2023-10-25 14:22:05,423 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:22:05,424 Model: "SequenceTagger( (embeddings): TransformerWordEmbeddings( (model): BertModel( (embeddings): BertEmbeddings( (word_embeddings): Embedding(64001, 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() )" 2023-10-25 14:22:05,424 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:22:05,424 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 2023-10-25 14:22:05,424 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:22:05,424 Train: 7142 sentences 2023-10-25 14:22:05,424 (train_with_dev=False, train_with_test=False) 2023-10-25 14:22:05,424 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:22:05,424 Training Params: 2023-10-25 14:22:05,425 - learning_rate: "5e-05" 2023-10-25 14:22:05,425 - mini_batch_size: "8" 2023-10-25 14:22:05,425 - max_epochs: "10" 2023-10-25 14:22:05,425 - shuffle: "True" 2023-10-25 14:22:05,425 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:22:05,425 Plugins: 2023-10-25 14:22:05,425 - TensorboardLogger 2023-10-25 14:22:05,425 - LinearScheduler | warmup_fraction: '0.1' 2023-10-25 14:22:05,425 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:22:05,425 Final evaluation on model from best epoch (best-model.pt) 2023-10-25 14:22:05,425 - metric: "('micro avg', 'f1-score')" 2023-10-25 14:22:05,425 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:22:05,425 Computation: 2023-10-25 14:22:05,425 - compute on device: cuda:0 2023-10-25 14:22:05,425 - embedding storage: none 2023-10-25 14:22:05,425 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:22:05,425 Model training base path: "hmbench-newseye/fr-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1" 2023-10-25 14:22:05,425 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:22:05,425 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:22:05,425 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-25 14:22:11,296 epoch 1 - iter 89/893 - loss 2.22174821 - time (sec): 5.87 - samples/sec: 4255.73 - lr: 0.000005 - momentum: 0.000000 2023-10-25 14:22:17,176 epoch 1 - iter 178/893 - loss 1.34900675 - time (sec): 11.75 - samples/sec: 4297.23 - lr: 0.000010 - momentum: 0.000000 2023-10-25 14:22:23,077 epoch 1 - iter 267/893 - loss 1.01310745 - time (sec): 17.65 - samples/sec: 4246.13 - lr: 0.000015 - momentum: 0.000000 2023-10-25 14:22:28,941 epoch 1 - iter 356/893 - loss 0.81673971 - time (sec): 23.52 - samples/sec: 4266.25 - lr: 0.000020 - momentum: 0.000000 2023-10-25 14:22:34,652 epoch 1 - iter 445/893 - loss 0.69153729 - time (sec): 29.23 - samples/sec: 4280.94 - lr: 0.000025 - momentum: 0.000000 2023-10-25 14:22:40,228 epoch 1 - iter 534/893 - loss 0.60855944 - time (sec): 34.80 - samples/sec: 4283.75 - lr: 0.000030 - momentum: 0.000000 2023-10-25 14:22:46,157 epoch 1 - iter 623/893 - loss 0.53908676 - time (sec): 40.73 - samples/sec: 4303.65 - lr: 0.000035 - momentum: 0.000000 2023-10-25 14:22:51,797 epoch 1 - iter 712/893 - loss 0.49305012 - time (sec): 46.37 - samples/sec: 4290.84 - lr: 0.000040 - momentum: 0.000000 2023-10-25 14:22:57,456 epoch 1 - iter 801/893 - loss 0.45348166 - time (sec): 52.03 - samples/sec: 4296.11 - lr: 0.000045 - momentum: 0.000000 2023-10-25 14:23:03,344 epoch 1 - iter 890/893 - loss 0.42527446 - time (sec): 57.92 - samples/sec: 4279.37 - lr: 0.000050 - momentum: 0.000000 2023-10-25 14:23:03,553 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:23:03,554 EPOCH 1 done: loss 0.4241 - lr: 0.000050 2023-10-25 14:23:07,042 DEV : loss 0.10224700719118118 - f1-score (micro avg) 0.7363 2023-10-25 14:23:07,064 saving best model 2023-10-25 14:23:07,584 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:23:13,537 epoch 2 - iter 89/893 - loss 0.10807083 - time (sec): 5.95 - samples/sec: 4146.32 - lr: 0.000049 - momentum: 0.000000 2023-10-25 14:23:19,506 epoch 2 - iter 178/893 - loss 0.10717817 - time (sec): 11.92 - samples/sec: 4148.68 - lr: 0.000049 - momentum: 0.000000 2023-10-25 14:23:25,279 epoch 2 - iter 267/893 - loss 0.10936825 - time (sec): 17.69 - samples/sec: 4151.97 - lr: 0.000048 - momentum: 0.000000 2023-10-25 14:23:31,126 epoch 2 - iter 356/893 - loss 0.11237855 - time (sec): 23.54 - samples/sec: 4166.50 - lr: 0.000048 - momentum: 0.000000 2023-10-25 14:23:36,586 epoch 2 - iter 445/893 - loss 0.10585519 - time (sec): 29.00 - samples/sec: 4220.66 - lr: 0.000047 - momentum: 0.000000 2023-10-25 14:23:42,078 epoch 2 - iter 534/893 - loss 0.10593438 - time (sec): 34.49 - samples/sec: 4248.78 - lr: 0.000047 - momentum: 0.000000 2023-10-25 14:23:47,781 epoch 2 - iter 623/893 - loss 0.10798060 - time (sec): 40.20 - samples/sec: 4281.71 - lr: 0.000046 - momentum: 0.000000 2023-10-25 14:23:53,567 epoch 2 - iter 712/893 - loss 0.10704577 - time (sec): 45.98 - samples/sec: 4307.87 - lr: 0.000046 - momentum: 0.000000 2023-10-25 14:23:59,120 epoch 2 - iter 801/893 - loss 0.10429214 - time (sec): 51.53 - samples/sec: 4347.28 - lr: 0.000045 - momentum: 0.000000 2023-10-25 14:24:04,739 epoch 2 - iter 890/893 - loss 0.10359464 - time (sec): 57.15 - samples/sec: 4337.39 - lr: 0.000044 - momentum: 0.000000 2023-10-25 14:24:04,955 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:24:04,955 EPOCH 2 done: loss 0.1033 - lr: 0.000044 2023-10-25 14:24:10,135 DEV : loss 0.0969092845916748 - f1-score (micro avg) 0.7786 2023-10-25 14:24:10,157 saving best model 2023-10-25 14:24:10,855 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:24:16,400 epoch 3 - iter 89/893 - loss 0.05486912 - time (sec): 5.54 - samples/sec: 4434.09 - lr: 0.000044 - momentum: 0.000000 2023-10-25 14:24:22,119 epoch 3 - iter 178/893 - loss 0.06296271 - time (sec): 11.26 - samples/sec: 4548.70 - lr: 0.000043 - momentum: 0.000000 2023-10-25 14:24:27,693 epoch 3 - iter 267/893 - loss 0.06237710 - time (sec): 16.83 - samples/sec: 4480.75 - lr: 0.000043 - momentum: 0.000000 2023-10-25 14:24:33,190 epoch 3 - iter 356/893 - loss 0.07078572 - time (sec): 22.33 - samples/sec: 4456.89 - lr: 0.000042 - momentum: 0.000000 2023-10-25 14:24:38,702 epoch 3 - iter 445/893 - loss 0.06966703 - time (sec): 27.84 - samples/sec: 4421.33 - lr: 0.000042 - momentum: 0.000000 2023-10-25 14:24:44,432 epoch 3 - iter 534/893 - loss 0.06853292 - time (sec): 33.57 - samples/sec: 4407.07 - lr: 0.000041 - momentum: 0.000000 2023-10-25 14:24:50,191 epoch 3 - iter 623/893 - loss 0.06693004 - time (sec): 39.33 - samples/sec: 4415.91 - lr: 0.000041 - momentum: 0.000000 2023-10-25 14:24:55,886 epoch 3 - iter 712/893 - loss 0.06730109 - time (sec): 45.03 - samples/sec: 4413.01 - lr: 0.000040 - momentum: 0.000000 2023-10-25 14:25:01,340 epoch 3 - iter 801/893 - loss 0.06610240 - time (sec): 50.48 - samples/sec: 4414.05 - lr: 0.000039 - momentum: 0.000000 2023-10-25 14:25:06,891 epoch 3 - iter 890/893 - loss 0.06743181 - time (sec): 56.03 - samples/sec: 4428.50 - lr: 0.000039 - momentum: 0.000000 2023-10-25 14:25:07,061 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:25:07,061 EPOCH 3 done: loss 0.0674 - lr: 0.000039 2023-10-25 14:25:10,865 DEV : loss 0.11728406697511673 - f1-score (micro avg) 0.7507 2023-10-25 14:25:10,887 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:25:16,941 epoch 4 - iter 89/893 - loss 0.05172042 - time (sec): 6.05 - samples/sec: 4265.61 - lr: 0.000038 - momentum: 0.000000 2023-10-25 14:25:22,824 epoch 4 - iter 178/893 - loss 0.04890968 - time (sec): 11.93 - samples/sec: 4188.03 - lr: 0.000038 - momentum: 0.000000 2023-10-25 14:25:28,627 epoch 4 - iter 267/893 - loss 0.04835057 - time (sec): 17.74 - samples/sec: 4153.56 - lr: 0.000037 - momentum: 0.000000 2023-10-25 14:25:34,625 epoch 4 - iter 356/893 - loss 0.04648503 - time (sec): 23.74 - samples/sec: 4206.05 - lr: 0.000037 - momentum: 0.000000 2023-10-25 14:25:41,203 epoch 4 - iter 445/893 - loss 0.04728395 - time (sec): 30.31 - samples/sec: 4135.62 - lr: 0.000036 - momentum: 0.000000 2023-10-25 14:25:46,943 epoch 4 - iter 534/893 - loss 0.04862880 - time (sec): 36.05 - samples/sec: 4139.07 - lr: 0.000036 - momentum: 0.000000 2023-10-25 14:25:52,737 epoch 4 - iter 623/893 - loss 0.04803766 - time (sec): 41.85 - samples/sec: 4149.38 - lr: 0.000035 - momentum: 0.000000 2023-10-25 14:25:58,381 epoch 4 - iter 712/893 - loss 0.04864147 - time (sec): 47.49 - samples/sec: 4175.34 - lr: 0.000034 - momentum: 0.000000 2023-10-25 14:26:04,318 epoch 4 - iter 801/893 - loss 0.04871377 - time (sec): 53.43 - samples/sec: 4193.31 - lr: 0.000034 - momentum: 0.000000 2023-10-25 14:26:09,795 epoch 4 - iter 890/893 - loss 0.04850206 - time (sec): 58.91 - samples/sec: 4210.00 - lr: 0.000033 - momentum: 0.000000 2023-10-25 14:26:09,973 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:26:09,973 EPOCH 4 done: loss 0.0486 - lr: 0.000033 2023-10-25 14:26:14,041 DEV : loss 0.15373246371746063 - f1-score (micro avg) 0.7653 2023-10-25 14:26:14,065 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:26:20,047 epoch 5 - iter 89/893 - loss 0.03855375 - time (sec): 5.98 - samples/sec: 4417.22 - lr: 0.000033 - momentum: 0.000000 2023-10-25 14:26:25,827 epoch 5 - iter 178/893 - loss 0.04024206 - time (sec): 11.76 - samples/sec: 4380.57 - lr: 0.000032 - momentum: 0.000000 2023-10-25 14:26:31,615 epoch 5 - iter 267/893 - loss 0.03624782 - time (sec): 17.55 - samples/sec: 4351.23 - lr: 0.000032 - momentum: 0.000000 2023-10-25 14:26:37,230 epoch 5 - iter 356/893 - loss 0.03803371 - time (sec): 23.16 - samples/sec: 4321.77 - lr: 0.000031 - momentum: 0.000000 2023-10-25 14:26:42,808 epoch 5 - iter 445/893 - loss 0.03757448 - time (sec): 28.74 - samples/sec: 4285.27 - lr: 0.000031 - momentum: 0.000000 2023-10-25 14:26:48,363 epoch 5 - iter 534/893 - loss 0.03668820 - time (sec): 34.30 - samples/sec: 4293.91 - lr: 0.000030 - momentum: 0.000000 2023-10-25 14:26:53,894 epoch 5 - iter 623/893 - loss 0.03705404 - time (sec): 39.83 - samples/sec: 4274.67 - lr: 0.000029 - momentum: 0.000000 2023-10-25 14:26:59,729 epoch 5 - iter 712/893 - loss 0.03634319 - time (sec): 45.66 - samples/sec: 4325.05 - lr: 0.000029 - momentum: 0.000000 2023-10-25 14:27:05,488 epoch 5 - iter 801/893 - loss 0.03547607 - time (sec): 51.42 - samples/sec: 4340.40 - lr: 0.000028 - momentum: 0.000000 2023-10-25 14:27:10,939 epoch 5 - iter 890/893 - loss 0.03563294 - time (sec): 56.87 - samples/sec: 4360.64 - lr: 0.000028 - momentum: 0.000000 2023-10-25 14:27:11,109 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:27:11,109 EPOCH 5 done: loss 0.0355 - lr: 0.000028 2023-10-25 14:27:15,806 DEV : loss 0.16320814192295074 - f1-score (micro avg) 0.7874 2023-10-25 14:27:15,826 saving best model 2023-10-25 14:27:16,529 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:27:22,218 epoch 6 - iter 89/893 - loss 0.02617113 - time (sec): 5.69 - samples/sec: 4700.94 - lr: 0.000027 - momentum: 0.000000 2023-10-25 14:27:28,467 epoch 6 - iter 178/893 - loss 0.02520431 - time (sec): 11.94 - samples/sec: 4332.41 - lr: 0.000027 - momentum: 0.000000 2023-10-25 14:27:33,984 epoch 6 - iter 267/893 - loss 0.02638010 - time (sec): 17.45 - samples/sec: 4320.59 - lr: 0.000026 - momentum: 0.000000 2023-10-25 14:27:39,878 epoch 6 - iter 356/893 - loss 0.02643239 - time (sec): 23.35 - samples/sec: 4349.87 - lr: 0.000026 - momentum: 0.000000 2023-10-25 14:27:45,656 epoch 6 - iter 445/893 - loss 0.02697284 - time (sec): 29.13 - samples/sec: 4332.53 - lr: 0.000025 - momentum: 0.000000 2023-10-25 14:27:51,208 epoch 6 - iter 534/893 - loss 0.02717458 - time (sec): 34.68 - samples/sec: 4355.89 - lr: 0.000024 - momentum: 0.000000 2023-10-25 14:27:56,893 epoch 6 - iter 623/893 - loss 0.02638170 - time (sec): 40.36 - samples/sec: 4340.97 - lr: 0.000024 - momentum: 0.000000 2023-10-25 14:28:02,464 epoch 6 - iter 712/893 - loss 0.02646998 - time (sec): 45.93 - samples/sec: 4345.70 - lr: 0.000023 - momentum: 0.000000 2023-10-25 14:28:08,304 epoch 6 - iter 801/893 - loss 0.02574741 - time (sec): 51.77 - samples/sec: 4344.49 - lr: 0.000023 - momentum: 0.000000 2023-10-25 14:28:13,959 epoch 6 - iter 890/893 - loss 0.02590536 - time (sec): 57.43 - samples/sec: 4318.36 - lr: 0.000022 - momentum: 0.000000 2023-10-25 14:28:14,125 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:28:14,126 EPOCH 6 done: loss 0.0259 - lr: 0.000022 2023-10-25 14:28:17,933 DEV : loss 0.18096403777599335 - f1-score (micro avg) 0.7885 2023-10-25 14:28:17,953 saving best model 2023-10-25 14:28:18,694 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:28:25,439 epoch 7 - iter 89/893 - loss 0.01678159 - time (sec): 6.74 - samples/sec: 3409.22 - lr: 0.000022 - momentum: 0.000000 2023-10-25 14:28:31,190 epoch 7 - iter 178/893 - loss 0.02086138 - time (sec): 12.49 - samples/sec: 3832.17 - lr: 0.000021 - momentum: 0.000000 2023-10-25 14:28:37,291 epoch 7 - iter 267/893 - loss 0.02080217 - time (sec): 18.59 - samples/sec: 4067.89 - lr: 0.000021 - momentum: 0.000000 2023-10-25 14:28:43,102 epoch 7 - iter 356/893 - loss 0.02143983 - time (sec): 24.40 - samples/sec: 4179.14 - lr: 0.000020 - momentum: 0.000000 2023-10-25 14:28:48,654 epoch 7 - iter 445/893 - loss 0.02119625 - time (sec): 29.96 - samples/sec: 4154.44 - lr: 0.000019 - momentum: 0.000000 2023-10-25 14:28:54,735 epoch 7 - iter 534/893 - loss 0.02007361 - time (sec): 36.04 - samples/sec: 4119.27 - lr: 0.000019 - momentum: 0.000000 2023-10-25 14:29:00,403 epoch 7 - iter 623/893 - loss 0.02020608 - time (sec): 41.71 - samples/sec: 4157.17 - lr: 0.000018 - momentum: 0.000000 2023-10-25 14:29:06,389 epoch 7 - iter 712/893 - loss 0.02032914 - time (sec): 47.69 - samples/sec: 4157.75 - lr: 0.000018 - momentum: 0.000000 2023-10-25 14:29:12,245 epoch 7 - iter 801/893 - loss 0.02130873 - time (sec): 53.55 - samples/sec: 4154.56 - lr: 0.000017 - momentum: 0.000000 2023-10-25 14:29:18,268 epoch 7 - iter 890/893 - loss 0.02140375 - time (sec): 59.57 - samples/sec: 4164.21 - lr: 0.000017 - momentum: 0.000000 2023-10-25 14:29:18,452 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:29:18,452 EPOCH 7 done: loss 0.0214 - lr: 0.000017 2023-10-25 14:29:22,302 DEV : loss 0.19584135711193085 - f1-score (micro avg) 0.7874 2023-10-25 14:29:22,325 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:29:28,254 epoch 8 - iter 89/893 - loss 0.01269647 - time (sec): 5.93 - samples/sec: 4175.01 - lr: 0.000016 - momentum: 0.000000 2023-10-25 14:29:34,170 epoch 8 - iter 178/893 - loss 0.01103982 - time (sec): 11.84 - samples/sec: 4330.76 - lr: 0.000016 - momentum: 0.000000 2023-10-25 14:29:40,172 epoch 8 - iter 267/893 - loss 0.01056955 - time (sec): 17.84 - samples/sec: 4320.95 - lr: 0.000015 - momentum: 0.000000 2023-10-25 14:29:46,086 epoch 8 - iter 356/893 - loss 0.01165606 - time (sec): 23.76 - samples/sec: 4333.99 - lr: 0.000014 - momentum: 0.000000 2023-10-25 14:29:51,817 epoch 8 - iter 445/893 - loss 0.01217482 - time (sec): 29.49 - samples/sec: 4280.96 - lr: 0.000014 - momentum: 0.000000 2023-10-25 14:29:57,619 epoch 8 - iter 534/893 - loss 0.01360158 - time (sec): 35.29 - samples/sec: 4255.30 - lr: 0.000013 - momentum: 0.000000 2023-10-25 14:30:03,211 epoch 8 - iter 623/893 - loss 0.01358451 - time (sec): 40.88 - samples/sec: 4280.98 - lr: 0.000013 - momentum: 0.000000 2023-10-25 14:30:09,010 epoch 8 - iter 712/893 - loss 0.01326304 - time (sec): 46.68 - samples/sec: 4279.43 - lr: 0.000012 - momentum: 0.000000 2023-10-25 14:30:14,815 epoch 8 - iter 801/893 - loss 0.01363656 - time (sec): 52.49 - samples/sec: 4280.60 - lr: 0.000012 - momentum: 0.000000 2023-10-25 14:30:20,749 epoch 8 - iter 890/893 - loss 0.01344285 - time (sec): 58.42 - samples/sec: 4244.81 - lr: 0.000011 - momentum: 0.000000 2023-10-25 14:30:20,929 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:30:20,929 EPOCH 8 done: loss 0.0134 - lr: 0.000011 2023-10-25 14:30:26,071 DEV : loss 0.21516965329647064 - f1-score (micro avg) 0.796 2023-10-25 14:30:26,096 saving best model 2023-10-25 14:30:26,883 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:30:32,597 epoch 9 - iter 89/893 - loss 0.01268252 - time (sec): 5.71 - samples/sec: 4284.31 - lr: 0.000011 - momentum: 0.000000 2023-10-25 14:30:38,584 epoch 9 - iter 178/893 - loss 0.01397325 - time (sec): 11.70 - samples/sec: 4298.37 - lr: 0.000010 - momentum: 0.000000 2023-10-25 14:30:44,380 epoch 9 - iter 267/893 - loss 0.01243272 - time (sec): 17.49 - samples/sec: 4354.73 - lr: 0.000009 - momentum: 0.000000 2023-10-25 14:30:49,932 epoch 9 - iter 356/893 - loss 0.01305437 - time (sec): 23.05 - samples/sec: 4343.10 - lr: 0.000009 - momentum: 0.000000 2023-10-25 14:30:55,731 epoch 9 - iter 445/893 - loss 0.01171325 - time (sec): 28.85 - samples/sec: 4364.70 - lr: 0.000008 - momentum: 0.000000 2023-10-25 14:31:01,450 epoch 9 - iter 534/893 - loss 0.01082927 - time (sec): 34.56 - samples/sec: 4348.58 - lr: 0.000008 - momentum: 0.000000 2023-10-25 14:31:07,263 epoch 9 - iter 623/893 - loss 0.01010485 - time (sec): 40.38 - samples/sec: 4345.62 - lr: 0.000007 - momentum: 0.000000 2023-10-25 14:31:12,771 epoch 9 - iter 712/893 - loss 0.01004408 - time (sec): 45.88 - samples/sec: 4345.42 - lr: 0.000007 - momentum: 0.000000 2023-10-25 14:31:18,573 epoch 9 - iter 801/893 - loss 0.01006474 - time (sec): 51.69 - samples/sec: 4340.94 - lr: 0.000006 - momentum: 0.000000 2023-10-25 14:31:24,256 epoch 9 - iter 890/893 - loss 0.01007528 - time (sec): 57.37 - samples/sec: 4323.30 - lr: 0.000006 - momentum: 0.000000 2023-10-25 14:31:24,441 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:31:24,441 EPOCH 9 done: loss 0.0101 - lr: 0.000006 2023-10-25 14:31:29,063 DEV : loss 0.23022107779979706 - f1-score (micro avg) 0.7853 2023-10-25 14:31:29,084 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:31:34,994 epoch 10 - iter 89/893 - loss 0.00710690 - time (sec): 5.91 - samples/sec: 4371.85 - lr: 0.000005 - momentum: 0.000000 2023-10-25 14:31:40,814 epoch 10 - iter 178/893 - loss 0.00583517 - time (sec): 11.73 - samples/sec: 4378.00 - lr: 0.000004 - momentum: 0.000000 2023-10-25 14:31:46,500 epoch 10 - iter 267/893 - loss 0.00596312 - time (sec): 17.41 - samples/sec: 4319.07 - lr: 0.000004 - momentum: 0.000000 2023-10-25 14:31:52,284 epoch 10 - iter 356/893 - loss 0.00716220 - time (sec): 23.20 - samples/sec: 4331.80 - lr: 0.000003 - momentum: 0.000000 2023-10-25 14:31:57,828 epoch 10 - iter 445/893 - loss 0.00782994 - time (sec): 28.74 - samples/sec: 4317.08 - lr: 0.000003 - momentum: 0.000000 2023-10-25 14:32:03,404 epoch 10 - iter 534/893 - loss 0.00818274 - time (sec): 34.32 - samples/sec: 4315.55 - lr: 0.000002 - momentum: 0.000000 2023-10-25 14:32:09,212 epoch 10 - iter 623/893 - loss 0.00842830 - time (sec): 40.13 - samples/sec: 4322.39 - lr: 0.000002 - momentum: 0.000000 2023-10-25 14:32:15,164 epoch 10 - iter 712/893 - loss 0.00783075 - time (sec): 46.08 - samples/sec: 4317.21 - lr: 0.000001 - momentum: 0.000000 2023-10-25 14:32:20,786 epoch 10 - iter 801/893 - loss 0.00742591 - time (sec): 51.70 - samples/sec: 4318.44 - lr: 0.000001 - momentum: 0.000000 2023-10-25 14:32:26,431 epoch 10 - iter 890/893 - loss 0.00733109 - time (sec): 57.35 - samples/sec: 4320.96 - lr: 0.000000 - momentum: 0.000000 2023-10-25 14:32:26,641 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:32:26,641 EPOCH 10 done: loss 0.0073 - lr: 0.000000 2023-10-25 14:32:30,523 DEV : loss 0.2292526513338089 - f1-score (micro avg) 0.7891 2023-10-25 14:32:31,069 ---------------------------------------------------------------------------------------------------- 2023-10-25 14:32:31,071 Loading model from best epoch ... 2023-10-25 14:32:33,012 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 2023-10-25 14:32:45,073 Results: - F-score (micro) 0.6915 - F-score (macro) 0.6191 - Accuracy 0.5445 By class: precision recall f1-score support LOC 0.7139 0.6767 0.6948 1095 PER 0.7722 0.7737 0.7730 1012 ORG 0.4444 0.5602 0.4957 357 HumanProd 0.4444 0.6061 0.5128 33 micro avg 0.6847 0.6984 0.6915 2497 macro avg 0.5937 0.6542 0.6191 2497 weighted avg 0.6954 0.6984 0.6956 2497 2023-10-25 14:32:45,073 ----------------------------------------------------------------------------------------------------