2023-10-25 17:12:05,512 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:12:05,513 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 17:12:05,514 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:12:05,514 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 17:12:05,514 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:12:05,514 Train: 7142 sentences 2023-10-25 17:12:05,514 (train_with_dev=False, train_with_test=False) 2023-10-25 17:12:05,514 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:12:05,514 Training Params: 2023-10-25 17:12:05,514 - learning_rate: "5e-05" 2023-10-25 17:12:05,514 - mini_batch_size: "8" 2023-10-25 17:12:05,514 - max_epochs: "10" 2023-10-25 17:12:05,514 - shuffle: "True" 2023-10-25 17:12:05,514 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:12:05,514 Plugins: 2023-10-25 17:12:05,514 - TensorboardLogger 2023-10-25 17:12:05,514 - LinearScheduler | warmup_fraction: '0.1' 2023-10-25 17:12:05,515 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:12:05,515 Final evaluation on model from best epoch (best-model.pt) 2023-10-25 17:12:05,515 - metric: "('micro avg', 'f1-score')" 2023-10-25 17:12:05,515 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:12:05,515 Computation: 2023-10-25 17:12:05,515 - compute on device: cuda:0 2023-10-25 17:12:05,515 - embedding storage: none 2023-10-25 17:12:05,515 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:12:05,515 Model training base path: "hmbench-newseye/fr-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4" 2023-10-25 17:12:05,515 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:12:05,515 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:12:05,515 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-25 17:12:11,754 epoch 1 - iter 89/893 - loss 1.86966354 - time (sec): 6.24 - samples/sec: 4048.40 - lr: 0.000005 - momentum: 0.000000 2023-10-25 17:12:18,158 epoch 1 - iter 178/893 - loss 1.17037592 - time (sec): 12.64 - samples/sec: 4015.73 - lr: 0.000010 - momentum: 0.000000 2023-10-25 17:12:24,252 epoch 1 - iter 267/893 - loss 0.89779536 - time (sec): 18.74 - samples/sec: 3990.87 - lr: 0.000015 - momentum: 0.000000 2023-10-25 17:12:30,281 epoch 1 - iter 356/893 - loss 0.72799361 - time (sec): 24.77 - samples/sec: 4021.65 - lr: 0.000020 - momentum: 0.000000 2023-10-25 17:12:36,084 epoch 1 - iter 445/893 - loss 0.62359043 - time (sec): 30.57 - samples/sec: 4037.40 - lr: 0.000025 - momentum: 0.000000 2023-10-25 17:12:41,987 epoch 1 - iter 534/893 - loss 0.54978542 - time (sec): 36.47 - samples/sec: 4059.93 - lr: 0.000030 - momentum: 0.000000 2023-10-25 17:12:48,656 epoch 1 - iter 623/893 - loss 0.49086057 - time (sec): 43.14 - samples/sec: 4008.81 - lr: 0.000035 - momentum: 0.000000 2023-10-25 17:12:54,781 epoch 1 - iter 712/893 - loss 0.44720365 - time (sec): 49.27 - samples/sec: 4036.10 - lr: 0.000040 - momentum: 0.000000 2023-10-25 17:13:00,618 epoch 1 - iter 801/893 - loss 0.41503176 - time (sec): 55.10 - samples/sec: 4063.39 - lr: 0.000045 - momentum: 0.000000 2023-10-25 17:13:06,507 epoch 1 - iter 890/893 - loss 0.38844168 - time (sec): 60.99 - samples/sec: 4060.90 - lr: 0.000050 - momentum: 0.000000 2023-10-25 17:13:06,717 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:13:06,717 EPOCH 1 done: loss 0.3873 - lr: 0.000050 2023-10-25 17:13:09,813 DEV : loss 0.10254143178462982 - f1-score (micro avg) 0.7386 2023-10-25 17:13:09,836 saving best model 2023-10-25 17:13:10,384 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:13:16,410 epoch 2 - iter 89/893 - loss 0.11297432 - time (sec): 6.02 - samples/sec: 4100.95 - lr: 0.000049 - momentum: 0.000000 2023-10-25 17:13:22,388 epoch 2 - iter 178/893 - loss 0.10095221 - time (sec): 12.00 - samples/sec: 4091.37 - lr: 0.000049 - momentum: 0.000000 2023-10-25 17:13:28,578 epoch 2 - iter 267/893 - loss 0.09904465 - time (sec): 18.19 - samples/sec: 4123.74 - lr: 0.000048 - momentum: 0.000000 2023-10-25 17:13:34,706 epoch 2 - iter 356/893 - loss 0.10304399 - time (sec): 24.32 - samples/sec: 4161.23 - lr: 0.000048 - momentum: 0.000000 2023-10-25 17:13:40,891 epoch 2 - iter 445/893 - loss 0.10290863 - time (sec): 30.50 - samples/sec: 4163.32 - lr: 0.000047 - momentum: 0.000000 2023-10-25 17:13:46,799 epoch 2 - iter 534/893 - loss 0.10253728 - time (sec): 36.41 - samples/sec: 4138.20 - lr: 0.000047 - momentum: 0.000000 2023-10-25 17:13:52,856 epoch 2 - iter 623/893 - loss 0.10517292 - time (sec): 42.47 - samples/sec: 4127.97 - lr: 0.000046 - momentum: 0.000000 2023-10-25 17:13:59,058 epoch 2 - iter 712/893 - loss 0.10448830 - time (sec): 48.67 - samples/sec: 4121.70 - lr: 0.000046 - momentum: 0.000000 2023-10-25 17:14:05,695 epoch 2 - iter 801/893 - loss 0.10541126 - time (sec): 55.31 - samples/sec: 4030.63 - lr: 0.000045 - momentum: 0.000000 2023-10-25 17:14:11,612 epoch 2 - iter 890/893 - loss 0.10497282 - time (sec): 61.23 - samples/sec: 4053.12 - lr: 0.000044 - momentum: 0.000000 2023-10-25 17:14:11,796 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:14:11,797 EPOCH 2 done: loss 0.1048 - lr: 0.000044 2023-10-25 17:14:15,732 DEV : loss 0.09835401922464371 - f1-score (micro avg) 0.7484 2023-10-25 17:14:15,753 saving best model 2023-10-25 17:14:17,077 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:14:22,720 epoch 3 - iter 89/893 - loss 0.06670513 - time (sec): 5.64 - samples/sec: 4160.09 - lr: 0.000044 - momentum: 0.000000 2023-10-25 17:14:28,626 epoch 3 - iter 178/893 - loss 0.06908344 - time (sec): 11.55 - samples/sec: 4278.75 - lr: 0.000043 - momentum: 0.000000 2023-10-25 17:14:34,039 epoch 3 - iter 267/893 - loss 0.06482755 - time (sec): 16.96 - samples/sec: 4378.26 - lr: 0.000043 - momentum: 0.000000 2023-10-25 17:14:39,692 epoch 3 - iter 356/893 - loss 0.06663973 - time (sec): 22.61 - samples/sec: 4381.76 - lr: 0.000042 - momentum: 0.000000 2023-10-25 17:14:45,211 epoch 3 - iter 445/893 - loss 0.06653035 - time (sec): 28.13 - samples/sec: 4430.59 - lr: 0.000042 - momentum: 0.000000 2023-10-25 17:14:50,834 epoch 3 - iter 534/893 - loss 0.06616108 - time (sec): 33.75 - samples/sec: 4434.83 - lr: 0.000041 - momentum: 0.000000 2023-10-25 17:14:56,338 epoch 3 - iter 623/893 - loss 0.06517711 - time (sec): 39.26 - samples/sec: 4441.20 - lr: 0.000041 - momentum: 0.000000 2023-10-25 17:15:01,626 epoch 3 - iter 712/893 - loss 0.06547935 - time (sec): 44.55 - samples/sec: 4424.27 - lr: 0.000040 - momentum: 0.000000 2023-10-25 17:15:07,582 epoch 3 - iter 801/893 - loss 0.06531523 - time (sec): 50.50 - samples/sec: 4426.12 - lr: 0.000039 - momentum: 0.000000 2023-10-25 17:15:13,138 epoch 3 - iter 890/893 - loss 0.06609427 - time (sec): 56.06 - samples/sec: 4424.89 - lr: 0.000039 - momentum: 0.000000 2023-10-25 17:15:13,306 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:15:13,306 EPOCH 3 done: loss 0.0663 - lr: 0.000039 2023-10-25 17:15:18,728 DEV : loss 0.10818858444690704 - f1-score (micro avg) 0.78 2023-10-25 17:15:18,750 saving best model 2023-10-25 17:15:19,433 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:15:25,434 epoch 4 - iter 89/893 - loss 0.04360221 - time (sec): 6.00 - samples/sec: 4325.89 - lr: 0.000038 - momentum: 0.000000 2023-10-25 17:15:31,204 epoch 4 - iter 178/893 - loss 0.04758612 - time (sec): 11.77 - samples/sec: 4382.46 - lr: 0.000038 - momentum: 0.000000 2023-10-25 17:15:36,707 epoch 4 - iter 267/893 - loss 0.04849685 - time (sec): 17.27 - samples/sec: 4342.52 - lr: 0.000037 - momentum: 0.000000 2023-10-25 17:15:42,413 epoch 4 - iter 356/893 - loss 0.04765784 - time (sec): 22.98 - samples/sec: 4307.89 - lr: 0.000037 - momentum: 0.000000 2023-10-25 17:15:48,360 epoch 4 - iter 445/893 - loss 0.04585792 - time (sec): 28.92 - samples/sec: 4301.93 - lr: 0.000036 - momentum: 0.000000 2023-10-25 17:15:53,987 epoch 4 - iter 534/893 - loss 0.04678008 - time (sec): 34.55 - samples/sec: 4339.31 - lr: 0.000036 - momentum: 0.000000 2023-10-25 17:15:59,524 epoch 4 - iter 623/893 - loss 0.04623631 - time (sec): 40.09 - samples/sec: 4325.05 - lr: 0.000035 - momentum: 0.000000 2023-10-25 17:16:05,227 epoch 4 - iter 712/893 - loss 0.04616789 - time (sec): 45.79 - samples/sec: 4335.78 - lr: 0.000034 - momentum: 0.000000 2023-10-25 17:16:10,897 epoch 4 - iter 801/893 - loss 0.04654488 - time (sec): 51.46 - samples/sec: 4351.22 - lr: 0.000034 - momentum: 0.000000 2023-10-25 17:16:16,408 epoch 4 - iter 890/893 - loss 0.04626882 - time (sec): 56.97 - samples/sec: 4340.98 - lr: 0.000033 - momentum: 0.000000 2023-10-25 17:16:16,693 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:16:16,694 EPOCH 4 done: loss 0.0460 - lr: 0.000033 2023-10-25 17:16:21,128 DEV : loss 0.14430440962314606 - f1-score (micro avg) 0.7763 2023-10-25 17:16:21,148 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:16:26,833 epoch 5 - iter 89/893 - loss 0.03076065 - time (sec): 5.68 - samples/sec: 4079.73 - lr: 0.000033 - momentum: 0.000000 2023-10-25 17:16:32,613 epoch 5 - iter 178/893 - loss 0.03387457 - time (sec): 11.46 - samples/sec: 4189.45 - lr: 0.000032 - momentum: 0.000000 2023-10-25 17:16:38,412 epoch 5 - iter 267/893 - loss 0.03857465 - time (sec): 17.26 - samples/sec: 4219.15 - lr: 0.000032 - momentum: 0.000000 2023-10-25 17:16:44,270 epoch 5 - iter 356/893 - loss 0.03668496 - time (sec): 23.12 - samples/sec: 4226.11 - lr: 0.000031 - momentum: 0.000000 2023-10-25 17:16:50,915 epoch 5 - iter 445/893 - loss 0.03779554 - time (sec): 29.77 - samples/sec: 4131.62 - lr: 0.000031 - momentum: 0.000000 2023-10-25 17:16:56,745 epoch 5 - iter 534/893 - loss 0.03736583 - time (sec): 35.60 - samples/sec: 4160.67 - lr: 0.000030 - momentum: 0.000000 2023-10-25 17:17:02,359 epoch 5 - iter 623/893 - loss 0.03619574 - time (sec): 41.21 - samples/sec: 4183.66 - lr: 0.000029 - momentum: 0.000000 2023-10-25 17:17:08,388 epoch 5 - iter 712/893 - loss 0.03616854 - time (sec): 47.24 - samples/sec: 4165.28 - lr: 0.000029 - momentum: 0.000000 2023-10-25 17:17:14,180 epoch 5 - iter 801/893 - loss 0.03605795 - time (sec): 53.03 - samples/sec: 4207.84 - lr: 0.000028 - momentum: 0.000000 2023-10-25 17:17:19,715 epoch 5 - iter 890/893 - loss 0.03609116 - time (sec): 58.57 - samples/sec: 4231.73 - lr: 0.000028 - momentum: 0.000000 2023-10-25 17:17:19,915 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:17:19,915 EPOCH 5 done: loss 0.0361 - lr: 0.000028 2023-10-25 17:17:23,906 DEV : loss 0.1808791607618332 - f1-score (micro avg) 0.7874 2023-10-25 17:17:23,926 saving best model 2023-10-25 17:17:24,575 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:17:30,427 epoch 6 - iter 89/893 - loss 0.03174096 - time (sec): 5.85 - samples/sec: 4049.45 - lr: 0.000027 - momentum: 0.000000 2023-10-25 17:17:36,184 epoch 6 - iter 178/893 - loss 0.02710066 - time (sec): 11.61 - samples/sec: 4011.60 - lr: 0.000027 - momentum: 0.000000 2023-10-25 17:17:42,186 epoch 6 - iter 267/893 - loss 0.02807563 - time (sec): 17.61 - samples/sec: 4098.40 - lr: 0.000026 - momentum: 0.000000 2023-10-25 17:17:48,150 epoch 6 - iter 356/893 - loss 0.02692728 - time (sec): 23.57 - samples/sec: 4121.36 - lr: 0.000026 - momentum: 0.000000 2023-10-25 17:17:54,081 epoch 6 - iter 445/893 - loss 0.02735570 - time (sec): 29.50 - samples/sec: 4159.09 - lr: 0.000025 - momentum: 0.000000 2023-10-25 17:18:00,130 epoch 6 - iter 534/893 - loss 0.02805044 - time (sec): 35.55 - samples/sec: 4171.09 - lr: 0.000024 - momentum: 0.000000 2023-10-25 17:18:06,195 epoch 6 - iter 623/893 - loss 0.02725646 - time (sec): 41.62 - samples/sec: 4157.30 - lr: 0.000024 - momentum: 0.000000 2023-10-25 17:18:12,254 epoch 6 - iter 712/893 - loss 0.02769298 - time (sec): 47.68 - samples/sec: 4162.52 - lr: 0.000023 - momentum: 0.000000 2023-10-25 17:18:18,094 epoch 6 - iter 801/893 - loss 0.02746510 - time (sec): 53.52 - samples/sec: 4161.49 - lr: 0.000023 - momentum: 0.000000 2023-10-25 17:18:24,001 epoch 6 - iter 890/893 - loss 0.02725677 - time (sec): 59.42 - samples/sec: 4178.69 - lr: 0.000022 - momentum: 0.000000 2023-10-25 17:18:24,190 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:18:24,190 EPOCH 6 done: loss 0.0274 - lr: 0.000022 2023-10-25 17:18:29,212 DEV : loss 0.18829816579818726 - f1-score (micro avg) 0.8008 2023-10-25 17:18:29,234 saving best model 2023-10-25 17:18:29,906 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:18:35,941 epoch 7 - iter 89/893 - loss 0.01515087 - time (sec): 6.03 - samples/sec: 3972.69 - lr: 0.000022 - momentum: 0.000000 2023-10-25 17:18:41,986 epoch 7 - iter 178/893 - loss 0.02051512 - time (sec): 12.08 - samples/sec: 4024.69 - lr: 0.000021 - momentum: 0.000000 2023-10-25 17:18:48,051 epoch 7 - iter 267/893 - loss 0.01998553 - time (sec): 18.14 - samples/sec: 4115.71 - lr: 0.000021 - momentum: 0.000000 2023-10-25 17:18:54,021 epoch 7 - iter 356/893 - loss 0.01986934 - time (sec): 24.11 - samples/sec: 4124.11 - lr: 0.000020 - momentum: 0.000000 2023-10-25 17:18:59,887 epoch 7 - iter 445/893 - loss 0.02129780 - time (sec): 29.98 - samples/sec: 4178.30 - lr: 0.000019 - momentum: 0.000000 2023-10-25 17:19:05,771 epoch 7 - iter 534/893 - loss 0.02089146 - time (sec): 35.86 - samples/sec: 4208.17 - lr: 0.000019 - momentum: 0.000000 2023-10-25 17:19:11,666 epoch 7 - iter 623/893 - loss 0.02180007 - time (sec): 41.76 - samples/sec: 4201.24 - lr: 0.000018 - momentum: 0.000000 2023-10-25 17:19:17,316 epoch 7 - iter 712/893 - loss 0.02130021 - time (sec): 47.41 - samples/sec: 4179.67 - lr: 0.000018 - momentum: 0.000000 2023-10-25 17:19:23,114 epoch 7 - iter 801/893 - loss 0.02092378 - time (sec): 53.21 - samples/sec: 4187.33 - lr: 0.000017 - momentum: 0.000000 2023-10-25 17:19:29,000 epoch 7 - iter 890/893 - loss 0.02059414 - time (sec): 59.09 - samples/sec: 4200.88 - lr: 0.000017 - momentum: 0.000000 2023-10-25 17:19:29,168 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:19:29,168 EPOCH 7 done: loss 0.0206 - lr: 0.000017 2023-10-25 17:19:33,135 DEV : loss 0.20971202850341797 - f1-score (micro avg) 0.7835 2023-10-25 17:19:33,158 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:19:39,046 epoch 8 - iter 89/893 - loss 0.01764633 - time (sec): 5.89 - samples/sec: 4379.59 - lr: 0.000016 - momentum: 0.000000 2023-10-25 17:19:45,066 epoch 8 - iter 178/893 - loss 0.01628569 - time (sec): 11.91 - samples/sec: 4235.59 - lr: 0.000016 - momentum: 0.000000 2023-10-25 17:19:51,714 epoch 8 - iter 267/893 - loss 0.01625886 - time (sec): 18.55 - samples/sec: 4031.94 - lr: 0.000015 - momentum: 0.000000 2023-10-25 17:19:57,412 epoch 8 - iter 356/893 - loss 0.01628296 - time (sec): 24.25 - samples/sec: 4045.87 - lr: 0.000014 - momentum: 0.000000 2023-10-25 17:20:03,072 epoch 8 - iter 445/893 - loss 0.01501426 - time (sec): 29.91 - samples/sec: 4086.07 - lr: 0.000014 - momentum: 0.000000 2023-10-25 17:20:08,856 epoch 8 - iter 534/893 - loss 0.01465639 - time (sec): 35.70 - samples/sec: 4134.41 - lr: 0.000013 - momentum: 0.000000 2023-10-25 17:20:14,511 epoch 8 - iter 623/893 - loss 0.01504650 - time (sec): 41.35 - samples/sec: 4158.41 - lr: 0.000013 - momentum: 0.000000 2023-10-25 17:20:20,232 epoch 8 - iter 712/893 - loss 0.01458875 - time (sec): 47.07 - samples/sec: 4166.37 - lr: 0.000012 - momentum: 0.000000 2023-10-25 17:20:26,137 epoch 8 - iter 801/893 - loss 0.01570324 - time (sec): 52.98 - samples/sec: 4186.22 - lr: 0.000012 - momentum: 0.000000 2023-10-25 17:20:32,101 epoch 8 - iter 890/893 - loss 0.01581706 - time (sec): 58.94 - samples/sec: 4207.71 - lr: 0.000011 - momentum: 0.000000 2023-10-25 17:20:32,280 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:20:32,281 EPOCH 8 done: loss 0.0158 - lr: 0.000011 2023-10-25 17:20:36,350 DEV : loss 0.21289943158626556 - f1-score (micro avg) 0.8 2023-10-25 17:20:36,375 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:20:42,196 epoch 9 - iter 89/893 - loss 0.00765889 - time (sec): 5.82 - samples/sec: 4349.70 - lr: 0.000011 - momentum: 0.000000 2023-10-25 17:20:48,053 epoch 9 - iter 178/893 - loss 0.00998425 - time (sec): 11.68 - samples/sec: 4306.03 - lr: 0.000010 - momentum: 0.000000 2023-10-25 17:20:54,019 epoch 9 - iter 267/893 - loss 0.01244333 - time (sec): 17.64 - samples/sec: 4202.39 - lr: 0.000009 - momentum: 0.000000 2023-10-25 17:20:59,745 epoch 9 - iter 356/893 - loss 0.01212745 - time (sec): 23.37 - samples/sec: 4282.98 - lr: 0.000009 - momentum: 0.000000 2023-10-25 17:21:05,354 epoch 9 - iter 445/893 - loss 0.01178149 - time (sec): 28.98 - samples/sec: 4326.46 - lr: 0.000008 - momentum: 0.000000 2023-10-25 17:21:10,965 epoch 9 - iter 534/893 - loss 0.01170529 - time (sec): 34.59 - samples/sec: 4302.74 - lr: 0.000008 - momentum: 0.000000 2023-10-25 17:21:16,851 epoch 9 - iter 623/893 - loss 0.01136873 - time (sec): 40.47 - samples/sec: 4324.76 - lr: 0.000007 - momentum: 0.000000 2023-10-25 17:21:22,488 epoch 9 - iter 712/893 - loss 0.01142673 - time (sec): 46.11 - samples/sec: 4298.22 - lr: 0.000007 - momentum: 0.000000 2023-10-25 17:21:28,182 epoch 9 - iter 801/893 - loss 0.01136731 - time (sec): 51.81 - samples/sec: 4291.71 - lr: 0.000006 - momentum: 0.000000 2023-10-25 17:21:34,076 epoch 9 - iter 890/893 - loss 0.01120456 - time (sec): 57.70 - samples/sec: 4294.92 - lr: 0.000006 - momentum: 0.000000 2023-10-25 17:21:34,259 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:21:34,259 EPOCH 9 done: loss 0.0112 - lr: 0.000006 2023-10-25 17:21:39,355 DEV : loss 0.22147664427757263 - f1-score (micro avg) 0.7981 2023-10-25 17:21:39,378 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:21:44,817 epoch 10 - iter 89/893 - loss 0.00613679 - time (sec): 5.44 - samples/sec: 4461.62 - lr: 0.000005 - momentum: 0.000000 2023-10-25 17:21:50,598 epoch 10 - iter 178/893 - loss 0.00627718 - time (sec): 11.22 - samples/sec: 4222.26 - lr: 0.000004 - momentum: 0.000000 2023-10-25 17:21:56,574 epoch 10 - iter 267/893 - loss 0.00759554 - time (sec): 17.19 - samples/sec: 4268.40 - lr: 0.000004 - momentum: 0.000000 2023-10-25 17:22:02,673 epoch 10 - iter 356/893 - loss 0.00807461 - time (sec): 23.29 - samples/sec: 4232.88 - lr: 0.000003 - momentum: 0.000000 2023-10-25 17:22:08,727 epoch 10 - iter 445/893 - loss 0.00789473 - time (sec): 29.35 - samples/sec: 4155.31 - lr: 0.000003 - momentum: 0.000000 2023-10-25 17:22:14,934 epoch 10 - iter 534/893 - loss 0.00783961 - time (sec): 35.55 - samples/sec: 4162.38 - lr: 0.000002 - momentum: 0.000000 2023-10-25 17:22:20,987 epoch 10 - iter 623/893 - loss 0.00764283 - time (sec): 41.61 - samples/sec: 4158.54 - lr: 0.000002 - momentum: 0.000000 2023-10-25 17:22:26,725 epoch 10 - iter 712/893 - loss 0.00703458 - time (sec): 47.35 - samples/sec: 4144.99 - lr: 0.000001 - momentum: 0.000000 2023-10-25 17:22:32,754 epoch 10 - iter 801/893 - loss 0.00689931 - time (sec): 53.37 - samples/sec: 4157.08 - lr: 0.000001 - momentum: 0.000000 2023-10-25 17:22:38,960 epoch 10 - iter 890/893 - loss 0.00666490 - time (sec): 59.58 - samples/sec: 4161.77 - lr: 0.000000 - momentum: 0.000000 2023-10-25 17:22:39,141 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:22:39,141 EPOCH 10 done: loss 0.0066 - lr: 0.000000 2023-10-25 17:22:43,875 DEV : loss 0.23105905950069427 - f1-score (micro avg) 0.8 2023-10-25 17:22:44,387 ---------------------------------------------------------------------------------------------------- 2023-10-25 17:22:44,388 Loading model from best epoch ... 2023-10-25 17:22:46,202 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 17:22:58,838 Results: - F-score (micro) 0.6773 - F-score (macro) 0.588 - Accuracy 0.5304 By class: precision recall f1-score support LOC 0.6839 0.6877 0.6858 1095 PER 0.7644 0.7500 0.7571 1012 ORG 0.4379 0.5434 0.4850 357 HumanProd 0.3182 0.6364 0.4242 33 micro avg 0.6635 0.6916 0.6773 2497 macro avg 0.5511 0.6544 0.5880 2497 weighted avg 0.6765 0.6916 0.6825 2497 2023-10-25 17:22:58,839 ----------------------------------------------------------------------------------------------------