2023-10-25 15:18:28,415 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:18:28,416 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 15:18:28,416 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:18:28,416 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 15:18:28,416 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:18:28,416 Train: 7142 sentences 2023-10-25 15:18:28,417 (train_with_dev=False, train_with_test=False) 2023-10-25 15:18:28,417 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:18:28,417 Training Params: 2023-10-25 15:18:28,417 - learning_rate: "5e-05" 2023-10-25 15:18:28,417 - mini_batch_size: "8" 2023-10-25 15:18:28,417 - max_epochs: "10" 2023-10-25 15:18:28,417 - shuffle: "True" 2023-10-25 15:18:28,417 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:18:28,417 Plugins: 2023-10-25 15:18:28,417 - TensorboardLogger 2023-10-25 15:18:28,417 - LinearScheduler | warmup_fraction: '0.1' 2023-10-25 15:18:28,417 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:18:28,417 Final evaluation on model from best epoch (best-model.pt) 2023-10-25 15:18:28,417 - metric: "('micro avg', 'f1-score')" 2023-10-25 15:18:28,417 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:18:28,417 Computation: 2023-10-25 15:18:28,418 - compute on device: cuda:0 2023-10-25 15:18:28,418 - embedding storage: none 2023-10-25 15:18:28,418 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:18:28,418 Model training base path: "hmbench-newseye/fr-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2" 2023-10-25 15:18:28,418 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:18:28,418 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:18:28,418 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-25 15:18:34,554 epoch 1 - iter 89/893 - loss 1.99536437 - time (sec): 6.13 - samples/sec: 4156.59 - lr: 0.000005 - momentum: 0.000000 2023-10-25 15:18:40,462 epoch 1 - iter 178/893 - loss 1.28347735 - time (sec): 12.04 - samples/sec: 4054.53 - lr: 0.000010 - momentum: 0.000000 2023-10-25 15:18:46,184 epoch 1 - iter 267/893 - loss 0.96621297 - time (sec): 17.77 - samples/sec: 4067.33 - lr: 0.000015 - momentum: 0.000000 2023-10-25 15:18:52,212 epoch 1 - iter 356/893 - loss 0.78153614 - time (sec): 23.79 - samples/sec: 4068.56 - lr: 0.000020 - momentum: 0.000000 2023-10-25 15:18:58,243 epoch 1 - iter 445/893 - loss 0.65825937 - time (sec): 29.82 - samples/sec: 4089.24 - lr: 0.000025 - momentum: 0.000000 2023-10-25 15:19:04,627 epoch 1 - iter 534/893 - loss 0.56876226 - time (sec): 36.21 - samples/sec: 4096.24 - lr: 0.000030 - momentum: 0.000000 2023-10-25 15:19:10,683 epoch 1 - iter 623/893 - loss 0.51115509 - time (sec): 42.26 - samples/sec: 4096.86 - lr: 0.000035 - momentum: 0.000000 2023-10-25 15:19:16,477 epoch 1 - iter 712/893 - loss 0.46644377 - time (sec): 48.06 - samples/sec: 4119.52 - lr: 0.000040 - momentum: 0.000000 2023-10-25 15:19:22,505 epoch 1 - iter 801/893 - loss 0.43085918 - time (sec): 54.09 - samples/sec: 4116.89 - lr: 0.000045 - momentum: 0.000000 2023-10-25 15:19:28,588 epoch 1 - iter 890/893 - loss 0.40247481 - time (sec): 60.17 - samples/sec: 4123.30 - lr: 0.000050 - momentum: 0.000000 2023-10-25 15:19:28,768 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:19:28,768 EPOCH 1 done: loss 0.4017 - lr: 0.000050 2023-10-25 15:19:32,698 DEV : loss 0.129494771361351 - f1-score (micro avg) 0.7147 2023-10-25 15:19:32,721 saving best model 2023-10-25 15:19:33,191 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:19:39,171 epoch 2 - iter 89/893 - loss 0.11570827 - time (sec): 5.98 - samples/sec: 4292.85 - lr: 0.000049 - momentum: 0.000000 2023-10-25 15:19:44,703 epoch 2 - iter 178/893 - loss 0.11300717 - time (sec): 11.51 - samples/sec: 4096.21 - lr: 0.000049 - momentum: 0.000000 2023-10-25 15:19:50,654 epoch 2 - iter 267/893 - loss 0.10732609 - time (sec): 17.46 - samples/sec: 4226.81 - lr: 0.000048 - momentum: 0.000000 2023-10-25 15:19:56,334 epoch 2 - iter 356/893 - loss 0.10846905 - time (sec): 23.14 - samples/sec: 4255.34 - lr: 0.000048 - momentum: 0.000000 2023-10-25 15:20:02,251 epoch 2 - iter 445/893 - loss 0.10556608 - time (sec): 29.06 - samples/sec: 4268.35 - lr: 0.000047 - momentum: 0.000000 2023-10-25 15:20:07,968 epoch 2 - iter 534/893 - loss 0.10460658 - time (sec): 34.78 - samples/sec: 4277.84 - lr: 0.000047 - momentum: 0.000000 2023-10-25 15:20:13,816 epoch 2 - iter 623/893 - loss 0.10460754 - time (sec): 40.62 - samples/sec: 4304.95 - lr: 0.000046 - momentum: 0.000000 2023-10-25 15:20:19,498 epoch 2 - iter 712/893 - loss 0.10363498 - time (sec): 46.31 - samples/sec: 4259.99 - lr: 0.000046 - momentum: 0.000000 2023-10-25 15:20:25,661 epoch 2 - iter 801/893 - loss 0.10398883 - time (sec): 52.47 - samples/sec: 4243.14 - lr: 0.000045 - momentum: 0.000000 2023-10-25 15:20:31,642 epoch 2 - iter 890/893 - loss 0.10345717 - time (sec): 58.45 - samples/sec: 4238.82 - lr: 0.000044 - momentum: 0.000000 2023-10-25 15:20:31,837 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:20:31,837 EPOCH 2 done: loss 0.1032 - lr: 0.000044 2023-10-25 15:20:36,043 DEV : loss 0.1055043414235115 - f1-score (micro avg) 0.7684 2023-10-25 15:20:36,066 saving best model 2023-10-25 15:20:36,737 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:20:43,647 epoch 3 - iter 89/893 - loss 0.06314996 - time (sec): 6.91 - samples/sec: 3683.92 - lr: 0.000044 - momentum: 0.000000 2023-10-25 15:20:49,448 epoch 3 - iter 178/893 - loss 0.05732830 - time (sec): 12.71 - samples/sec: 3851.88 - lr: 0.000043 - momentum: 0.000000 2023-10-25 15:20:55,350 epoch 3 - iter 267/893 - loss 0.06133970 - time (sec): 18.61 - samples/sec: 4029.84 - lr: 0.000043 - momentum: 0.000000 2023-10-25 15:21:01,032 epoch 3 - iter 356/893 - loss 0.06089097 - time (sec): 24.29 - samples/sec: 4104.42 - lr: 0.000042 - momentum: 0.000000 2023-10-25 15:21:06,667 epoch 3 - iter 445/893 - loss 0.06302484 - time (sec): 29.93 - samples/sec: 4110.82 - lr: 0.000042 - momentum: 0.000000 2023-10-25 15:21:12,590 epoch 3 - iter 534/893 - loss 0.06411251 - time (sec): 35.85 - samples/sec: 4120.08 - lr: 0.000041 - momentum: 0.000000 2023-10-25 15:21:18,784 epoch 3 - iter 623/893 - loss 0.06511810 - time (sec): 42.04 - samples/sec: 4134.48 - lr: 0.000041 - momentum: 0.000000 2023-10-25 15:21:24,821 epoch 3 - iter 712/893 - loss 0.06623456 - time (sec): 48.08 - samples/sec: 4153.85 - lr: 0.000040 - momentum: 0.000000 2023-10-25 15:21:30,738 epoch 3 - iter 801/893 - loss 0.06510690 - time (sec): 54.00 - samples/sec: 4169.43 - lr: 0.000039 - momentum: 0.000000 2023-10-25 15:21:36,527 epoch 3 - iter 890/893 - loss 0.06434040 - time (sec): 59.79 - samples/sec: 4151.20 - lr: 0.000039 - momentum: 0.000000 2023-10-25 15:21:36,707 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:21:36,707 EPOCH 3 done: loss 0.0643 - lr: 0.000039 2023-10-25 15:21:40,732 DEV : loss 0.10246531665325165 - f1-score (micro avg) 0.7631 2023-10-25 15:21:40,752 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:21:46,872 epoch 4 - iter 89/893 - loss 0.04679698 - time (sec): 6.12 - samples/sec: 4077.93 - lr: 0.000038 - momentum: 0.000000 2023-10-25 15:21:52,842 epoch 4 - iter 178/893 - loss 0.04843306 - time (sec): 12.09 - samples/sec: 4130.07 - lr: 0.000038 - momentum: 0.000000 2023-10-25 15:21:58,578 epoch 4 - iter 267/893 - loss 0.04773681 - time (sec): 17.82 - samples/sec: 4119.76 - lr: 0.000037 - momentum: 0.000000 2023-10-25 15:22:04,436 epoch 4 - iter 356/893 - loss 0.04706353 - time (sec): 23.68 - samples/sec: 4188.67 - lr: 0.000037 - momentum: 0.000000 2023-10-25 15:22:10,388 epoch 4 - iter 445/893 - loss 0.04712251 - time (sec): 29.63 - samples/sec: 4192.99 - lr: 0.000036 - momentum: 0.000000 2023-10-25 15:22:16,548 epoch 4 - iter 534/893 - loss 0.04626453 - time (sec): 35.79 - samples/sec: 4217.43 - lr: 0.000036 - momentum: 0.000000 2023-10-25 15:22:22,420 epoch 4 - iter 623/893 - loss 0.04772297 - time (sec): 41.67 - samples/sec: 4200.40 - lr: 0.000035 - momentum: 0.000000 2023-10-25 15:22:28,169 epoch 4 - iter 712/893 - loss 0.04802724 - time (sec): 47.42 - samples/sec: 4177.73 - lr: 0.000034 - momentum: 0.000000 2023-10-25 15:22:34,271 epoch 4 - iter 801/893 - loss 0.04644315 - time (sec): 53.52 - samples/sec: 4194.85 - lr: 0.000034 - momentum: 0.000000 2023-10-25 15:22:40,085 epoch 4 - iter 890/893 - loss 0.04640298 - time (sec): 59.33 - samples/sec: 4183.14 - lr: 0.000033 - momentum: 0.000000 2023-10-25 15:22:40,257 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:22:40,257 EPOCH 4 done: loss 0.0465 - lr: 0.000033 2023-10-25 15:22:45,236 DEV : loss 0.12659873068332672 - f1-score (micro avg) 0.795 2023-10-25 15:22:45,256 saving best model 2023-10-25 15:22:45,952 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:22:51,820 epoch 5 - iter 89/893 - loss 0.03385452 - time (sec): 5.86 - samples/sec: 4049.42 - lr: 0.000033 - momentum: 0.000000 2023-10-25 15:22:57,539 epoch 5 - iter 178/893 - loss 0.03418212 - time (sec): 11.58 - samples/sec: 4271.19 - lr: 0.000032 - momentum: 0.000000 2023-10-25 15:23:03,297 epoch 5 - iter 267/893 - loss 0.03554924 - time (sec): 17.34 - samples/sec: 4336.00 - lr: 0.000032 - momentum: 0.000000 2023-10-25 15:23:08,625 epoch 5 - iter 356/893 - loss 0.03505123 - time (sec): 22.67 - samples/sec: 4387.92 - lr: 0.000031 - momentum: 0.000000 2023-10-25 15:23:14,192 epoch 5 - iter 445/893 - loss 0.03409722 - time (sec): 28.24 - samples/sec: 4378.75 - lr: 0.000031 - momentum: 0.000000 2023-10-25 15:23:20,011 epoch 5 - iter 534/893 - loss 0.03585306 - time (sec): 34.05 - samples/sec: 4402.79 - lr: 0.000030 - momentum: 0.000000 2023-10-25 15:23:25,675 epoch 5 - iter 623/893 - loss 0.03523579 - time (sec): 39.72 - samples/sec: 4383.02 - lr: 0.000029 - momentum: 0.000000 2023-10-25 15:23:31,217 epoch 5 - iter 712/893 - loss 0.03457740 - time (sec): 45.26 - samples/sec: 4379.15 - lr: 0.000029 - momentum: 0.000000 2023-10-25 15:23:36,764 epoch 5 - iter 801/893 - loss 0.03595039 - time (sec): 50.81 - samples/sec: 4386.59 - lr: 0.000028 - momentum: 0.000000 2023-10-25 15:23:42,344 epoch 5 - iter 890/893 - loss 0.03641465 - time (sec): 56.39 - samples/sec: 4398.95 - lr: 0.000028 - momentum: 0.000000 2023-10-25 15:23:42,509 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:23:42,510 EPOCH 5 done: loss 0.0363 - lr: 0.000028 2023-10-25 15:23:46,433 DEV : loss 0.15950827300548553 - f1-score (micro avg) 0.8068 2023-10-25 15:23:46,454 saving best model 2023-10-25 15:23:47,110 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:23:53,621 epoch 6 - iter 89/893 - loss 0.01843165 - time (sec): 6.51 - samples/sec: 3866.75 - lr: 0.000027 - momentum: 0.000000 2023-10-25 15:23:59,070 epoch 6 - iter 178/893 - loss 0.02060189 - time (sec): 11.96 - samples/sec: 4081.75 - lr: 0.000027 - momentum: 0.000000 2023-10-25 15:24:04,597 epoch 6 - iter 267/893 - loss 0.02287458 - time (sec): 17.48 - samples/sec: 4238.78 - lr: 0.000026 - momentum: 0.000000 2023-10-25 15:24:10,175 epoch 6 - iter 356/893 - loss 0.02618184 - time (sec): 23.06 - samples/sec: 4299.53 - lr: 0.000026 - momentum: 0.000000 2023-10-25 15:24:15,632 epoch 6 - iter 445/893 - loss 0.02572160 - time (sec): 28.52 - samples/sec: 4391.01 - lr: 0.000025 - momentum: 0.000000 2023-10-25 15:24:20,990 epoch 6 - iter 534/893 - loss 0.02653895 - time (sec): 33.88 - samples/sec: 4380.81 - lr: 0.000024 - momentum: 0.000000 2023-10-25 15:24:26,774 epoch 6 - iter 623/893 - loss 0.02582608 - time (sec): 39.66 - samples/sec: 4384.91 - lr: 0.000024 - momentum: 0.000000 2023-10-25 15:24:32,333 epoch 6 - iter 712/893 - loss 0.02556832 - time (sec): 45.22 - samples/sec: 4406.91 - lr: 0.000023 - momentum: 0.000000 2023-10-25 15:24:37,881 epoch 6 - iter 801/893 - loss 0.02666351 - time (sec): 50.77 - samples/sec: 4411.11 - lr: 0.000023 - momentum: 0.000000 2023-10-25 15:24:43,421 epoch 6 - iter 890/893 - loss 0.02679083 - time (sec): 56.31 - samples/sec: 4408.04 - lr: 0.000022 - momentum: 0.000000 2023-10-25 15:24:43,612 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:24:43,612 EPOCH 6 done: loss 0.0269 - lr: 0.000022 2023-10-25 15:24:47,464 DEV : loss 0.16690844297409058 - f1-score (micro avg) 0.7946 2023-10-25 15:24:47,486 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:24:53,278 epoch 7 - iter 89/893 - loss 0.01604361 - time (sec): 5.79 - samples/sec: 4603.66 - lr: 0.000022 - momentum: 0.000000 2023-10-25 15:24:58,745 epoch 7 - iter 178/893 - loss 0.01903158 - time (sec): 11.26 - samples/sec: 4502.09 - lr: 0.000021 - momentum: 0.000000 2023-10-25 15:25:04,355 epoch 7 - iter 267/893 - loss 0.01762160 - time (sec): 16.87 - samples/sec: 4447.11 - lr: 0.000021 - momentum: 0.000000 2023-10-25 15:25:10,018 epoch 7 - iter 356/893 - loss 0.01799719 - time (sec): 22.53 - samples/sec: 4449.66 - lr: 0.000020 - momentum: 0.000000 2023-10-25 15:25:15,579 epoch 7 - iter 445/893 - loss 0.01816463 - time (sec): 28.09 - samples/sec: 4437.15 - lr: 0.000019 - momentum: 0.000000 2023-10-25 15:25:21,206 epoch 7 - iter 534/893 - loss 0.01762382 - time (sec): 33.72 - samples/sec: 4408.92 - lr: 0.000019 - momentum: 0.000000 2023-10-25 15:25:26,582 epoch 7 - iter 623/893 - loss 0.01818342 - time (sec): 39.09 - samples/sec: 4391.43 - lr: 0.000018 - momentum: 0.000000 2023-10-25 15:25:32,470 epoch 7 - iter 712/893 - loss 0.01864109 - time (sec): 44.98 - samples/sec: 4401.35 - lr: 0.000018 - momentum: 0.000000 2023-10-25 15:25:37,922 epoch 7 - iter 801/893 - loss 0.01859268 - time (sec): 50.43 - samples/sec: 4410.62 - lr: 0.000017 - momentum: 0.000000 2023-10-25 15:25:43,454 epoch 7 - iter 890/893 - loss 0.01876551 - time (sec): 55.97 - samples/sec: 4432.61 - lr: 0.000017 - momentum: 0.000000 2023-10-25 15:25:43,621 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:25:43,621 EPOCH 7 done: loss 0.0187 - lr: 0.000017 2023-10-25 15:25:48,547 DEV : loss 0.20097887516021729 - f1-score (micro avg) 0.797 2023-10-25 15:25:48,569 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:25:54,290 epoch 8 - iter 89/893 - loss 0.01460596 - time (sec): 5.72 - samples/sec: 4188.29 - lr: 0.000016 - momentum: 0.000000 2023-10-25 15:25:59,972 epoch 8 - iter 178/893 - loss 0.01450563 - time (sec): 11.40 - samples/sec: 4260.99 - lr: 0.000016 - momentum: 0.000000 2023-10-25 15:26:05,738 epoch 8 - iter 267/893 - loss 0.01437589 - time (sec): 17.17 - samples/sec: 4309.25 - lr: 0.000015 - momentum: 0.000000 2023-10-25 15:26:11,493 epoch 8 - iter 356/893 - loss 0.01437436 - time (sec): 22.92 - samples/sec: 4260.09 - lr: 0.000014 - momentum: 0.000000 2023-10-25 15:26:17,302 epoch 8 - iter 445/893 - loss 0.01387661 - time (sec): 28.73 - samples/sec: 4259.05 - lr: 0.000014 - momentum: 0.000000 2023-10-25 15:26:23,205 epoch 8 - iter 534/893 - loss 0.01340798 - time (sec): 34.63 - samples/sec: 4266.55 - lr: 0.000013 - momentum: 0.000000 2023-10-25 15:26:29,267 epoch 8 - iter 623/893 - loss 0.01346100 - time (sec): 40.70 - samples/sec: 4280.38 - lr: 0.000013 - momentum: 0.000000 2023-10-25 15:26:34,847 epoch 8 - iter 712/893 - loss 0.01427616 - time (sec): 46.28 - samples/sec: 4271.12 - lr: 0.000012 - momentum: 0.000000 2023-10-25 15:26:40,330 epoch 8 - iter 801/893 - loss 0.01452412 - time (sec): 51.76 - samples/sec: 4297.64 - lr: 0.000012 - momentum: 0.000000 2023-10-25 15:26:45,869 epoch 8 - iter 890/893 - loss 0.01415925 - time (sec): 57.30 - samples/sec: 4328.86 - lr: 0.000011 - momentum: 0.000000 2023-10-25 15:26:46,048 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:26:46,049 EPOCH 8 done: loss 0.0142 - lr: 0.000011 2023-10-25 15:26:50,095 DEV : loss 0.20207209885120392 - f1-score (micro avg) 0.8049 2023-10-25 15:26:50,116 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:26:56,812 epoch 9 - iter 89/893 - loss 0.00563361 - time (sec): 6.69 - samples/sec: 3687.01 - lr: 0.000011 - momentum: 0.000000 2023-10-25 15:27:02,376 epoch 9 - iter 178/893 - loss 0.00548688 - time (sec): 12.26 - samples/sec: 3995.04 - lr: 0.000010 - momentum: 0.000000 2023-10-25 15:27:07,978 epoch 9 - iter 267/893 - loss 0.00713590 - time (sec): 17.86 - samples/sec: 4125.43 - lr: 0.000009 - momentum: 0.000000 2023-10-25 15:27:13,974 epoch 9 - iter 356/893 - loss 0.00914907 - time (sec): 23.86 - samples/sec: 4219.03 - lr: 0.000009 - momentum: 0.000000 2023-10-25 15:27:19,731 epoch 9 - iter 445/893 - loss 0.01028740 - time (sec): 29.61 - samples/sec: 4249.69 - lr: 0.000008 - momentum: 0.000000 2023-10-25 15:27:25,466 epoch 9 - iter 534/893 - loss 0.01006046 - time (sec): 35.35 - samples/sec: 4282.64 - lr: 0.000008 - momentum: 0.000000 2023-10-25 15:27:31,112 epoch 9 - iter 623/893 - loss 0.01076164 - time (sec): 40.99 - samples/sec: 4271.27 - lr: 0.000007 - momentum: 0.000000 2023-10-25 15:27:36,878 epoch 9 - iter 712/893 - loss 0.01047859 - time (sec): 46.76 - samples/sec: 4289.54 - lr: 0.000007 - momentum: 0.000000 2023-10-25 15:27:42,412 epoch 9 - iter 801/893 - loss 0.01028654 - time (sec): 52.29 - samples/sec: 4294.61 - lr: 0.000006 - momentum: 0.000000 2023-10-25 15:27:48,005 epoch 9 - iter 890/893 - loss 0.01012660 - time (sec): 57.89 - samples/sec: 4285.49 - lr: 0.000006 - momentum: 0.000000 2023-10-25 15:27:48,168 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:27:48,169 EPOCH 9 done: loss 0.0101 - lr: 0.000006 2023-10-25 15:27:52,536 DEV : loss 0.21088457107543945 - f1-score (micro avg) 0.8116 2023-10-25 15:27:52,560 saving best model 2023-10-25 15:27:53,280 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:27:58,925 epoch 10 - iter 89/893 - loss 0.00743254 - time (sec): 5.64 - samples/sec: 4235.48 - lr: 0.000005 - momentum: 0.000000 2023-10-25 15:28:04,835 epoch 10 - iter 178/893 - loss 0.00751409 - time (sec): 11.55 - samples/sec: 4297.28 - lr: 0.000004 - momentum: 0.000000 2023-10-25 15:28:10,905 epoch 10 - iter 267/893 - loss 0.00573510 - time (sec): 17.62 - samples/sec: 4241.46 - lr: 0.000004 - momentum: 0.000000 2023-10-25 15:28:16,625 epoch 10 - iter 356/893 - loss 0.00576235 - time (sec): 23.34 - samples/sec: 4280.64 - lr: 0.000003 - momentum: 0.000000 2023-10-25 15:28:22,156 epoch 10 - iter 445/893 - loss 0.00542024 - time (sec): 28.87 - samples/sec: 4228.88 - lr: 0.000003 - momentum: 0.000000 2023-10-25 15:28:27,932 epoch 10 - iter 534/893 - loss 0.00569832 - time (sec): 34.65 - samples/sec: 4272.05 - lr: 0.000002 - momentum: 0.000000 2023-10-25 15:28:33,724 epoch 10 - iter 623/893 - loss 0.00590285 - time (sec): 40.44 - samples/sec: 4273.13 - lr: 0.000002 - momentum: 0.000000 2023-10-25 15:28:39,373 epoch 10 - iter 712/893 - loss 0.00577265 - time (sec): 46.09 - samples/sec: 4289.95 - lr: 0.000001 - momentum: 0.000000 2023-10-25 15:28:45,228 epoch 10 - iter 801/893 - loss 0.00558124 - time (sec): 51.94 - samples/sec: 4264.95 - lr: 0.000001 - momentum: 0.000000 2023-10-25 15:28:51,161 epoch 10 - iter 890/893 - loss 0.00610789 - time (sec): 57.88 - samples/sec: 4286.41 - lr: 0.000000 - momentum: 0.000000 2023-10-25 15:28:51,332 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:28:51,332 EPOCH 10 done: loss 0.0061 - lr: 0.000000 2023-10-25 15:28:56,407 DEV : loss 0.21562287211418152 - f1-score (micro avg) 0.8067 2023-10-25 15:28:56,870 ---------------------------------------------------------------------------------------------------- 2023-10-25 15:28:56,871 Loading model from best epoch ... 2023-10-25 15:28:58,698 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 15:29:10,894 Results: - F-score (micro) 0.6867 - F-score (macro) 0.6046 - Accuracy 0.5377 By class: precision recall f1-score support LOC 0.6919 0.6667 0.6791 1095 PER 0.7771 0.7717 0.7744 1012 ORG 0.4685 0.5630 0.5115 357 HumanProd 0.3438 0.6667 0.4536 33 micro avg 0.6792 0.6944 0.6867 2497 macro avg 0.5703 0.6670 0.6046 2497 weighted avg 0.6899 0.6944 0.6908 2497 2023-10-25 15:29:10,895 ----------------------------------------------------------------------------------------------------