2023-10-17 13:08:39,919 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:08:39,921 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 13:08:39,921 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:08:39,921 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 13:08:39,921 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:08:39,921 Train: 6183 sentences 2023-10-17 13:08:39,921 (train_with_dev=False, train_with_test=False) 2023-10-17 13:08:39,922 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:08:39,922 Training Params: 2023-10-17 13:08:39,922 - learning_rate: "3e-05" 2023-10-17 13:08:39,922 - mini_batch_size: "8" 2023-10-17 13:08:39,922 - max_epochs: "10" 2023-10-17 13:08:39,922 - shuffle: "True" 2023-10-17 13:08:39,922 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:08:39,922 Plugins: 2023-10-17 13:08:39,922 - TensorboardLogger 2023-10-17 13:08:39,922 - LinearScheduler | warmup_fraction: '0.1' 2023-10-17 13:08:39,922 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:08:39,922 Final evaluation on model from best epoch (best-model.pt) 2023-10-17 13:08:39,922 - metric: "('micro avg', 'f1-score')" 2023-10-17 13:08:39,922 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:08:39,922 Computation: 2023-10-17 13:08:39,922 - compute on device: cuda:0 2023-10-17 13:08:39,923 - embedding storage: none 2023-10-17 13:08:39,923 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:08:39,923 Model training base path: "hmbench-topres19th/en-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4" 2023-10-17 13:08:39,923 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:08:39,923 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:08:39,923 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-17 13:08:46,795 epoch 1 - iter 77/773 - loss 2.97835929 - time (sec): 6.87 - samples/sec: 1745.08 - lr: 0.000003 - momentum: 0.000000 2023-10-17 13:08:54,296 epoch 1 - iter 154/773 - loss 1.87469467 - time (sec): 14.37 - samples/sec: 1632.78 - lr: 0.000006 - momentum: 0.000000 2023-10-17 13:09:01,982 epoch 1 - iter 231/773 - loss 1.29875250 - time (sec): 22.06 - samples/sec: 1621.70 - lr: 0.000009 - momentum: 0.000000 2023-10-17 13:09:09,388 epoch 1 - iter 308/773 - loss 1.01072496 - time (sec): 29.46 - samples/sec: 1621.66 - lr: 0.000012 - momentum: 0.000000 2023-10-17 13:09:17,277 epoch 1 - iter 385/773 - loss 0.81975956 - time (sec): 37.35 - samples/sec: 1635.34 - lr: 0.000015 - momentum: 0.000000 2023-10-17 13:09:25,191 epoch 1 - iter 462/773 - loss 0.70447128 - time (sec): 45.27 - samples/sec: 1627.15 - lr: 0.000018 - momentum: 0.000000 2023-10-17 13:09:32,881 epoch 1 - iter 539/773 - loss 0.62180756 - time (sec): 52.96 - samples/sec: 1620.95 - lr: 0.000021 - momentum: 0.000000 2023-10-17 13:09:41,089 epoch 1 - iter 616/773 - loss 0.55646217 - time (sec): 61.16 - samples/sec: 1609.64 - lr: 0.000024 - momentum: 0.000000 2023-10-17 13:09:48,700 epoch 1 - iter 693/773 - loss 0.50448744 - time (sec): 68.78 - samples/sec: 1614.09 - lr: 0.000027 - momentum: 0.000000 2023-10-17 13:09:56,705 epoch 1 - iter 770/773 - loss 0.46197740 - time (sec): 76.78 - samples/sec: 1611.60 - lr: 0.000030 - momentum: 0.000000 2023-10-17 13:09:57,056 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:09:57,057 EPOCH 1 done: loss 0.4601 - lr: 0.000030 2023-10-17 13:10:00,370 DEV : loss 0.055146388709545135 - f1-score (micro avg) 0.7536 2023-10-17 13:10:00,402 saving best model 2023-10-17 13:10:01,040 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:10:08,616 epoch 2 - iter 77/773 - loss 0.08891357 - time (sec): 7.57 - samples/sec: 1619.07 - lr: 0.000030 - momentum: 0.000000 2023-10-17 13:10:16,519 epoch 2 - iter 154/773 - loss 0.08025883 - time (sec): 15.48 - samples/sec: 1618.18 - lr: 0.000029 - momentum: 0.000000 2023-10-17 13:10:23,631 epoch 2 - iter 231/773 - loss 0.07291651 - time (sec): 22.59 - samples/sec: 1636.92 - lr: 0.000029 - momentum: 0.000000 2023-10-17 13:10:30,834 epoch 2 - iter 308/773 - loss 0.07668126 - time (sec): 29.79 - samples/sec: 1651.60 - lr: 0.000029 - momentum: 0.000000 2023-10-17 13:10:38,528 epoch 2 - iter 385/773 - loss 0.07749694 - time (sec): 37.48 - samples/sec: 1640.58 - lr: 0.000028 - momentum: 0.000000 2023-10-17 13:10:46,505 epoch 2 - iter 462/773 - loss 0.07639212 - time (sec): 45.46 - samples/sec: 1618.29 - lr: 0.000028 - momentum: 0.000000 2023-10-17 13:10:54,438 epoch 2 - iter 539/773 - loss 0.07525245 - time (sec): 53.40 - samples/sec: 1615.54 - lr: 0.000028 - momentum: 0.000000 2023-10-17 13:11:02,973 epoch 2 - iter 616/773 - loss 0.07570975 - time (sec): 61.93 - samples/sec: 1583.02 - lr: 0.000027 - momentum: 0.000000 2023-10-17 13:11:11,049 epoch 2 - iter 693/773 - loss 0.07390561 - time (sec): 70.01 - samples/sec: 1586.06 - lr: 0.000027 - momentum: 0.000000 2023-10-17 13:11:18,885 epoch 2 - iter 770/773 - loss 0.07331939 - time (sec): 77.84 - samples/sec: 1592.96 - lr: 0.000027 - momentum: 0.000000 2023-10-17 13:11:19,172 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:11:19,173 EPOCH 2 done: loss 0.0733 - lr: 0.000027 2023-10-17 13:11:22,464 DEV : loss 0.0555521659553051 - f1-score (micro avg) 0.7828 2023-10-17 13:11:22,493 saving best model 2023-10-17 13:11:23,970 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:11:31,790 epoch 3 - iter 77/773 - loss 0.04302712 - time (sec): 7.82 - samples/sec: 1736.74 - lr: 0.000026 - momentum: 0.000000 2023-10-17 13:11:39,410 epoch 3 - iter 154/773 - loss 0.04215121 - time (sec): 15.44 - samples/sec: 1646.04 - lr: 0.000026 - momentum: 0.000000 2023-10-17 13:11:47,665 epoch 3 - iter 231/773 - loss 0.04258486 - time (sec): 23.69 - samples/sec: 1601.66 - lr: 0.000026 - momentum: 0.000000 2023-10-17 13:11:55,609 epoch 3 - iter 308/773 - loss 0.04530148 - time (sec): 31.64 - samples/sec: 1574.86 - lr: 0.000025 - momentum: 0.000000 2023-10-17 13:12:04,204 epoch 3 - iter 385/773 - loss 0.04445010 - time (sec): 40.23 - samples/sec: 1585.64 - lr: 0.000025 - momentum: 0.000000 2023-10-17 13:12:12,128 epoch 3 - iter 462/773 - loss 0.04672927 - time (sec): 48.16 - samples/sec: 1566.26 - lr: 0.000025 - momentum: 0.000000 2023-10-17 13:12:19,104 epoch 3 - iter 539/773 - loss 0.04829116 - time (sec): 55.13 - samples/sec: 1593.13 - lr: 0.000024 - momentum: 0.000000 2023-10-17 13:12:26,282 epoch 3 - iter 616/773 - loss 0.04728436 - time (sec): 62.31 - samples/sec: 1603.68 - lr: 0.000024 - momentum: 0.000000 2023-10-17 13:12:34,349 epoch 3 - iter 693/773 - loss 0.04777688 - time (sec): 70.38 - samples/sec: 1592.63 - lr: 0.000024 - momentum: 0.000000 2023-10-17 13:12:41,971 epoch 3 - iter 770/773 - loss 0.04728166 - time (sec): 78.00 - samples/sec: 1588.52 - lr: 0.000023 - momentum: 0.000000 2023-10-17 13:12:42,251 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:12:42,252 EPOCH 3 done: loss 0.0471 - lr: 0.000023 2023-10-17 13:12:45,338 DEV : loss 0.06297728419303894 - f1-score (micro avg) 0.7748 2023-10-17 13:12:45,370 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:12:53,983 epoch 4 - iter 77/773 - loss 0.02732323 - time (sec): 8.61 - samples/sec: 1445.84 - lr: 0.000023 - momentum: 0.000000 2023-10-17 13:13:02,146 epoch 4 - iter 154/773 - loss 0.03181965 - time (sec): 16.77 - samples/sec: 1493.94 - lr: 0.000023 - momentum: 0.000000 2023-10-17 13:13:10,153 epoch 4 - iter 231/773 - loss 0.03214623 - time (sec): 24.78 - samples/sec: 1480.43 - lr: 0.000022 - momentum: 0.000000 2023-10-17 13:13:17,716 epoch 4 - iter 308/773 - loss 0.02980114 - time (sec): 32.34 - samples/sec: 1528.22 - lr: 0.000022 - momentum: 0.000000 2023-10-17 13:13:25,433 epoch 4 - iter 385/773 - loss 0.02875251 - time (sec): 40.06 - samples/sec: 1551.31 - lr: 0.000022 - momentum: 0.000000 2023-10-17 13:13:33,344 epoch 4 - iter 462/773 - loss 0.03233554 - time (sec): 47.97 - samples/sec: 1560.68 - lr: 0.000021 - momentum: 0.000000 2023-10-17 13:13:41,357 epoch 4 - iter 539/773 - loss 0.03122506 - time (sec): 55.99 - samples/sec: 1560.44 - lr: 0.000021 - momentum: 0.000000 2023-10-17 13:13:49,884 epoch 4 - iter 616/773 - loss 0.03233573 - time (sec): 64.51 - samples/sec: 1539.64 - lr: 0.000021 - momentum: 0.000000 2023-10-17 13:13:57,748 epoch 4 - iter 693/773 - loss 0.03209738 - time (sec): 72.38 - samples/sec: 1538.03 - lr: 0.000020 - momentum: 0.000000 2023-10-17 13:14:04,906 epoch 4 - iter 770/773 - loss 0.03230222 - time (sec): 79.53 - samples/sec: 1557.31 - lr: 0.000020 - momentum: 0.000000 2023-10-17 13:14:05,197 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:14:05,197 EPOCH 4 done: loss 0.0325 - lr: 0.000020 2023-10-17 13:14:08,468 DEV : loss 0.09397705644369125 - f1-score (micro avg) 0.7876 2023-10-17 13:14:08,497 saving best model 2023-10-17 13:14:09,959 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:14:17,907 epoch 5 - iter 77/773 - loss 0.02528085 - time (sec): 7.94 - samples/sec: 1628.49 - lr: 0.000020 - momentum: 0.000000 2023-10-17 13:14:25,737 epoch 5 - iter 154/773 - loss 0.02527426 - time (sec): 15.77 - samples/sec: 1580.91 - lr: 0.000019 - momentum: 0.000000 2023-10-17 13:14:33,774 epoch 5 - iter 231/773 - loss 0.02362748 - time (sec): 23.81 - samples/sec: 1552.99 - lr: 0.000019 - momentum: 0.000000 2023-10-17 13:14:41,673 epoch 5 - iter 308/773 - loss 0.02239320 - time (sec): 31.71 - samples/sec: 1559.77 - lr: 0.000019 - momentum: 0.000000 2023-10-17 13:14:49,222 epoch 5 - iter 385/773 - loss 0.02330606 - time (sec): 39.26 - samples/sec: 1554.33 - lr: 0.000018 - momentum: 0.000000 2023-10-17 13:14:56,980 epoch 5 - iter 462/773 - loss 0.02385192 - time (sec): 47.02 - samples/sec: 1562.03 - lr: 0.000018 - momentum: 0.000000 2023-10-17 13:15:05,089 epoch 5 - iter 539/773 - loss 0.02449892 - time (sec): 55.13 - samples/sec: 1542.51 - lr: 0.000018 - momentum: 0.000000 2023-10-17 13:15:13,885 epoch 5 - iter 616/773 - loss 0.02394531 - time (sec): 63.92 - samples/sec: 1527.94 - lr: 0.000017 - momentum: 0.000000 2023-10-17 13:15:21,845 epoch 5 - iter 693/773 - loss 0.02348430 - time (sec): 71.88 - samples/sec: 1546.02 - lr: 0.000017 - momentum: 0.000000 2023-10-17 13:15:29,820 epoch 5 - iter 770/773 - loss 0.02290727 - time (sec): 79.86 - samples/sec: 1550.24 - lr: 0.000017 - momentum: 0.000000 2023-10-17 13:15:30,108 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:15:30,108 EPOCH 5 done: loss 0.0231 - lr: 0.000017 2023-10-17 13:15:33,581 DEV : loss 0.09673922508955002 - f1-score (micro avg) 0.7871 2023-10-17 13:15:33,614 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:15:41,113 epoch 6 - iter 77/773 - loss 0.01282970 - time (sec): 7.50 - samples/sec: 1576.16 - lr: 0.000016 - momentum: 0.000000 2023-10-17 13:15:48,997 epoch 6 - iter 154/773 - loss 0.01426688 - time (sec): 15.38 - samples/sec: 1607.45 - lr: 0.000016 - momentum: 0.000000 2023-10-17 13:15:57,056 epoch 6 - iter 231/773 - loss 0.01542611 - time (sec): 23.44 - samples/sec: 1597.10 - lr: 0.000016 - momentum: 0.000000 2023-10-17 13:16:04,734 epoch 6 - iter 308/773 - loss 0.01498317 - time (sec): 31.12 - samples/sec: 1618.59 - lr: 0.000015 - momentum: 0.000000 2023-10-17 13:16:11,813 epoch 6 - iter 385/773 - loss 0.01554285 - time (sec): 38.20 - samples/sec: 1621.76 - lr: 0.000015 - momentum: 0.000000 2023-10-17 13:16:19,174 epoch 6 - iter 462/773 - loss 0.01570361 - time (sec): 45.56 - samples/sec: 1618.77 - lr: 0.000015 - momentum: 0.000000 2023-10-17 13:16:26,803 epoch 6 - iter 539/773 - loss 0.01538039 - time (sec): 53.19 - samples/sec: 1608.65 - lr: 0.000014 - momentum: 0.000000 2023-10-17 13:16:34,412 epoch 6 - iter 616/773 - loss 0.01501943 - time (sec): 60.80 - samples/sec: 1610.50 - lr: 0.000014 - momentum: 0.000000 2023-10-17 13:16:42,341 epoch 6 - iter 693/773 - loss 0.01478103 - time (sec): 68.72 - samples/sec: 1614.46 - lr: 0.000014 - momentum: 0.000000 2023-10-17 13:16:50,931 epoch 6 - iter 770/773 - loss 0.01499341 - time (sec): 77.31 - samples/sec: 1601.28 - lr: 0.000013 - momentum: 0.000000 2023-10-17 13:16:51,209 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:16:51,209 EPOCH 6 done: loss 0.0149 - lr: 0.000013 2023-10-17 13:16:54,440 DEV : loss 0.10239903628826141 - f1-score (micro avg) 0.8145 2023-10-17 13:16:54,472 saving best model 2023-10-17 13:16:55,963 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:17:03,813 epoch 7 - iter 77/773 - loss 0.00993657 - time (sec): 7.85 - samples/sec: 1575.71 - lr: 0.000013 - momentum: 0.000000 2023-10-17 13:17:11,672 epoch 7 - iter 154/773 - loss 0.01083244 - time (sec): 15.71 - samples/sec: 1559.76 - lr: 0.000013 - momentum: 0.000000 2023-10-17 13:17:20,126 epoch 7 - iter 231/773 - loss 0.01440738 - time (sec): 24.16 - samples/sec: 1534.55 - lr: 0.000012 - momentum: 0.000000 2023-10-17 13:17:28,474 epoch 7 - iter 308/773 - loss 0.01310777 - time (sec): 32.51 - samples/sec: 1509.11 - lr: 0.000012 - momentum: 0.000000 2023-10-17 13:17:36,571 epoch 7 - iter 385/773 - loss 0.01226929 - time (sec): 40.61 - samples/sec: 1518.86 - lr: 0.000012 - momentum: 0.000000 2023-10-17 13:17:44,264 epoch 7 - iter 462/773 - loss 0.01168054 - time (sec): 48.30 - samples/sec: 1512.85 - lr: 0.000011 - momentum: 0.000000 2023-10-17 13:17:51,543 epoch 7 - iter 539/773 - loss 0.01131055 - time (sec): 55.58 - samples/sec: 1550.49 - lr: 0.000011 - momentum: 0.000000 2023-10-17 13:17:59,150 epoch 7 - iter 616/773 - loss 0.01118152 - time (sec): 63.18 - samples/sec: 1570.31 - lr: 0.000011 - momentum: 0.000000 2023-10-17 13:18:06,493 epoch 7 - iter 693/773 - loss 0.01095252 - time (sec): 70.53 - samples/sec: 1583.36 - lr: 0.000010 - momentum: 0.000000 2023-10-17 13:18:14,096 epoch 7 - iter 770/773 - loss 0.01093922 - time (sec): 78.13 - samples/sec: 1584.50 - lr: 0.000010 - momentum: 0.000000 2023-10-17 13:18:14,408 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:18:14,408 EPOCH 7 done: loss 0.0109 - lr: 0.000010 2023-10-17 13:18:17,632 DEV : loss 0.11184219270944595 - f1-score (micro avg) 0.8057 2023-10-17 13:18:17,665 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:18:25,406 epoch 8 - iter 77/773 - loss 0.00400052 - time (sec): 7.74 - samples/sec: 1593.59 - lr: 0.000010 - momentum: 0.000000 2023-10-17 13:18:33,177 epoch 8 - iter 154/773 - loss 0.00447152 - time (sec): 15.51 - samples/sec: 1569.21 - lr: 0.000009 - momentum: 0.000000 2023-10-17 13:18:40,878 epoch 8 - iter 231/773 - loss 0.00490799 - time (sec): 23.21 - samples/sec: 1565.41 - lr: 0.000009 - momentum: 0.000000 2023-10-17 13:18:48,509 epoch 8 - iter 308/773 - loss 0.00494344 - time (sec): 30.84 - samples/sec: 1567.92 - lr: 0.000009 - momentum: 0.000000 2023-10-17 13:18:56,212 epoch 8 - iter 385/773 - loss 0.00625829 - time (sec): 38.54 - samples/sec: 1578.24 - lr: 0.000008 - momentum: 0.000000 2023-10-17 13:19:04,489 epoch 8 - iter 462/773 - loss 0.00613140 - time (sec): 46.82 - samples/sec: 1564.49 - lr: 0.000008 - momentum: 0.000000 2023-10-17 13:19:13,691 epoch 8 - iter 539/773 - loss 0.00633907 - time (sec): 56.02 - samples/sec: 1552.09 - lr: 0.000008 - momentum: 0.000000 2023-10-17 13:19:22,168 epoch 8 - iter 616/773 - loss 0.00615248 - time (sec): 64.50 - samples/sec: 1546.11 - lr: 0.000007 - momentum: 0.000000 2023-10-17 13:19:30,578 epoch 8 - iter 693/773 - loss 0.00641877 - time (sec): 72.91 - samples/sec: 1533.68 - lr: 0.000007 - momentum: 0.000000 2023-10-17 13:19:37,585 epoch 8 - iter 770/773 - loss 0.00648931 - time (sec): 79.92 - samples/sec: 1550.50 - lr: 0.000007 - momentum: 0.000000 2023-10-17 13:19:37,865 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:19:37,865 EPOCH 8 done: loss 0.0066 - lr: 0.000007 2023-10-17 13:19:40,914 DEV : loss 0.1238672062754631 - f1-score (micro avg) 0.7701 2023-10-17 13:19:40,948 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:19:47,862 epoch 9 - iter 77/773 - loss 0.00422951 - time (sec): 6.91 - samples/sec: 1776.61 - lr: 0.000006 - momentum: 0.000000 2023-10-17 13:19:55,147 epoch 9 - iter 154/773 - loss 0.00410486 - time (sec): 14.20 - samples/sec: 1755.62 - lr: 0.000006 - momentum: 0.000000 2023-10-17 13:20:02,671 epoch 9 - iter 231/773 - loss 0.00365045 - time (sec): 21.72 - samples/sec: 1765.62 - lr: 0.000006 - momentum: 0.000000 2023-10-17 13:20:10,380 epoch 9 - iter 308/773 - loss 0.00392863 - time (sec): 29.43 - samples/sec: 1757.76 - lr: 0.000005 - momentum: 0.000000 2023-10-17 13:20:17,497 epoch 9 - iter 385/773 - loss 0.00446919 - time (sec): 36.55 - samples/sec: 1704.88 - lr: 0.000005 - momentum: 0.000000 2023-10-17 13:20:24,563 epoch 9 - iter 462/773 - loss 0.00449122 - time (sec): 43.61 - samples/sec: 1700.22 - lr: 0.000005 - momentum: 0.000000 2023-10-17 13:20:31,806 epoch 9 - iter 539/773 - loss 0.00450468 - time (sec): 50.86 - samples/sec: 1698.95 - lr: 0.000004 - momentum: 0.000000 2023-10-17 13:20:39,046 epoch 9 - iter 616/773 - loss 0.00433734 - time (sec): 58.10 - samples/sec: 1694.55 - lr: 0.000004 - momentum: 0.000000 2023-10-17 13:20:46,096 epoch 9 - iter 693/773 - loss 0.00402114 - time (sec): 65.15 - samples/sec: 1695.80 - lr: 0.000004 - momentum: 0.000000 2023-10-17 13:20:53,731 epoch 9 - iter 770/773 - loss 0.00408108 - time (sec): 72.78 - samples/sec: 1699.18 - lr: 0.000003 - momentum: 0.000000 2023-10-17 13:20:54,012 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:20:54,013 EPOCH 9 done: loss 0.0041 - lr: 0.000003 2023-10-17 13:20:56,844 DEV : loss 0.12823046743869781 - f1-score (micro avg) 0.7897 2023-10-17 13:20:56,875 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:21:04,003 epoch 10 - iter 77/773 - loss 0.00299248 - time (sec): 7.13 - samples/sec: 1816.45 - lr: 0.000003 - momentum: 0.000000 2023-10-17 13:21:11,274 epoch 10 - iter 154/773 - loss 0.00431058 - time (sec): 14.40 - samples/sec: 1760.63 - lr: 0.000003 - momentum: 0.000000 2023-10-17 13:21:18,785 epoch 10 - iter 231/773 - loss 0.00336600 - time (sec): 21.91 - samples/sec: 1768.91 - lr: 0.000002 - momentum: 0.000000 2023-10-17 13:21:26,524 epoch 10 - iter 308/773 - loss 0.00313090 - time (sec): 29.65 - samples/sec: 1721.62 - lr: 0.000002 - momentum: 0.000000 2023-10-17 13:21:33,931 epoch 10 - iter 385/773 - loss 0.00278220 - time (sec): 37.05 - samples/sec: 1705.96 - lr: 0.000002 - momentum: 0.000000 2023-10-17 13:21:40,847 epoch 10 - iter 462/773 - loss 0.00299155 - time (sec): 43.97 - samples/sec: 1724.70 - lr: 0.000001 - momentum: 0.000000 2023-10-17 13:21:48,002 epoch 10 - iter 539/773 - loss 0.00280852 - time (sec): 51.12 - samples/sec: 1716.44 - lr: 0.000001 - momentum: 0.000000 2023-10-17 13:21:54,933 epoch 10 - iter 616/773 - loss 0.00285345 - time (sec): 58.06 - samples/sec: 1706.62 - lr: 0.000001 - momentum: 0.000000 2023-10-17 13:22:02,114 epoch 10 - iter 693/773 - loss 0.00280881 - time (sec): 65.24 - samples/sec: 1701.66 - lr: 0.000000 - momentum: 0.000000 2023-10-17 13:22:09,626 epoch 10 - iter 770/773 - loss 0.00273836 - time (sec): 72.75 - samples/sec: 1701.49 - lr: 0.000000 - momentum: 0.000000 2023-10-17 13:22:09,899 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:22:09,899 EPOCH 10 done: loss 0.0028 - lr: 0.000000 2023-10-17 13:22:12,830 DEV : loss 0.12347615510225296 - f1-score (micro avg) 0.8024 2023-10-17 13:22:13,409 ---------------------------------------------------------------------------------------------------- 2023-10-17 13:22:13,411 Loading model from best epoch ... 2023-10-17 13:22:15,686 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 13:22:24,113 Results: - F-score (micro) 0.8043 - F-score (macro) 0.7283 - Accuracy 0.691 By class: precision recall f1-score support LOC 0.8384 0.8499 0.8441 946 BUILDING 0.6218 0.6486 0.6349 185 STREET 0.6667 0.7500 0.7059 56 micro avg 0.7951 0.8138 0.8043 1187 macro avg 0.7089 0.7495 0.7283 1187 weighted avg 0.7965 0.8138 0.8050 1187 2023-10-17 13:22:24,113 ----------------------------------------------------------------------------------------------------