2023-10-18 20:56:03,815 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:56:03,816 Model: "SequenceTagger( (embeddings): TransformerWordEmbeddings( (model): BertModel( (embeddings): BertEmbeddings( (word_embeddings): Embedding(32001, 128) (position_embeddings): Embedding(512, 128) (token_type_embeddings): Embedding(2, 128) (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) (encoder): BertEncoder( (layer): ModuleList( (0-1): 2 x BertLayer( (attention): BertAttention( (self): BertSelfAttention( (query): Linear(in_features=128, out_features=128, bias=True) (key): Linear(in_features=128, out_features=128, bias=True) (value): Linear(in_features=128, out_features=128, bias=True) (dropout): Dropout(p=0.1, inplace=False) ) (output): BertSelfOutput( (dense): Linear(in_features=128, out_features=128, bias=True) (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) (intermediate): BertIntermediate( (dense): Linear(in_features=128, out_features=512, bias=True) (intermediate_act_fn): GELUActivation() ) (output): BertOutput( (dense): Linear(in_features=512, out_features=128, bias=True) (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) ) (pooler): BertPooler( (dense): Linear(in_features=128, out_features=128, bias=True) (activation): Tanh() ) ) ) (locked_dropout): LockedDropout(p=0.5) (linear): Linear(in_features=128, out_features=13, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-18 20:56:03,816 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:56:03,816 MultiCorpus: 7936 train + 992 dev + 992 test sentences - NER_ICDAR_EUROPEANA Corpus: 7936 train + 992 dev + 992 test sentences - /root/.flair/datasets/ner_icdar_europeana/fr 2023-10-18 20:56:03,816 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:56:03,816 Train: 7936 sentences 2023-10-18 20:56:03,816 (train_with_dev=False, train_with_test=False) 2023-10-18 20:56:03,816 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:56:03,816 Training Params: 2023-10-18 20:56:03,816 - learning_rate: "5e-05" 2023-10-18 20:56:03,816 - mini_batch_size: "8" 2023-10-18 20:56:03,816 - max_epochs: "10" 2023-10-18 20:56:03,816 - shuffle: "True" 2023-10-18 20:56:03,816 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:56:03,816 Plugins: 2023-10-18 20:56:03,816 - TensorboardLogger 2023-10-18 20:56:03,816 - LinearScheduler | warmup_fraction: '0.1' 2023-10-18 20:56:03,816 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:56:03,816 Final evaluation on model from best epoch (best-model.pt) 2023-10-18 20:56:03,816 - metric: "('micro avg', 'f1-score')" 2023-10-18 20:56:03,816 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:56:03,816 Computation: 2023-10-18 20:56:03,816 - compute on device: cuda:0 2023-10-18 20:56:03,816 - embedding storage: none 2023-10-18 20:56:03,816 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:56:03,816 Model training base path: "hmbench-icdar/fr-dbmdz/bert-tiny-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2" 2023-10-18 20:56:03,816 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:56:03,816 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:56:03,817 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-18 20:56:06,181 epoch 1 - iter 99/992 - loss 2.37417653 - time (sec): 2.36 - samples/sec: 6970.36 - lr: 0.000005 - momentum: 0.000000 2023-10-18 20:56:08,505 epoch 1 - iter 198/992 - loss 2.04547181 - time (sec): 4.69 - samples/sec: 6978.26 - lr: 0.000010 - momentum: 0.000000 2023-10-18 20:56:10,702 epoch 1 - iter 297/992 - loss 1.63644393 - time (sec): 6.88 - samples/sec: 7234.43 - lr: 0.000015 - momentum: 0.000000 2023-10-18 20:56:12,678 epoch 1 - iter 396/992 - loss 1.35011565 - time (sec): 8.86 - samples/sec: 7467.10 - lr: 0.000020 - momentum: 0.000000 2023-10-18 20:56:14,632 epoch 1 - iter 495/992 - loss 1.18194660 - time (sec): 10.82 - samples/sec: 7693.85 - lr: 0.000025 - momentum: 0.000000 2023-10-18 20:56:16,785 epoch 1 - iter 594/992 - loss 1.05850892 - time (sec): 12.97 - samples/sec: 7664.55 - lr: 0.000030 - momentum: 0.000000 2023-10-18 20:56:18,922 epoch 1 - iter 693/992 - loss 0.96209046 - time (sec): 15.11 - samples/sec: 7664.05 - lr: 0.000035 - momentum: 0.000000 2023-10-18 20:56:21,060 epoch 1 - iter 792/992 - loss 0.88404985 - time (sec): 17.24 - samples/sec: 7634.19 - lr: 0.000040 - momentum: 0.000000 2023-10-18 20:56:23,256 epoch 1 - iter 891/992 - loss 0.82833250 - time (sec): 19.44 - samples/sec: 7576.98 - lr: 0.000045 - momentum: 0.000000 2023-10-18 20:56:25,516 epoch 1 - iter 990/992 - loss 0.77917868 - time (sec): 21.70 - samples/sec: 7543.24 - lr: 0.000050 - momentum: 0.000000 2023-10-18 20:56:25,565 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:56:25,565 EPOCH 1 done: loss 0.7778 - lr: 0.000050 2023-10-18 20:56:27,108 DEV : loss 0.2251289188861847 - f1-score (micro avg) 0.2771 2023-10-18 20:56:27,126 saving best model 2023-10-18 20:56:27,158 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:56:29,368 epoch 2 - iter 99/992 - loss 0.32167945 - time (sec): 2.21 - samples/sec: 7301.80 - lr: 0.000049 - momentum: 0.000000 2023-10-18 20:56:31,577 epoch 2 - iter 198/992 - loss 0.29255043 - time (sec): 4.42 - samples/sec: 7535.62 - lr: 0.000049 - momentum: 0.000000 2023-10-18 20:56:33,795 epoch 2 - iter 297/992 - loss 0.29067708 - time (sec): 6.64 - samples/sec: 7386.21 - lr: 0.000048 - momentum: 0.000000 2023-10-18 20:56:36,015 epoch 2 - iter 396/992 - loss 0.29025565 - time (sec): 8.86 - samples/sec: 7363.21 - lr: 0.000048 - momentum: 0.000000 2023-10-18 20:56:38,278 epoch 2 - iter 495/992 - loss 0.28143687 - time (sec): 11.12 - samples/sec: 7310.90 - lr: 0.000047 - momentum: 0.000000 2023-10-18 20:56:40,527 epoch 2 - iter 594/992 - loss 0.27945389 - time (sec): 13.37 - samples/sec: 7270.36 - lr: 0.000047 - momentum: 0.000000 2023-10-18 20:56:42,728 epoch 2 - iter 693/992 - loss 0.27975877 - time (sec): 15.57 - samples/sec: 7261.00 - lr: 0.000046 - momentum: 0.000000 2023-10-18 20:56:45,070 epoch 2 - iter 792/992 - loss 0.27735348 - time (sec): 17.91 - samples/sec: 7256.35 - lr: 0.000046 - momentum: 0.000000 2023-10-18 20:56:47,388 epoch 2 - iter 891/992 - loss 0.27681776 - time (sec): 20.23 - samples/sec: 7272.50 - lr: 0.000045 - momentum: 0.000000 2023-10-18 20:56:49,647 epoch 2 - iter 990/992 - loss 0.27381910 - time (sec): 22.49 - samples/sec: 7277.31 - lr: 0.000044 - momentum: 0.000000 2023-10-18 20:56:49,690 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:56:49,691 EPOCH 2 done: loss 0.2739 - lr: 0.000044 2023-10-18 20:56:51,906 DEV : loss 0.18277420103549957 - f1-score (micro avg) 0.3656 2023-10-18 20:56:51,924 saving best model 2023-10-18 20:56:51,958 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:56:54,200 epoch 3 - iter 99/992 - loss 0.22769296 - time (sec): 2.24 - samples/sec: 7466.59 - lr: 0.000044 - momentum: 0.000000 2023-10-18 20:56:56,425 epoch 3 - iter 198/992 - loss 0.23820508 - time (sec): 4.47 - samples/sec: 7317.55 - lr: 0.000043 - momentum: 0.000000 2023-10-18 20:56:58,657 epoch 3 - iter 297/992 - loss 0.24041108 - time (sec): 6.70 - samples/sec: 7300.94 - lr: 0.000043 - momentum: 0.000000 2023-10-18 20:57:00,926 epoch 3 - iter 396/992 - loss 0.23852730 - time (sec): 8.97 - samples/sec: 7228.49 - lr: 0.000042 - momentum: 0.000000 2023-10-18 20:57:03,244 epoch 3 - iter 495/992 - loss 0.23174392 - time (sec): 11.29 - samples/sec: 7235.37 - lr: 0.000042 - momentum: 0.000000 2023-10-18 20:57:05,436 epoch 3 - iter 594/992 - loss 0.23397501 - time (sec): 13.48 - samples/sec: 7202.93 - lr: 0.000041 - momentum: 0.000000 2023-10-18 20:57:07,666 epoch 3 - iter 693/992 - loss 0.23953902 - time (sec): 15.71 - samples/sec: 7229.79 - lr: 0.000041 - momentum: 0.000000 2023-10-18 20:57:09,905 epoch 3 - iter 792/992 - loss 0.23601306 - time (sec): 17.95 - samples/sec: 7263.88 - lr: 0.000040 - momentum: 0.000000 2023-10-18 20:57:12,148 epoch 3 - iter 891/992 - loss 0.23207443 - time (sec): 20.19 - samples/sec: 7280.65 - lr: 0.000039 - momentum: 0.000000 2023-10-18 20:57:14,478 epoch 3 - iter 990/992 - loss 0.23020989 - time (sec): 22.52 - samples/sec: 7262.19 - lr: 0.000039 - momentum: 0.000000 2023-10-18 20:57:14,524 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:57:14,525 EPOCH 3 done: loss 0.2301 - lr: 0.000039 2023-10-18 20:57:16,353 DEV : loss 0.1740075945854187 - f1-score (micro avg) 0.3899 2023-10-18 20:57:16,372 saving best model 2023-10-18 20:57:16,407 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:57:18,696 epoch 4 - iter 99/992 - loss 0.21972604 - time (sec): 2.29 - samples/sec: 6880.22 - lr: 0.000038 - momentum: 0.000000 2023-10-18 20:57:20,907 epoch 4 - iter 198/992 - loss 0.21771922 - time (sec): 4.50 - samples/sec: 7255.82 - lr: 0.000038 - momentum: 0.000000 2023-10-18 20:57:23,071 epoch 4 - iter 297/992 - loss 0.21406813 - time (sec): 6.66 - samples/sec: 7184.34 - lr: 0.000037 - momentum: 0.000000 2023-10-18 20:57:25,280 epoch 4 - iter 396/992 - loss 0.21159074 - time (sec): 8.87 - samples/sec: 7161.50 - lr: 0.000037 - momentum: 0.000000 2023-10-18 20:57:27,521 epoch 4 - iter 495/992 - loss 0.21280319 - time (sec): 11.11 - samples/sec: 7252.30 - lr: 0.000036 - momentum: 0.000000 2023-10-18 20:57:29,763 epoch 4 - iter 594/992 - loss 0.21029475 - time (sec): 13.36 - samples/sec: 7261.19 - lr: 0.000036 - momentum: 0.000000 2023-10-18 20:57:31,997 epoch 4 - iter 693/992 - loss 0.20859482 - time (sec): 15.59 - samples/sec: 7267.30 - lr: 0.000035 - momentum: 0.000000 2023-10-18 20:57:34,226 epoch 4 - iter 792/992 - loss 0.20972295 - time (sec): 17.82 - samples/sec: 7290.54 - lr: 0.000034 - momentum: 0.000000 2023-10-18 20:57:36,517 epoch 4 - iter 891/992 - loss 0.20628428 - time (sec): 20.11 - samples/sec: 7298.49 - lr: 0.000034 - momentum: 0.000000 2023-10-18 20:57:38,726 epoch 4 - iter 990/992 - loss 0.20626360 - time (sec): 22.32 - samples/sec: 7330.91 - lr: 0.000033 - momentum: 0.000000 2023-10-18 20:57:38,772 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:57:38,772 EPOCH 4 done: loss 0.2061 - lr: 0.000033 2023-10-18 20:57:40,612 DEV : loss 0.16479910910129547 - f1-score (micro avg) 0.4101 2023-10-18 20:57:40,630 saving best model 2023-10-18 20:57:40,666 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:57:42,943 epoch 5 - iter 99/992 - loss 0.17177104 - time (sec): 2.28 - samples/sec: 7353.67 - lr: 0.000033 - momentum: 0.000000 2023-10-18 20:57:45,125 epoch 5 - iter 198/992 - loss 0.18004855 - time (sec): 4.46 - samples/sec: 7278.40 - lr: 0.000032 - momentum: 0.000000 2023-10-18 20:57:47,298 epoch 5 - iter 297/992 - loss 0.18559241 - time (sec): 6.63 - samples/sec: 7192.47 - lr: 0.000032 - momentum: 0.000000 2023-10-18 20:57:49,624 epoch 5 - iter 396/992 - loss 0.18522087 - time (sec): 8.96 - samples/sec: 7199.80 - lr: 0.000031 - momentum: 0.000000 2023-10-18 20:57:51,819 epoch 5 - iter 495/992 - loss 0.18653077 - time (sec): 11.15 - samples/sec: 7216.72 - lr: 0.000031 - momentum: 0.000000 2023-10-18 20:57:54,038 epoch 5 - iter 594/992 - loss 0.18784304 - time (sec): 13.37 - samples/sec: 7224.68 - lr: 0.000030 - momentum: 0.000000 2023-10-18 20:57:56,273 epoch 5 - iter 693/992 - loss 0.18868299 - time (sec): 15.61 - samples/sec: 7255.64 - lr: 0.000029 - momentum: 0.000000 2023-10-18 20:57:58,607 epoch 5 - iter 792/992 - loss 0.18764218 - time (sec): 17.94 - samples/sec: 7244.21 - lr: 0.000029 - momentum: 0.000000 2023-10-18 20:58:00,866 epoch 5 - iter 891/992 - loss 0.18960836 - time (sec): 20.20 - samples/sec: 7275.43 - lr: 0.000028 - momentum: 0.000000 2023-10-18 20:58:03,136 epoch 5 - iter 990/992 - loss 0.19225274 - time (sec): 22.47 - samples/sec: 7281.35 - lr: 0.000028 - momentum: 0.000000 2023-10-18 20:58:03,191 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:58:03,191 EPOCH 5 done: loss 0.1923 - lr: 0.000028 2023-10-18 20:58:05,000 DEV : loss 0.15571151673793793 - f1-score (micro avg) 0.429 2023-10-18 20:58:05,018 saving best model 2023-10-18 20:58:05,051 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:58:07,306 epoch 6 - iter 99/992 - loss 0.17941164 - time (sec): 2.25 - samples/sec: 7240.83 - lr: 0.000027 - momentum: 0.000000 2023-10-18 20:58:09,525 epoch 6 - iter 198/992 - loss 0.17285384 - time (sec): 4.47 - samples/sec: 7150.39 - lr: 0.000027 - momentum: 0.000000 2023-10-18 20:58:11,791 epoch 6 - iter 297/992 - loss 0.17178395 - time (sec): 6.74 - samples/sec: 7272.32 - lr: 0.000026 - momentum: 0.000000 2023-10-18 20:58:14,001 epoch 6 - iter 396/992 - loss 0.17127824 - time (sec): 8.95 - samples/sec: 7269.39 - lr: 0.000026 - momentum: 0.000000 2023-10-18 20:58:16,238 epoch 6 - iter 495/992 - loss 0.17366195 - time (sec): 11.19 - samples/sec: 7241.92 - lr: 0.000025 - momentum: 0.000000 2023-10-18 20:58:18,505 epoch 6 - iter 594/992 - loss 0.17202885 - time (sec): 13.45 - samples/sec: 7246.43 - lr: 0.000024 - momentum: 0.000000 2023-10-18 20:58:20,637 epoch 6 - iter 693/992 - loss 0.17315227 - time (sec): 15.59 - samples/sec: 7274.96 - lr: 0.000024 - momentum: 0.000000 2023-10-18 20:58:22,852 epoch 6 - iter 792/992 - loss 0.17320411 - time (sec): 17.80 - samples/sec: 7271.31 - lr: 0.000023 - momentum: 0.000000 2023-10-18 20:58:25,041 epoch 6 - iter 891/992 - loss 0.17490211 - time (sec): 19.99 - samples/sec: 7285.57 - lr: 0.000023 - momentum: 0.000000 2023-10-18 20:58:27,314 epoch 6 - iter 990/992 - loss 0.17714492 - time (sec): 22.26 - samples/sec: 7351.83 - lr: 0.000022 - momentum: 0.000000 2023-10-18 20:58:27,362 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:58:27,363 EPOCH 6 done: loss 0.1770 - lr: 0.000022 2023-10-18 20:58:29,203 DEV : loss 0.15208245813846588 - f1-score (micro avg) 0.4522 2023-10-18 20:58:29,221 saving best model 2023-10-18 20:58:29,259 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:58:31,472 epoch 7 - iter 99/992 - loss 0.18307759 - time (sec): 2.21 - samples/sec: 7285.65 - lr: 0.000022 - momentum: 0.000000 2023-10-18 20:58:33,665 epoch 7 - iter 198/992 - loss 0.17513144 - time (sec): 4.41 - samples/sec: 7284.45 - lr: 0.000021 - momentum: 0.000000 2023-10-18 20:58:35,924 epoch 7 - iter 297/992 - loss 0.17363848 - time (sec): 6.66 - samples/sec: 7347.88 - lr: 0.000021 - momentum: 0.000000 2023-10-18 20:58:38,152 epoch 7 - iter 396/992 - loss 0.17849995 - time (sec): 8.89 - samples/sec: 7356.97 - lr: 0.000020 - momentum: 0.000000 2023-10-18 20:58:40,363 epoch 7 - iter 495/992 - loss 0.17302936 - time (sec): 11.10 - samples/sec: 7373.04 - lr: 0.000019 - momentum: 0.000000 2023-10-18 20:58:42,608 epoch 7 - iter 594/992 - loss 0.17173804 - time (sec): 13.35 - samples/sec: 7350.59 - lr: 0.000019 - momentum: 0.000000 2023-10-18 20:58:44,842 epoch 7 - iter 693/992 - loss 0.17176977 - time (sec): 15.58 - samples/sec: 7321.56 - lr: 0.000018 - momentum: 0.000000 2023-10-18 20:58:47,119 epoch 7 - iter 792/992 - loss 0.17139044 - time (sec): 17.86 - samples/sec: 7282.96 - lr: 0.000018 - momentum: 0.000000 2023-10-18 20:58:49,409 epoch 7 - iter 891/992 - loss 0.16779114 - time (sec): 20.15 - samples/sec: 7332.95 - lr: 0.000017 - momentum: 0.000000 2023-10-18 20:58:51,632 epoch 7 - iter 990/992 - loss 0.16865105 - time (sec): 22.37 - samples/sec: 7315.20 - lr: 0.000017 - momentum: 0.000000 2023-10-18 20:58:51,674 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:58:51,675 EPOCH 7 done: loss 0.1684 - lr: 0.000017 2023-10-18 20:58:53,875 DEV : loss 0.14835909008979797 - f1-score (micro avg) 0.4563 2023-10-18 20:58:53,893 saving best model 2023-10-18 20:58:53,930 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:58:56,137 epoch 8 - iter 99/992 - loss 0.16559554 - time (sec): 2.21 - samples/sec: 7584.76 - lr: 0.000016 - momentum: 0.000000 2023-10-18 20:58:58,402 epoch 8 - iter 198/992 - loss 0.15713025 - time (sec): 4.47 - samples/sec: 7401.59 - lr: 0.000016 - momentum: 0.000000 2023-10-18 20:59:00,698 epoch 8 - iter 297/992 - loss 0.15886284 - time (sec): 6.77 - samples/sec: 7505.02 - lr: 0.000015 - momentum: 0.000000 2023-10-18 20:59:02,941 epoch 8 - iter 396/992 - loss 0.16091614 - time (sec): 9.01 - samples/sec: 7381.27 - lr: 0.000014 - momentum: 0.000000 2023-10-18 20:59:05,171 epoch 8 - iter 495/992 - loss 0.15847892 - time (sec): 11.24 - samples/sec: 7290.06 - lr: 0.000014 - momentum: 0.000000 2023-10-18 20:59:07,413 epoch 8 - iter 594/992 - loss 0.16400434 - time (sec): 13.48 - samples/sec: 7279.96 - lr: 0.000013 - momentum: 0.000000 2023-10-18 20:59:09,638 epoch 8 - iter 693/992 - loss 0.16260905 - time (sec): 15.71 - samples/sec: 7267.37 - lr: 0.000013 - momentum: 0.000000 2023-10-18 20:59:11,837 epoch 8 - iter 792/992 - loss 0.16132712 - time (sec): 17.91 - samples/sec: 7275.69 - lr: 0.000012 - momentum: 0.000000 2023-10-18 20:59:14,069 epoch 8 - iter 891/992 - loss 0.16199039 - time (sec): 20.14 - samples/sec: 7290.01 - lr: 0.000012 - momentum: 0.000000 2023-10-18 20:59:16,290 epoch 8 - iter 990/992 - loss 0.16199479 - time (sec): 22.36 - samples/sec: 7317.37 - lr: 0.000011 - momentum: 0.000000 2023-10-18 20:59:16,333 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:59:16,334 EPOCH 8 done: loss 0.1618 - lr: 0.000011 2023-10-18 20:59:18,140 DEV : loss 0.14870117604732513 - f1-score (micro avg) 0.4715 2023-10-18 20:59:18,158 saving best model 2023-10-18 20:59:18,192 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:59:20,399 epoch 9 - iter 99/992 - loss 0.16182854 - time (sec): 2.21 - samples/sec: 7607.41 - lr: 0.000011 - momentum: 0.000000 2023-10-18 20:59:22,613 epoch 9 - iter 198/992 - loss 0.16350312 - time (sec): 4.42 - samples/sec: 7814.63 - lr: 0.000010 - momentum: 0.000000 2023-10-18 20:59:24,553 epoch 9 - iter 297/992 - loss 0.15746054 - time (sec): 6.36 - samples/sec: 7922.20 - lr: 0.000009 - momentum: 0.000000 2023-10-18 20:59:26,606 epoch 9 - iter 396/992 - loss 0.15577802 - time (sec): 8.41 - samples/sec: 7832.63 - lr: 0.000009 - momentum: 0.000000 2023-10-18 20:59:28,852 epoch 9 - iter 495/992 - loss 0.15518974 - time (sec): 10.66 - samples/sec: 7730.86 - lr: 0.000008 - momentum: 0.000000 2023-10-18 20:59:31,075 epoch 9 - iter 594/992 - loss 0.16021182 - time (sec): 12.88 - samples/sec: 7660.07 - lr: 0.000008 - momentum: 0.000000 2023-10-18 20:59:33,419 epoch 9 - iter 693/992 - loss 0.16018544 - time (sec): 15.23 - samples/sec: 7495.81 - lr: 0.000007 - momentum: 0.000000 2023-10-18 20:59:35,680 epoch 9 - iter 792/992 - loss 0.15779920 - time (sec): 17.49 - samples/sec: 7456.85 - lr: 0.000007 - momentum: 0.000000 2023-10-18 20:59:37,888 epoch 9 - iter 891/992 - loss 0.15767464 - time (sec): 19.70 - samples/sec: 7472.73 - lr: 0.000006 - momentum: 0.000000 2023-10-18 20:59:40,148 epoch 9 - iter 990/992 - loss 0.15624567 - time (sec): 21.96 - samples/sec: 7453.85 - lr: 0.000006 - momentum: 0.000000 2023-10-18 20:59:40,194 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:59:40,194 EPOCH 9 done: loss 0.1560 - lr: 0.000006 2023-10-18 20:59:42,035 DEV : loss 0.14903637766838074 - f1-score (micro avg) 0.4845 2023-10-18 20:59:42,054 saving best model 2023-10-18 20:59:42,087 ---------------------------------------------------------------------------------------------------- 2023-10-18 20:59:44,366 epoch 10 - iter 99/992 - loss 0.16504639 - time (sec): 2.28 - samples/sec: 6883.04 - lr: 0.000005 - momentum: 0.000000 2023-10-18 20:59:46,843 epoch 10 - iter 198/992 - loss 0.15098993 - time (sec): 4.76 - samples/sec: 6839.12 - lr: 0.000004 - momentum: 0.000000 2023-10-18 20:59:49,054 epoch 10 - iter 297/992 - loss 0.15199686 - time (sec): 6.97 - samples/sec: 7033.05 - lr: 0.000004 - momentum: 0.000000 2023-10-18 20:59:51,283 epoch 10 - iter 396/992 - loss 0.15271679 - time (sec): 9.19 - samples/sec: 7085.91 - lr: 0.000003 - momentum: 0.000000 2023-10-18 20:59:53,521 epoch 10 - iter 495/992 - loss 0.15018940 - time (sec): 11.43 - samples/sec: 7162.47 - lr: 0.000003 - momentum: 0.000000 2023-10-18 20:59:55,709 epoch 10 - iter 594/992 - loss 0.15324209 - time (sec): 13.62 - samples/sec: 7237.12 - lr: 0.000002 - momentum: 0.000000 2023-10-18 20:59:57,949 epoch 10 - iter 693/992 - loss 0.15234940 - time (sec): 15.86 - samples/sec: 7258.78 - lr: 0.000002 - momentum: 0.000000 2023-10-18 21:00:00,169 epoch 10 - iter 792/992 - loss 0.15171088 - time (sec): 18.08 - samples/sec: 7283.86 - lr: 0.000001 - momentum: 0.000000 2023-10-18 21:00:02,422 epoch 10 - iter 891/992 - loss 0.15226960 - time (sec): 20.33 - samples/sec: 7245.68 - lr: 0.000001 - momentum: 0.000000 2023-10-18 21:00:04,665 epoch 10 - iter 990/992 - loss 0.15307673 - time (sec): 22.58 - samples/sec: 7252.76 - lr: 0.000000 - momentum: 0.000000 2023-10-18 21:00:04,718 ---------------------------------------------------------------------------------------------------- 2023-10-18 21:00:04,718 EPOCH 10 done: loss 0.1532 - lr: 0.000000 2023-10-18 21:00:06,578 DEV : loss 0.1486775279045105 - f1-score (micro avg) 0.4917 2023-10-18 21:00:06,597 saving best model 2023-10-18 21:00:06,663 ---------------------------------------------------------------------------------------------------- 2023-10-18 21:00:06,664 Loading model from best epoch ... 2023-10-18 21:00:06,735 SequenceTagger predicts: Dictionary with 13 tags: O, S-PER, B-PER, E-PER, I-PER, S-LOC, B-LOC, E-LOC, I-LOC, S-ORG, B-ORG, E-ORG, I-ORG 2023-10-18 21:00:08,257 Results: - F-score (micro) 0.545 - F-score (macro) 0.3587 - Accuracy 0.4189 By class: precision recall f1-score support LOC 0.6909 0.6824 0.6866 655 PER 0.2964 0.5157 0.3764 223 ORG 0.0385 0.0079 0.0131 127 micro avg 0.5306 0.5602 0.5450 1005 macro avg 0.3419 0.4020 0.3587 1005 weighted avg 0.5209 0.5602 0.5327 1005 2023-10-18 21:00:08,257 ----------------------------------------------------------------------------------------------------