2023-10-16 12:45:52,143 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:45:52,144 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() )" 2023-10-16 12:45:52,144 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:45:52,145 MultiCorpus: 7142 train + 698 dev + 2570 test sentences - NER_HIPE_2022 Corpus: 7142 train + 698 dev + 2570 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fr/with_doc_seperator 2023-10-16 12:45:52,145 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:45:52,145 Train: 7142 sentences 2023-10-16 12:45:52,145 (train_with_dev=False, train_with_test=False) 2023-10-16 12:45:52,145 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:45:52,145 Training Params: 2023-10-16 12:45:52,145 - learning_rate: "3e-05" 2023-10-16 12:45:52,145 - mini_batch_size: "8" 2023-10-16 12:45:52,145 - max_epochs: "10" 2023-10-16 12:45:52,145 - shuffle: "True" 2023-10-16 12:45:52,145 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:45:52,145 Plugins: 2023-10-16 12:45:52,145 - LinearScheduler | warmup_fraction: '0.1' 2023-10-16 12:45:52,145 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:45:52,145 Final evaluation on model from best epoch (best-model.pt) 2023-10-16 12:45:52,145 - metric: "('micro avg', 'f1-score')" 2023-10-16 12:45:52,145 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:45:52,145 Computation: 2023-10-16 12:45:52,145 - compute on device: cuda:0 2023-10-16 12:45:52,145 - embedding storage: none 2023-10-16 12:45:52,145 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:45:52,145 Model training base path: "hmbench-newseye/fr-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4" 2023-10-16 12:45:52,145 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:45:52,145 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:46:00,431 epoch 1 - iter 89/893 - loss 2.55511339 - time (sec): 8.28 - samples/sec: 2909.27 - lr: 0.000003 - momentum: 0.000000 2023-10-16 12:46:07,327 epoch 1 - iter 178/893 - loss 1.61235074 - time (sec): 15.18 - samples/sec: 3224.15 - lr: 0.000006 - momentum: 0.000000 2023-10-16 12:46:14,109 epoch 1 - iter 267/893 - loss 1.21408701 - time (sec): 21.96 - samples/sec: 3363.25 - lr: 0.000009 - momentum: 0.000000 2023-10-16 12:46:21,076 epoch 1 - iter 356/893 - loss 0.98196317 - time (sec): 28.93 - samples/sec: 3425.09 - lr: 0.000012 - momentum: 0.000000 2023-10-16 12:46:27,573 epoch 1 - iter 445/893 - loss 0.84813604 - time (sec): 35.43 - samples/sec: 3461.42 - lr: 0.000015 - momentum: 0.000000 2023-10-16 12:46:34,458 epoch 1 - iter 534/893 - loss 0.73801609 - time (sec): 42.31 - samples/sec: 3505.89 - lr: 0.000018 - momentum: 0.000000 2023-10-16 12:46:41,293 epoch 1 - iter 623/893 - loss 0.65814262 - time (sec): 49.15 - samples/sec: 3539.49 - lr: 0.000021 - momentum: 0.000000 2023-10-16 12:46:48,032 epoch 1 - iter 712/893 - loss 0.60020660 - time (sec): 55.89 - samples/sec: 3546.66 - lr: 0.000024 - momentum: 0.000000 2023-10-16 12:46:54,998 epoch 1 - iter 801/893 - loss 0.55223454 - time (sec): 62.85 - samples/sec: 3546.82 - lr: 0.000027 - momentum: 0.000000 2023-10-16 12:47:01,870 epoch 1 - iter 890/893 - loss 0.51136141 - time (sec): 69.72 - samples/sec: 3557.52 - lr: 0.000030 - momentum: 0.000000 2023-10-16 12:47:02,083 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:47:02,083 EPOCH 1 done: loss 0.5100 - lr: 0.000030 2023-10-16 12:47:04,728 DEV : loss 0.11952730268239975 - f1-score (micro avg) 0.6946 2023-10-16 12:47:04,745 saving best model 2023-10-16 12:47:05,151 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:47:12,143 epoch 2 - iter 89/893 - loss 0.11614387 - time (sec): 6.99 - samples/sec: 3765.39 - lr: 0.000030 - momentum: 0.000000 2023-10-16 12:47:18,713 epoch 2 - iter 178/893 - loss 0.11703984 - time (sec): 13.56 - samples/sec: 3705.32 - lr: 0.000029 - momentum: 0.000000 2023-10-16 12:47:25,911 epoch 2 - iter 267/893 - loss 0.11593650 - time (sec): 20.76 - samples/sec: 3665.76 - lr: 0.000029 - momentum: 0.000000 2023-10-16 12:47:33,003 epoch 2 - iter 356/893 - loss 0.11898584 - time (sec): 27.85 - samples/sec: 3575.78 - lr: 0.000029 - momentum: 0.000000 2023-10-16 12:47:40,167 epoch 2 - iter 445/893 - loss 0.11380050 - time (sec): 35.01 - samples/sec: 3595.39 - lr: 0.000028 - momentum: 0.000000 2023-10-16 12:47:46,495 epoch 2 - iter 534/893 - loss 0.11094488 - time (sec): 41.34 - samples/sec: 3622.50 - lr: 0.000028 - momentum: 0.000000 2023-10-16 12:47:53,145 epoch 2 - iter 623/893 - loss 0.11041525 - time (sec): 47.99 - samples/sec: 3621.92 - lr: 0.000028 - momentum: 0.000000 2023-10-16 12:47:59,812 epoch 2 - iter 712/893 - loss 0.10964599 - time (sec): 54.66 - samples/sec: 3624.04 - lr: 0.000027 - momentum: 0.000000 2023-10-16 12:48:06,674 epoch 2 - iter 801/893 - loss 0.10815829 - time (sec): 61.52 - samples/sec: 3626.54 - lr: 0.000027 - momentum: 0.000000 2023-10-16 12:48:13,580 epoch 2 - iter 890/893 - loss 0.10593895 - time (sec): 68.43 - samples/sec: 3619.64 - lr: 0.000027 - momentum: 0.000000 2023-10-16 12:48:13,861 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:48:13,861 EPOCH 2 done: loss 0.1057 - lr: 0.000027 2023-10-16 12:48:18,043 DEV : loss 0.10355333238840103 - f1-score (micro avg) 0.7514 2023-10-16 12:48:18,059 saving best model 2023-10-16 12:48:18,614 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:48:25,574 epoch 3 - iter 89/893 - loss 0.06161232 - time (sec): 6.96 - samples/sec: 3465.77 - lr: 0.000026 - momentum: 0.000000 2023-10-16 12:48:33,414 epoch 3 - iter 178/893 - loss 0.05931401 - time (sec): 14.80 - samples/sec: 3535.00 - lr: 0.000026 - momentum: 0.000000 2023-10-16 12:48:40,156 epoch 3 - iter 267/893 - loss 0.06373319 - time (sec): 21.54 - samples/sec: 3554.90 - lr: 0.000026 - momentum: 0.000000 2023-10-16 12:48:46,889 epoch 3 - iter 356/893 - loss 0.06176266 - time (sec): 28.27 - samples/sec: 3583.15 - lr: 0.000025 - momentum: 0.000000 2023-10-16 12:48:53,244 epoch 3 - iter 445/893 - loss 0.06271250 - time (sec): 34.63 - samples/sec: 3585.52 - lr: 0.000025 - momentum: 0.000000 2023-10-16 12:48:59,780 epoch 3 - iter 534/893 - loss 0.06428874 - time (sec): 41.16 - samples/sec: 3599.68 - lr: 0.000025 - momentum: 0.000000 2023-10-16 12:49:06,534 epoch 3 - iter 623/893 - loss 0.06294223 - time (sec): 47.92 - samples/sec: 3638.07 - lr: 0.000024 - momentum: 0.000000 2023-10-16 12:49:12,998 epoch 3 - iter 712/893 - loss 0.06425486 - time (sec): 54.38 - samples/sec: 3652.85 - lr: 0.000024 - momentum: 0.000000 2023-10-16 12:49:19,311 epoch 3 - iter 801/893 - loss 0.06474582 - time (sec): 60.70 - samples/sec: 3655.96 - lr: 0.000024 - momentum: 0.000000 2023-10-16 12:49:26,634 epoch 3 - iter 890/893 - loss 0.06378154 - time (sec): 68.02 - samples/sec: 3646.66 - lr: 0.000023 - momentum: 0.000000 2023-10-16 12:49:26,907 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:49:26,907 EPOCH 3 done: loss 0.0640 - lr: 0.000023 2023-10-16 12:49:31,625 DEV : loss 0.12550972402095795 - f1-score (micro avg) 0.7835 2023-10-16 12:49:31,640 saving best model 2023-10-16 12:49:32,212 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:49:39,003 epoch 4 - iter 89/893 - loss 0.04612929 - time (sec): 6.79 - samples/sec: 3721.21 - lr: 0.000023 - momentum: 0.000000 2023-10-16 12:49:45,662 epoch 4 - iter 178/893 - loss 0.04670224 - time (sec): 13.45 - samples/sec: 3748.05 - lr: 0.000023 - momentum: 0.000000 2023-10-16 12:49:52,483 epoch 4 - iter 267/893 - loss 0.04673840 - time (sec): 20.27 - samples/sec: 3689.22 - lr: 0.000022 - momentum: 0.000000 2023-10-16 12:49:59,270 epoch 4 - iter 356/893 - loss 0.04627007 - time (sec): 27.06 - samples/sec: 3678.67 - lr: 0.000022 - momentum: 0.000000 2023-10-16 12:50:06,421 epoch 4 - iter 445/893 - loss 0.04530410 - time (sec): 34.21 - samples/sec: 3675.49 - lr: 0.000022 - momentum: 0.000000 2023-10-16 12:50:13,138 epoch 4 - iter 534/893 - loss 0.04603225 - time (sec): 40.92 - samples/sec: 3680.66 - lr: 0.000021 - momentum: 0.000000 2023-10-16 12:50:19,943 epoch 4 - iter 623/893 - loss 0.04600487 - time (sec): 47.73 - samples/sec: 3649.96 - lr: 0.000021 - momentum: 0.000000 2023-10-16 12:50:26,861 epoch 4 - iter 712/893 - loss 0.04696416 - time (sec): 54.65 - samples/sec: 3624.44 - lr: 0.000021 - momentum: 0.000000 2023-10-16 12:50:33,465 epoch 4 - iter 801/893 - loss 0.04647850 - time (sec): 61.25 - samples/sec: 3635.01 - lr: 0.000020 - momentum: 0.000000 2023-10-16 12:50:40,628 epoch 4 - iter 890/893 - loss 0.04696665 - time (sec): 68.41 - samples/sec: 3626.05 - lr: 0.000020 - momentum: 0.000000 2023-10-16 12:50:40,819 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:50:40,819 EPOCH 4 done: loss 0.0469 - lr: 0.000020 2023-10-16 12:50:45,113 DEV : loss 0.14134925603866577 - f1-score (micro avg) 0.7836 2023-10-16 12:50:45,130 saving best model 2023-10-16 12:50:45,700 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:50:52,682 epoch 5 - iter 89/893 - loss 0.02860295 - time (sec): 6.98 - samples/sec: 3752.60 - lr: 0.000020 - momentum: 0.000000 2023-10-16 12:50:59,524 epoch 5 - iter 178/893 - loss 0.03232203 - time (sec): 13.82 - samples/sec: 3696.27 - lr: 0.000019 - momentum: 0.000000 2023-10-16 12:51:06,414 epoch 5 - iter 267/893 - loss 0.03327893 - time (sec): 20.71 - samples/sec: 3549.51 - lr: 0.000019 - momentum: 0.000000 2023-10-16 12:51:13,137 epoch 5 - iter 356/893 - loss 0.03240572 - time (sec): 27.44 - samples/sec: 3555.85 - lr: 0.000019 - momentum: 0.000000 2023-10-16 12:51:20,796 epoch 5 - iter 445/893 - loss 0.03319621 - time (sec): 35.09 - samples/sec: 3560.42 - lr: 0.000018 - momentum: 0.000000 2023-10-16 12:51:27,811 epoch 5 - iter 534/893 - loss 0.03409730 - time (sec): 42.11 - samples/sec: 3572.12 - lr: 0.000018 - momentum: 0.000000 2023-10-16 12:51:34,927 epoch 5 - iter 623/893 - loss 0.03454180 - time (sec): 49.23 - samples/sec: 3592.93 - lr: 0.000018 - momentum: 0.000000 2023-10-16 12:51:41,334 epoch 5 - iter 712/893 - loss 0.03417524 - time (sec): 55.63 - samples/sec: 3589.84 - lr: 0.000017 - momentum: 0.000000 2023-10-16 12:51:48,137 epoch 5 - iter 801/893 - loss 0.03472578 - time (sec): 62.44 - samples/sec: 3578.49 - lr: 0.000017 - momentum: 0.000000 2023-10-16 12:51:54,801 epoch 5 - iter 890/893 - loss 0.03490357 - time (sec): 69.10 - samples/sec: 3588.45 - lr: 0.000017 - momentum: 0.000000 2023-10-16 12:51:55,044 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:51:55,044 EPOCH 5 done: loss 0.0348 - lr: 0.000017 2023-10-16 12:51:59,372 DEV : loss 0.17483924329280853 - f1-score (micro avg) 0.7945 2023-10-16 12:51:59,392 saving best model 2023-10-16 12:51:59,954 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:52:06,819 epoch 6 - iter 89/893 - loss 0.02626343 - time (sec): 6.86 - samples/sec: 3670.10 - lr: 0.000016 - momentum: 0.000000 2023-10-16 12:52:13,623 epoch 6 - iter 178/893 - loss 0.03047495 - time (sec): 13.67 - samples/sec: 3610.14 - lr: 0.000016 - momentum: 0.000000 2023-10-16 12:52:20,194 epoch 6 - iter 267/893 - loss 0.03214194 - time (sec): 20.24 - samples/sec: 3673.48 - lr: 0.000016 - momentum: 0.000000 2023-10-16 12:52:27,056 epoch 6 - iter 356/893 - loss 0.03185153 - time (sec): 27.10 - samples/sec: 3671.57 - lr: 0.000015 - momentum: 0.000000 2023-10-16 12:52:34,130 epoch 6 - iter 445/893 - loss 0.02932483 - time (sec): 34.17 - samples/sec: 3672.16 - lr: 0.000015 - momentum: 0.000000 2023-10-16 12:52:40,891 epoch 6 - iter 534/893 - loss 0.02993883 - time (sec): 40.93 - samples/sec: 3659.20 - lr: 0.000015 - momentum: 0.000000 2023-10-16 12:52:47,512 epoch 6 - iter 623/893 - loss 0.02884411 - time (sec): 47.56 - samples/sec: 3683.06 - lr: 0.000014 - momentum: 0.000000 2023-10-16 12:52:54,195 epoch 6 - iter 712/893 - loss 0.02850536 - time (sec): 54.24 - samples/sec: 3668.78 - lr: 0.000014 - momentum: 0.000000 2023-10-16 12:53:01,056 epoch 6 - iter 801/893 - loss 0.02830873 - time (sec): 61.10 - samples/sec: 3657.82 - lr: 0.000014 - momentum: 0.000000 2023-10-16 12:53:08,041 epoch 6 - iter 890/893 - loss 0.02857723 - time (sec): 68.08 - samples/sec: 3645.85 - lr: 0.000013 - momentum: 0.000000 2023-10-16 12:53:08,211 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:53:08,212 EPOCH 6 done: loss 0.0287 - lr: 0.000013 2023-10-16 12:53:12,859 DEV : loss 0.19940702617168427 - f1-score (micro avg) 0.7727 2023-10-16 12:53:12,875 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:53:19,596 epoch 7 - iter 89/893 - loss 0.02500460 - time (sec): 6.72 - samples/sec: 3541.40 - lr: 0.000013 - momentum: 0.000000 2023-10-16 12:53:26,259 epoch 7 - iter 178/893 - loss 0.02277466 - time (sec): 13.38 - samples/sec: 3591.82 - lr: 0.000013 - momentum: 0.000000 2023-10-16 12:53:33,085 epoch 7 - iter 267/893 - loss 0.02090527 - time (sec): 20.21 - samples/sec: 3616.67 - lr: 0.000012 - momentum: 0.000000 2023-10-16 12:53:39,862 epoch 7 - iter 356/893 - loss 0.02011475 - time (sec): 26.99 - samples/sec: 3636.71 - lr: 0.000012 - momentum: 0.000000 2023-10-16 12:53:46,751 epoch 7 - iter 445/893 - loss 0.02069675 - time (sec): 33.88 - samples/sec: 3629.54 - lr: 0.000012 - momentum: 0.000000 2023-10-16 12:53:53,473 epoch 7 - iter 534/893 - loss 0.02079766 - time (sec): 40.60 - samples/sec: 3626.98 - lr: 0.000011 - momentum: 0.000000 2023-10-16 12:54:00,256 epoch 7 - iter 623/893 - loss 0.02128472 - time (sec): 47.38 - samples/sec: 3641.22 - lr: 0.000011 - momentum: 0.000000 2023-10-16 12:54:07,283 epoch 7 - iter 712/893 - loss 0.02254890 - time (sec): 54.41 - samples/sec: 3648.20 - lr: 0.000011 - momentum: 0.000000 2023-10-16 12:54:13,922 epoch 7 - iter 801/893 - loss 0.02269062 - time (sec): 61.05 - samples/sec: 3639.12 - lr: 0.000010 - momentum: 0.000000 2023-10-16 12:54:20,953 epoch 7 - iter 890/893 - loss 0.02194304 - time (sec): 68.08 - samples/sec: 3640.63 - lr: 0.000010 - momentum: 0.000000 2023-10-16 12:54:21,171 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:54:21,171 EPOCH 7 done: loss 0.0220 - lr: 0.000010 2023-10-16 12:54:25,732 DEV : loss 0.19999033212661743 - f1-score (micro avg) 0.7912 2023-10-16 12:54:25,755 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:54:32,512 epoch 8 - iter 89/893 - loss 0.01168160 - time (sec): 6.76 - samples/sec: 3588.80 - lr: 0.000010 - momentum: 0.000000 2023-10-16 12:54:39,559 epoch 8 - iter 178/893 - loss 0.01427644 - time (sec): 13.80 - samples/sec: 3476.32 - lr: 0.000009 - momentum: 0.000000 2023-10-16 12:54:46,333 epoch 8 - iter 267/893 - loss 0.01591158 - time (sec): 20.58 - samples/sec: 3543.59 - lr: 0.000009 - momentum: 0.000000 2023-10-16 12:54:53,376 epoch 8 - iter 356/893 - loss 0.01732768 - time (sec): 27.62 - samples/sec: 3605.76 - lr: 0.000009 - momentum: 0.000000 2023-10-16 12:55:00,019 epoch 8 - iter 445/893 - loss 0.01967539 - time (sec): 34.26 - samples/sec: 3629.85 - lr: 0.000008 - momentum: 0.000000 2023-10-16 12:55:07,011 epoch 8 - iter 534/893 - loss 0.01853002 - time (sec): 41.25 - samples/sec: 3637.34 - lr: 0.000008 - momentum: 0.000000 2023-10-16 12:55:13,914 epoch 8 - iter 623/893 - loss 0.01808121 - time (sec): 48.16 - samples/sec: 3620.58 - lr: 0.000008 - momentum: 0.000000 2023-10-16 12:55:20,357 epoch 8 - iter 712/893 - loss 0.01781687 - time (sec): 54.60 - samples/sec: 3625.12 - lr: 0.000007 - momentum: 0.000000 2023-10-16 12:55:27,362 epoch 8 - iter 801/893 - loss 0.01794196 - time (sec): 61.61 - samples/sec: 3612.50 - lr: 0.000007 - momentum: 0.000000 2023-10-16 12:55:34,282 epoch 8 - iter 890/893 - loss 0.01754005 - time (sec): 68.53 - samples/sec: 3621.46 - lr: 0.000007 - momentum: 0.000000 2023-10-16 12:55:34,488 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:55:34,489 EPOCH 8 done: loss 0.0175 - lr: 0.000007 2023-10-16 12:55:38,684 DEV : loss 0.20922400057315826 - f1-score (micro avg) 0.8038 2023-10-16 12:55:38,700 saving best model 2023-10-16 12:55:39,186 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:55:45,857 epoch 9 - iter 89/893 - loss 0.00982904 - time (sec): 6.67 - samples/sec: 3621.30 - lr: 0.000006 - momentum: 0.000000 2023-10-16 12:55:52,816 epoch 9 - iter 178/893 - loss 0.01033872 - time (sec): 13.63 - samples/sec: 3702.90 - lr: 0.000006 - momentum: 0.000000 2023-10-16 12:56:00,091 epoch 9 - iter 267/893 - loss 0.01214149 - time (sec): 20.90 - samples/sec: 3649.27 - lr: 0.000006 - momentum: 0.000000 2023-10-16 12:56:06,660 epoch 9 - iter 356/893 - loss 0.01195331 - time (sec): 27.47 - samples/sec: 3654.09 - lr: 0.000005 - momentum: 0.000000 2023-10-16 12:56:13,921 epoch 9 - iter 445/893 - loss 0.01203697 - time (sec): 34.73 - samples/sec: 3632.41 - lr: 0.000005 - momentum: 0.000000 2023-10-16 12:56:20,404 epoch 9 - iter 534/893 - loss 0.01189900 - time (sec): 41.22 - samples/sec: 3644.72 - lr: 0.000005 - momentum: 0.000000 2023-10-16 12:56:27,229 epoch 9 - iter 623/893 - loss 0.01250389 - time (sec): 48.04 - samples/sec: 3649.49 - lr: 0.000004 - momentum: 0.000000 2023-10-16 12:56:34,221 epoch 9 - iter 712/893 - loss 0.01230678 - time (sec): 55.03 - samples/sec: 3640.52 - lr: 0.000004 - momentum: 0.000000 2023-10-16 12:56:40,816 epoch 9 - iter 801/893 - loss 0.01224281 - time (sec): 61.63 - samples/sec: 3637.30 - lr: 0.000004 - momentum: 0.000000 2023-10-16 12:56:47,523 epoch 9 - iter 890/893 - loss 0.01274520 - time (sec): 68.34 - samples/sec: 3631.08 - lr: 0.000003 - momentum: 0.000000 2023-10-16 12:56:47,696 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:56:47,696 EPOCH 9 done: loss 0.0128 - lr: 0.000003 2023-10-16 12:56:52,288 DEV : loss 0.21604780852794647 - f1-score (micro avg) 0.7879 2023-10-16 12:56:52,314 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:56:59,282 epoch 10 - iter 89/893 - loss 0.00670445 - time (sec): 6.97 - samples/sec: 3448.15 - lr: 0.000003 - momentum: 0.000000 2023-10-16 12:57:05,865 epoch 10 - iter 178/893 - loss 0.00793855 - time (sec): 13.55 - samples/sec: 3566.79 - lr: 0.000003 - momentum: 0.000000 2023-10-16 12:57:12,745 epoch 10 - iter 267/893 - loss 0.00788264 - time (sec): 20.43 - samples/sec: 3594.85 - lr: 0.000002 - momentum: 0.000000 2023-10-16 12:57:19,300 epoch 10 - iter 356/893 - loss 0.00725310 - time (sec): 26.98 - samples/sec: 3573.51 - lr: 0.000002 - momentum: 0.000000 2023-10-16 12:57:26,244 epoch 10 - iter 445/893 - loss 0.00808299 - time (sec): 33.93 - samples/sec: 3599.24 - lr: 0.000002 - momentum: 0.000000 2023-10-16 12:57:32,839 epoch 10 - iter 534/893 - loss 0.00893934 - time (sec): 40.52 - samples/sec: 3616.94 - lr: 0.000001 - momentum: 0.000000 2023-10-16 12:57:39,669 epoch 10 - iter 623/893 - loss 0.00976410 - time (sec): 47.35 - samples/sec: 3625.68 - lr: 0.000001 - momentum: 0.000000 2023-10-16 12:57:46,950 epoch 10 - iter 712/893 - loss 0.01000539 - time (sec): 54.63 - samples/sec: 3628.40 - lr: 0.000001 - momentum: 0.000000 2023-10-16 12:57:54,074 epoch 10 - iter 801/893 - loss 0.00983205 - time (sec): 61.76 - samples/sec: 3622.71 - lr: 0.000000 - momentum: 0.000000 2023-10-16 12:58:00,863 epoch 10 - iter 890/893 - loss 0.00969370 - time (sec): 68.55 - samples/sec: 3621.54 - lr: 0.000000 - momentum: 0.000000 2023-10-16 12:58:01,015 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:58:01,015 EPOCH 10 done: loss 0.0097 - lr: 0.000000 2023-10-16 12:58:05,669 DEV : loss 0.2217082679271698 - f1-score (micro avg) 0.7952 2023-10-16 12:58:06,080 ---------------------------------------------------------------------------------------------------- 2023-10-16 12:58:06,081 Loading model from best epoch ... 2023-10-16 12:58:07,923 SequenceTagger predicts: Dictionary with 17 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, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd 2023-10-16 12:58:17,287 Results: - F-score (micro) 0.6906 - F-score (macro) 0.6137 - Accuracy 0.5455 By class: precision recall f1-score support LOC 0.7176 0.6868 0.7018 1095 PER 0.7696 0.7658 0.7677 1012 ORG 0.4312 0.5882 0.4976 357 HumanProd 0.4082 0.6061 0.4878 33 micro avg 0.6781 0.7036 0.6906 2497 macro avg 0.5816 0.6617 0.6137 2497 weighted avg 0.6936 0.7036 0.6965 2497 2023-10-16 12:58:17,287 ----------------------------------------------------------------------------------------------------