch 2 - iter 1540/1546 - loss 0.09639554 - time (sec): 69.11 - samples/sec: 1791.82 - lr: 0.000044 - momentum: 0.000000 2023-10-16 21:46:30,812 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:46:30,812 EPOCH 2 done: loss 0.0961 - lr: 0.000044 2023-10-16 21:46:32,814 DEV : loss 0.0685592070221901 - f1-score (micro avg) 0.7696 2023-10-16 21:46:32,827 saving best model 2023-10-16 21:46:33,306 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:46:40,223 epoch 3 - iter 154/1546 - loss 0.06577964 - time (sec): 6.91 - samples/sec: 1853.17 - lr: 0.000044 - momentum: 0.000000 2023-10-16 21:46:46,999 epoch 3 - iter 308/1546 - loss 0.06885278 - time (sec): 13.69 - samples/sec: 1832.42 - lr: 0.000043 - momentum: 0.000000 2023-10-16 21:46:53,930 epoch 3 - iter 462/1546 - loss 0.07613517 - time (sec): 20.62 - samples/sec: 1828.52 - lr: 0.000043 - momentum: 0.000000 2023-10-16 21:47:00,686 epoch 3 - iter 616/1546 - loss 0.07339670 - time (sec): 27.38 - samples/sec: 1822.81 - lr: 0.000042 - momentum: 0.000000 2023-10-16 21:47:07,536 epoch 3 - iter 770/1546 - loss 0.07279666 - time (sec): 34.23 - samples/sec: 1827.46 - lr: 0.000042 - momentum: 0.000000 2023-10-16 21:47:14,494 epoch 3 - iter 924/1546 - loss 0.07159335 - time (sec): 41.19 - samples/sec: 1818.28 - lr: 0.000041 - momentum: 0.000000 2023-10-16 21:47:21,360 epoch 3 - iter 1078/1546 - loss 0.06968918 - time (sec): 48.05 - samples/sec: 1814.87 - lr: 0.000041 - momentum: 0.000000 2023-10-16 21:47:28,153 epoch 3 - iter 1232/1546 - loss 0.06933368 - time (sec): 54.84 - samples/sec: 1802.51 - lr: 0.000040 - momentum: 0.000000 2023-10-16 21:47:35,020 epoch 3 - iter 1386/1546 - loss 0.06859729 - time (sec): 61.71 - samples/sec: 1812.58 - lr: 0.000039 - momentum: 0.000000 2023-10-16 21:47:41,777 epoch 3 - iter 1540/1546 - loss 0.06849329 - time (sec): 68.47 - samples/sec: 1810.30 - lr: 0.000039 - momentum: 0.000000 2023-10-16 21:47:42,035 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:47:42,035 EPOCH 3 done: loss 0.0685 - lr: 0.000039 2023-10-16 21:47:44,051 DEV : loss 0.08517798036336899 - f1-score (micro avg) 0.7447 2023-10-16 21:47:44,063 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:47:51,034 epoch 4 - iter 154/1546 - loss 0.05538366 - time (sec): 6.97 - samples/sec: 1785.10 - lr: 0.000038 - momentum: 0.000000 2023-10-16 21:47:57,870 epoch 4 - iter 308/1546 - loss 0.05887323 - time (sec): 13.81 - samples/sec: 1724.84 - lr: 0.000038 - momentum: 0.000000 2023-10-16 21:48:04,737 epoch 4 - iter 462/1546 - loss 0.05491630 - time (sec): 20.67 - samples/sec: 1741.61 - lr: 0.000037 - momentum: 0.000000 2023-10-16 21:48:11,492 epoch 4 - iter 616/1546 - loss 0.05243751 - time (sec): 27.43 - samples/sec: 1758.25 - lr: 0.000037 - momentum: 0.000000 2023-10-16 21:48:18,347 epoch 4 - iter 770/1546 - loss 0.05291705 - time (sec): 34.28 - samples/sec: 1778.29 - lr: 0.000036 - momentum: 0.000000 2023-10-16 21:48:25,115 epoch 4 - iter 924/1546 - loss 0.05316064 - time (sec): 41.05 - samples/sec: 1782.75 - lr: 0.000036 - momentum: 0.000000 2023-10-16 21:48:31,936 epoch 4 - iter 1078/1546 - loss 0.05226355 - time (sec): 47.87 - samples/sec: 1787.50 - lr: 0.000035 - momentum: 0.000000 2023-10-16 21:48:38,834 epoch 4 - iter 1232/1546 - loss 0.05147292 - time (sec): 54.77 - samples/sec: 1793.52 - lr: 0.000034 - momentum: 0.000000 2023-10-16 21:48:45,749 epoch 4 - iter 1386/1546 - loss 0.04971427 - time (sec): 61.68 - samples/sec: 1797.52 - lr: 0.000034 - momentum: 0.000000 2023-10-16 21:48:52,720 epoch 4 - iter 1540/1546 - loss 0.04975106 - time (sec): 68.66 - samples/sec: 1804.48 - lr: 0.000033 - momentum: 0.000000 2023-10-16 21:48:52,978 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:48:52,978 EPOCH 4 done: loss 0.0497 - lr: 0.000033 2023-10-16 21:48:55,028 DEV : loss 0.09497705101966858 - f1-score (micro avg) 0.7821 2023-10-16 21:48:55,040 saving best model 2023-10-16 21:48:55,509 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:49:02,349 epoch 5 - iter 154/1546 - loss 0.01734854 - time (sec): 6.83 - samples/sec: 1839.45 - lr: 0.000033 - momentum: 0.000000 2023-10-16 21:49:09,103 epoch 5 - iter 308/1546 - loss 0.02591889 - time (sec): 13.58 - samples/sec: 1813.31 - lr: 0.000032 - momentum: 0.000000 2023-10-16 21:49:15,969 epoch 5 - iter 462/1546 - loss 0.03215375 - time (sec): 20.45 - samples/sec: 1762.51 - lr: 0.000032 - momentum: 0.000000 2023-10-16 21:49:22,877 epoch 5 - iter 616/1546 - loss 0.03561342 - time (sec): 27.36 - samples/sec: 1798.27 - lr: 0.000031 - momentum: 0.000000 2023-10-16 21:49:29,717 epoch 5 - iter 770/1546 - loss 0.04014613 - time (sec): 34.20 - samples/sec: 1807.31 - lr: 0.000031 - momentum: 0.000000 2023-10-16 21:49:36,641 epoch 5 - iter 924/1546 - loss 0.04102622 - time (sec): 41.12 - samples/sec: 1818.94 - lr: 0.000030 - momentum: 0.000000 2023-10-16 21:49:43,561 epoch 5 - iter 1078/1546 - loss 0.03999871 - time (sec): 48.04 - samples/sec: 1820.62 - lr: 0.000029 - momentum: 0.000000 2023-10-16 21:49:50,438 epoch 5 - iter 1232/1546 - loss 0.04053831 - time (sec): 54.92 - samples/sec: 1816.84 - lr: 0.000029 - momentum: 0.000000 2023-10-16 21:49:57,266 epoch 5 - iter 1386/1546 - loss 0.04007914 - time (sec): 61.75 - samples/sec: 1811.89 - lr: 0.000028 - momentum: 0.000000 2023-10-16 21:50:04,082 epoch 5 - iter 1540/1546 - loss 0.03801735 - time (sec): 68.56 - samples/sec: 1807.74 - lr: 0.000028 - momentum: 0.000000 2023-10-16 21:50:04,338 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:50:04,338 EPOCH 5 done: loss 0.0380 - lr: 0.000028 2023-10-16 21:50:06,356 DEV : loss 0.09318046271800995 - f1-score (micro avg) 0.7782 2023-10-16 21:50:06,368 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:50:13,194 epoch 6 - iter 154/1546 - loss 0.02556469 - time (sec): 6.82 - samples/sec: 1809.18 - lr: 0.000027 - momentum: 0.000000 2023-10-16 21:50:20,094 epoch 6 - iter 308/1546 - loss 0.03097773 - time (sec): 13.72 - samples/sec: 1736.11 - lr: 0.000027 - momentum: 0.000000 2023-10-16 21:50:26,958 epoch 6 - iter 462/1546 - loss 0.02965652 - time (sec): 20.59 - samples/sec: 1746.62 - lr: 0.000026 - momentum: 0.000000 2023-10-16 21:50:33,830 epoch 6 - iter 616/1546 - loss 0.02790282 - time (sec): 27.46 - samples/sec: 1771.83 - lr: 0.000026 - momentum: 0.000000 2023-10-16 21:50:40,624 epoch 6 - iter 770/1546 - loss 0.02749592 - time (sec): 34.25 - samples/sec: 1771.92 - lr: 0.000025 - momentum: 0.000000 2023-10-16 21:50:47,433 epoch 6 - iter 924/1546 - loss 0.02610977 - time (sec): 41.06 - samples/sec: 1768.73 - lr: 0.000024 - momentum: 0.000000 2023-10-16 21:50:54,284 epoch 6 - iter 1078/1546 - loss 0.02639060 - time (sec): 47.91 - samples/sec: 1772.33 - lr: 0.000024 - momentum: 0.000000 2023-10-16 21:51:01,306 epoch 6 - iter 1232/1546 - loss 0.02605220 - time (sec): 54.94 - samples/sec: 1777.47 - lr: 0.000023 - momentum: 0.000000 2023-10-16 21:51:08,170 epoch 6 - iter 1386/1546 - loss 0.02522412 - time (sec): 61.80 - samples/sec: 1771.19 - lr: 0.000023 - momentum: 0.000000 2023-10-16 21:51:15,235 epoch 6 - iter 1540/1546 - loss 0.02642731 - time (sec): 68.87 - samples/sec: 1795.68 - lr: 0.000022 - momentum: 0.000000 2023-10-16 21:51:15,502 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:51:15,502 EPOCH 6 done: loss 0.0266 - lr: 0.000022 2023-10-16 21:51:17,895 DEV : loss 0.10446853190660477 - f1-score (micro avg) 0.7795 2023-10-16 21:51:17,908 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:51:24,779 epoch 7 - iter 154/1546 - loss 0.02404472 - time (sec): 6.87 - samples/sec: 1839.81 - lr: 0.000022 - momentum: 0.000000 2023-10-16 21:51:31,664 epoch 7 - iter 308/1546 - loss 0.02056482 - time (sec): 13.75 - samples/sec: 1861.72 - lr: 0.000021 - momentum: 0.000000 2023-10-16 21:51:38,496 epoch 7 - iter 462/1546 - loss 0.02046697 - time (sec): 20.59 - samples/sec: 1840.24 - lr: 0.000021 - momentum: 0.000000 2023-10-16 21:51:45,451 epoch 7 - iter 616/1546 - loss 0.02166477 - time (sec): 27.54 - samples/sec: 1809.86 - lr: 0.000020 - momentum: 0.000000 2023-10-16 21:51:52,463 epoch 7 - iter 770/1546 - loss 0.02146336 - time (sec): 34.55 - samples/sec: 1805.13 - lr: 0.000019 - momentum: 0.000000 2023-10-16 21:51:59,383 epoch 7 - iter 924/1546 - loss 0.02253466 - time (sec): 41.47 - samples/sec: 1791.51 - lr: 0.000019 - momentum: 0.000000 2023-10-16 21:52:06,243 epoch 7 - iter 1078/1546 - loss 0.02170121 - time (sec): 48.33 - samples/sec: 1787.33 - lr: 0.000018 - momentum: 0.000000 2023-10-16 21:52:13,045 epoch 7 - iter 1232/1546 - loss 0.02115814 - time (sec): 55.14 - samples/sec: 1788.46 - lr: 0.000018 - momentum: 0.000000 2023-10-16 21:52:19,960 epoch 7 - iter 1386/1546 - loss 0.02075815 - time (sec): 62.05 - samples/sec: 1795.57 - lr: 0.000017 - momentum: 0.000000 2023-10-16 21:52:26,876 epoch 7 - iter 1540/1546 - loss 0.02048075 - time (sec): 68.97 - samples/sec: 1796.06 - lr: 0.000017 - momentum: 0.000000 2023-10-16 21:52:27,154 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:52:27,154 EPOCH 7 done: loss 0.0204 - lr: 0.000017 2023-10-16 21:52:29,150 DEV : loss 0.11760783195495605 - f1-score (micro avg) 0.7724 2023-10-16 21:52:29,163 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:52:36,119 epoch 8 - iter 154/1546 - loss 0.00904914 - time (sec): 6.96 - samples/sec: 1799.83 - lr: 0.000016 - momentum: 0.000000 2023-10-16 21:52:43,087 epoch 8 - iter 308/1546 - loss 0.01120019 - time (sec): 13.92 - samples/sec: 1821.61 - lr: 0.000016 - momentum: 0.000000 2023-10-16 21:52:49,903 epoch 8 - iter 462/1546 - loss 0.01332359 - time (sec): 20.74 - samples/sec: 1822.53 - lr: 0.000015 - momentum: 0.000000 2023-10-16 21:52:56,846 epoch 8 - iter 616/1546 - loss 0.01192538 - time (sec): 27.68 - samples/sec: 1842.43 - lr: 0.000014 - momentum: 0.000000 2023-10-16 21:53:03,643 epoch 8 - iter 770/1546 - loss 0.01078426 - time (sec): 34.48 - samples/sec: 1815.73 - lr: 0.000014 - momentum: 0.000000 2023-10-16 21:53:10,438 epoch 8 - iter 924/1546 - loss 0.01161029 - time (sec): 41.27 - samples/sec: 1813.84 - lr: 0.000013 - momentum: 0.000000 2023-10-16 21:53:17,145 epoch 8 - iter 1078/1546 - loss 0.01243620 - time (sec): 47.98 - samples/sec: 1810.28 - lr: 0.000013 - momentum: 0.000000 2023-10-16 21:53:23,967 epoch 8 - iter 1232/1546 - loss 0.01233856 - time (sec): 54.80 - samples/sec: 1804.05 - lr: 0.000012 - momentum: 0.000000 2023-10-16 21:53:30,776 epoch 8 - iter 1386/1546 - loss 0.01272431 - time (sec): 61.61 - samples/sec: 1800.38 - lr: 0.000012 - momentum: 0.000000 2023-10-16 21:53:37,834 epoch 8 - iter 1540/1546 - loss 0.01300626 - time (sec): 68.67 - samples/sec: 1804.05 - lr: 0.000011 - momentum: 0.000000 2023-10-16 21:53:38,096 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:53:38,096 EPOCH 8 done: loss 0.0130 - lr: 0.000011 2023-10-16 21:53:40,162 DEV : loss 0.11045785248279572 - f1-score (micro avg) 0.7898 2023-10-16 21:53:40,175 saving best model 2023-10-16 21:53:40,649 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:53:47,486 epoch 9 - iter 154/1546 - loss 0.00775423 - time (sec): 6.83 - samples/sec: 1831.86 - lr: 0.000011 - momentum: 0.000000 2023-10-16 21:53:54,019 epoch 9 - iter 308/1546 - loss 0.00790082 - time (sec): 13.36 - samples/sec: 1827.72 - lr: 0.000010 - momentum: 0.000000 2023-10-16 21:54:00,556 epoch 9 - iter 462/1546 - loss 0.00714745 - time (sec): 19.90 - samples/sec: 1837.52 - lr: 0.000009 - momentum: 0.000000 2023-10-16 21:54:07,065 epoch 9 - iter 616/1546 - loss 0.00761023 - time (sec): 26.41 - samples/sec: 1842.09 - lr: 0.000009 - momentum: 0.000000 2023-10-16 21:54:13,656 epoch 9 - iter 770/1546 - loss 0.00769417 - time (sec): 33.00 - samples/sec: 1841.10 - lr: 0.000008 - momentum: 0.000000 2023-10-16 21:54:20,344 epoch 9 - iter 924/1546 - loss 0.00778298 - time (sec): 39.69 - samples/sec: 1855.33 - lr: 0.000008 - momentum: 0.000000 2023-10-16 21:54:26,945 epoch 9 - iter 1078/1546 - loss 0.00794647 - time (sec): 46.29 - samples/sec: 1864.63 - lr: 0.000007 - momentum: 0.000000 2023-10-16 21:54:33,540 epoch 9 - iter 1232/1546 - loss 0.00813811 - time (sec): 52.89 - samples/sec: 1864.94 - lr: 0.000007 - momentum: 0.000000 2023-10-16 21:54:40,134 epoch 9 - iter 1386/1546 - loss 0.00823860 - time (sec): 59.48 - samples/sec: 1871.30 - lr: 0.000006 - momentum: 0.000000 2023-10-16 21:54:46,726 epoch 9 - iter 1540/1546 - loss 0.00873508 - time (sec): 66.07 - samples/sec: 1876.32 - lr: 0.000006 - momentum: 0.000000 2023-10-16 21:54:46,975 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:54:46,975 EPOCH 9 done: loss 0.0087 - lr: 0.000006 2023-10-16 21:54:48,998 DEV : loss 0.11864420771598816 - f1-score (micro avg) 0.7941 2023-10-16 21:54:49,011 saving best model 2023-10-16 21:54:49,452 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:54:56,289 epoch 10 - iter 154/1546 - loss 0.00586620 - time (sec): 6.83 - samples/sec: 1818.10 - lr: 0.000005 - momentum: 0.000000 2023-10-16 21:55:03,238 epoch 10 - iter 308/1546 - loss 0.00524542 - time (sec): 13.78 - samples/sec: 1808.72 - lr: 0.000004 - momentum: 0.000000 2023-10-16 21:55:10,143 epoch 10 - iter 462/1546 - loss 0.00563260 - time (sec): 20.69 - samples/sec: 1780.72 - lr: 0.000004 - momentum: 0.000000 2023-10-16 21:55:17,083 epoch 10 - iter 616/1546 - loss 0.00560671 - time (sec): 27.63 - samples/sec: 1810.14 - lr: 0.000003 - momentum: 0.000000 2023-10-16 21:55:23,981 epoch 10 - iter 770/1546 - loss 0.00551268 - time (sec): 34.53 - samples/sec: 1798.02 - lr: 0.000003 - momentum: 0.000000 2023-10-16 21:55:30,938 epoch 10 - iter 924/1546 - loss 0.00541465 - time (sec): 41.48 - samples/sec: 1806.94 - lr: 0.000002 - momentum: 0.000000 2023-10-16 21:55:37,821 epoch 10 - iter 1078/1546 - loss 0.00581212 - time (sec): 48.37 - samples/sec: 1798.64 - lr: 0.000002 - momentum: 0.000000 2023-10-16 21:55:44,637 epoch 10 - iter 1232/1546 - loss 0.00550965 - time (sec): 55.18 - samples/sec: 1802.49 - lr: 0.000001 - momentum: 0.000000 2023-10-16 21:55:51,547 epoch 10 - iter 1386/1546 - loss 0.00535177 - time (sec): 62.09 - samples/sec: 1793.62 - lr: 0.000001 - momentum: 0.000000 2023-10-16 21:55:58,452 epoch 10 - iter 1540/1546 - loss 0.00556052 - time (sec): 69.00 - samples/sec: 1791.86 - lr: 0.000000 - momentum: 0.000000 2023-10-16 21:55:58,718 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:55:58,719 EPOCH 10 done: loss 0.0055 - lr: 0.000000 2023-10-16 21:56:00,754 DEV : loss 0.1224876418709755 - f1-score (micro avg) 0.7966 2023-10-16 21:56:00,767 saving best model 2023-10-16 21:56:01,611 ---------------------------------------------------------------------------------------------------- 2023-10-16 21:56:01,612 Loading model from best epoch ... 2023-10-16 21:56:03,412 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 2023-10-16 21:56:09,090 Results: - F-score (micro) 0.8125 - F-score (macro) 0.7224 - Accuracy 0.7061 By class: precision recall f1-score support LOC 0.8716 0.8467 0.8590 946 BUILDING 0.6347 0.5730 0.6023 185 STREET 0.6667 0.7500 0.7059 56 micro avg 0.8259 0.7995 0.8125 1187 macro avg 0.7243 0.7232 0.7224 1187 weighted avg 0.8250 0.7995 0.8117 1187 2023-10-16 21:56:09,090 ----------------------------------------------------------------------------------------------------