2023-10-11 20:52:35,997 ---------------------------------------------------------------------------------------------------- 2023-10-11 20:52:35,999 Model: "SequenceTagger( (embeddings): ByT5Embeddings( (model): T5EncoderModel( (shared): Embedding(384, 1472) (encoder): T5Stack( (embed_tokens): Embedding(384, 1472) (block): ModuleList( (0): T5Block( (layer): ModuleList( (0): T5LayerSelfAttention( (SelfAttention): T5Attention( (q): Linear(in_features=1472, out_features=384, bias=False) (k): Linear(in_features=1472, out_features=384, bias=False) (v): Linear(in_features=1472, out_features=384, bias=False) (o): Linear(in_features=384, out_features=1472, bias=False) (relative_attention_bias): Embedding(32, 6) ) (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) (1): T5LayerFF( (DenseReluDense): T5DenseGatedActDense( (wi_0): Linear(in_features=1472, out_features=3584, bias=False) (wi_1): Linear(in_features=1472, out_features=3584, bias=False) (wo): Linear(in_features=3584, out_features=1472, bias=False) (dropout): Dropout(p=0.1, inplace=False) (act): NewGELUActivation() ) (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) (1-11): 11 x T5Block( (layer): ModuleList( (0): T5LayerSelfAttention( (SelfAttention): T5Attention( (q): Linear(in_features=1472, out_features=384, bias=False) (k): Linear(in_features=1472, out_features=384, bias=False) (v): Linear(in_features=1472, out_features=384, bias=False) (o): Linear(in_features=384, out_features=1472, bias=False) ) (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) (1): T5LayerFF( (DenseReluDense): T5DenseGatedActDense( (wi_0): Linear(in_features=1472, out_features=3584, bias=False) (wi_1): Linear(in_features=1472, out_features=3584, bias=False) (wo): Linear(in_features=3584, out_features=1472, bias=False) (dropout): Dropout(p=0.1, inplace=False) (act): NewGELUActivation() ) (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) ) (final_layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) (locked_dropout): LockedDropout(p=0.5) (linear): Linear(in_features=1472, out_features=17, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-11 20:52:36,000 ---------------------------------------------------------------------------------------------------- 2023-10-11 20:52:36,000 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 2023-10-11 20:52:36,000 ---------------------------------------------------------------------------------------------------- 2023-10-11 20:52:36,000 Train: 7142 sentences 2023-10-11 20:52:36,000 (train_with_dev=False, train_with_test=False) 2023-10-11 20:52:36,000 ---------------------------------------------------------------------------------------------------- 2023-10-11 20:52:36,000 Training Params: 2023-10-11 20:52:36,000 - learning_rate: "0.00015" 2023-10-11 20:52:36,000 - mini_batch_size: "8" 2023-10-11 20:52:36,000 - max_epochs: "10" 2023-10-11 20:52:36,000 - shuffle: "True" 2023-10-11 20:52:36,000 ---------------------------------------------------------------------------------------------------- 2023-10-11 20:52:36,001 Plugins: 2023-10-11 20:52:36,001 - TensorboardLogger 2023-10-11 20:52:36,001 - LinearScheduler | warmup_fraction: '0.1' 2023-10-11 20:52:36,001 ---------------------------------------------------------------------------------------------------- 2023-10-11 20:52:36,001 Final evaluation on model from best epoch (best-model.pt) 2023-10-11 20:52:36,001 - metric: "('micro avg', 'f1-score')" 2023-10-11 20:52:36,001 ---------------------------------------------------------------------------------------------------- 2023-10-11 20:52:36,001 Computation: 2023-10-11 20:52:36,001 - compute on device: cuda:0 2023-10-11 20:52:36,001 - embedding storage: none 2023-10-11 20:52:36,001 ---------------------------------------------------------------------------------------------------- 2023-10-11 20:52:36,001 Model training base path: "hmbench-newseye/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-5" 2023-10-11 20:52:36,001 ---------------------------------------------------------------------------------------------------- 2023-10-11 20:52:36,001 ---------------------------------------------------------------------------------------------------- 2023-10-11 20:52:36,002 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-11 20:53:30,393 epoch 1 - iter 89/893 - loss 2.81613488 - time (sec): 54.39 - samples/sec: 498.00 - lr: 0.000015 - momentum: 0.000000 2023-10-11 20:54:22,518 epoch 1 - iter 178/893 - loss 2.73845748 - time (sec): 106.51 - samples/sec: 496.29 - lr: 0.000030 - momentum: 0.000000 2023-10-11 20:55:13,898 epoch 1 - iter 267/893 - loss 2.54046487 - time (sec): 157.89 - samples/sec: 504.70 - lr: 0.000045 - momentum: 0.000000 2023-10-11 20:56:03,634 epoch 1 - iter 356/893 - loss 2.33753858 - time (sec): 207.63 - samples/sec: 504.86 - lr: 0.000060 - momentum: 0.000000 2023-10-11 20:56:52,756 epoch 1 - iter 445/893 - loss 2.11803692 - time (sec): 256.75 - samples/sec: 503.90 - lr: 0.000075 - momentum: 0.000000 2023-10-11 20:57:41,859 epoch 1 - iter 534/893 - loss 1.90825297 - time (sec): 305.86 - samples/sec: 500.33 - lr: 0.000090 - momentum: 0.000000 2023-10-11 20:58:30,865 epoch 1 - iter 623/893 - loss 1.73583373 - time (sec): 354.86 - samples/sec: 498.90 - lr: 0.000104 - momentum: 0.000000 2023-10-11 20:59:18,284 epoch 1 - iter 712/893 - loss 1.60212408 - time (sec): 402.28 - samples/sec: 495.84 - lr: 0.000119 - momentum: 0.000000 2023-10-11 21:00:07,780 epoch 1 - iter 801/893 - loss 1.47431684 - time (sec): 451.78 - samples/sec: 494.97 - lr: 0.000134 - momentum: 0.000000 2023-10-11 21:00:56,330 epoch 1 - iter 890/893 - loss 1.36510703 - time (sec): 500.33 - samples/sec: 495.86 - lr: 0.000149 - momentum: 0.000000 2023-10-11 21:00:57,770 ---------------------------------------------------------------------------------------------------- 2023-10-11 21:00:57,771 EPOCH 1 done: loss 1.3620 - lr: 0.000149 2023-10-11 21:01:17,012 DEV : loss 0.2528855502605438 - f1-score (micro avg) 0.3972 2023-10-11 21:01:17,041 saving best model 2023-10-11 21:01:17,905 ---------------------------------------------------------------------------------------------------- 2023-10-11 21:02:06,967 epoch 2 - iter 89/893 - loss 0.31703165 - time (sec): 49.06 - samples/sec: 508.66 - lr: 0.000148 - momentum: 0.000000 2023-10-11 21:02:56,720 epoch 2 - iter 178/893 - loss 0.29370793 - time (sec): 98.81 - samples/sec: 509.36 - lr: 0.000147 - momentum: 0.000000 2023-10-11 21:03:47,381 epoch 2 - iter 267/893 - loss 0.26771875 - time (sec): 149.47 - samples/sec: 504.46 - lr: 0.000145 - momentum: 0.000000 2023-10-11 21:04:35,944 epoch 2 - iter 356/893 - loss 0.24749931 - time (sec): 198.04 - samples/sec: 504.18 - lr: 0.000143 - momentum: 0.000000 2023-10-11 21:05:25,721 epoch 2 - iter 445/893 - loss 0.22753805 - time (sec): 247.81 - samples/sec: 509.20 - lr: 0.000142 - momentum: 0.000000 2023-10-11 21:06:13,587 epoch 2 - iter 534/893 - loss 0.21653155 - time (sec): 295.68 - samples/sec: 505.52 - lr: 0.000140 - momentum: 0.000000 2023-10-11 21:07:01,206 epoch 2 - iter 623/893 - loss 0.20530533 - time (sec): 343.30 - samples/sec: 504.85 - lr: 0.000138 - momentum: 0.000000 2023-10-11 21:07:49,818 epoch 2 - iter 712/893 - loss 0.19565496 - time (sec): 391.91 - samples/sec: 507.23 - lr: 0.000137 - momentum: 0.000000 2023-10-11 21:08:37,917 epoch 2 - iter 801/893 - loss 0.18851018 - time (sec): 440.01 - samples/sec: 506.57 - lr: 0.000135 - momentum: 0.000000 2023-10-11 21:09:26,153 epoch 2 - iter 890/893 - loss 0.17990919 - time (sec): 488.25 - samples/sec: 507.13 - lr: 0.000133 - momentum: 0.000000 2023-10-11 21:09:27,932 ---------------------------------------------------------------------------------------------------- 2023-10-11 21:09:27,933 EPOCH 2 done: loss 0.1797 - lr: 0.000133 2023-10-11 21:09:49,130 DEV : loss 0.10362720489501953 - f1-score (micro avg) 0.7507 2023-10-11 21:09:49,162 saving best model 2023-10-11 21:09:51,732 ---------------------------------------------------------------------------------------------------- 2023-10-11 21:10:41,847 epoch 3 - iter 89/893 - loss 0.08229146 - time (sec): 50.11 - samples/sec: 489.78 - lr: 0.000132 - momentum: 0.000000 2023-10-11 21:11:31,776 epoch 3 - iter 178/893 - loss 0.08114947 - time (sec): 100.04 - samples/sec: 500.73 - lr: 0.000130 - momentum: 0.000000 2023-10-11 21:12:20,073 epoch 3 - iter 267/893 - loss 0.08051125 - time (sec): 148.34 - samples/sec: 497.04 - lr: 0.000128 - momentum: 0.000000 2023-10-11 21:13:08,650 epoch 3 - iter 356/893 - loss 0.07991414 - time (sec): 196.91 - samples/sec: 497.08 - lr: 0.000127 - momentum: 0.000000 2023-10-11 21:13:59,030 epoch 3 - iter 445/893 - loss 0.07654832 - time (sec): 247.29 - samples/sec: 499.98 - lr: 0.000125 - momentum: 0.000000 2023-10-11 21:14:49,928 epoch 3 - iter 534/893 - loss 0.07698540 - time (sec): 298.19 - samples/sec: 500.71 - lr: 0.000123 - momentum: 0.000000 2023-10-11 21:15:39,596 epoch 3 - iter 623/893 - loss 0.07568197 - time (sec): 347.86 - samples/sec: 500.44 - lr: 0.000122 - momentum: 0.000000 2023-10-11 21:16:26,772 epoch 3 - iter 712/893 - loss 0.07552582 - time (sec): 395.04 - samples/sec: 499.42 - lr: 0.000120 - momentum: 0.000000 2023-10-11 21:17:14,772 epoch 3 - iter 801/893 - loss 0.07615903 - time (sec): 443.04 - samples/sec: 500.35 - lr: 0.000118 - momentum: 0.000000 2023-10-11 21:18:04,158 epoch 3 - iter 890/893 - loss 0.07422971 - time (sec): 492.42 - samples/sec: 503.52 - lr: 0.000117 - momentum: 0.000000 2023-10-11 21:18:05,654 ---------------------------------------------------------------------------------------------------- 2023-10-11 21:18:05,654 EPOCH 3 done: loss 0.0744 - lr: 0.000117 2023-10-11 21:18:26,816 DEV : loss 0.10866602510213852 - f1-score (micro avg) 0.7824 2023-10-11 21:18:26,845 saving best model 2023-10-11 21:18:29,409 ---------------------------------------------------------------------------------------------------- 2023-10-11 21:19:18,462 epoch 4 - iter 89/893 - loss 0.05155868 - time (sec): 49.05 - samples/sec: 546.85 - lr: 0.000115 - momentum: 0.000000 2023-10-11 21:20:06,688 epoch 4 - iter 178/893 - loss 0.04973704 - time (sec): 97.27 - samples/sec: 522.35 - lr: 0.000113 - momentum: 0.000000 2023-10-11 21:20:54,538 epoch 4 - iter 267/893 - loss 0.04803400 - time (sec): 145.12 - samples/sec: 523.03 - lr: 0.000112 - momentum: 0.000000 2023-10-11 21:21:42,141 epoch 4 - iter 356/893 - loss 0.04813800 - time (sec): 192.73 - samples/sec: 521.30 - lr: 0.000110 - momentum: 0.000000 2023-10-11 21:22:29,528 epoch 4 - iter 445/893 - loss 0.05071141 - time (sec): 240.11 - samples/sec: 516.67 - lr: 0.000108 - momentum: 0.000000 2023-10-11 21:23:17,733 epoch 4 - iter 534/893 - loss 0.05008976 - time (sec): 288.32 - samples/sec: 518.01 - lr: 0.000107 - momentum: 0.000000 2023-10-11 21:24:04,878 epoch 4 - iter 623/893 - loss 0.05021815 - time (sec): 335.46 - samples/sec: 516.12 - lr: 0.000105 - momentum: 0.000000 2023-10-11 21:24:52,076 epoch 4 - iter 712/893 - loss 0.05021718 - time (sec): 382.66 - samples/sec: 515.30 - lr: 0.000103 - momentum: 0.000000 2023-10-11 21:25:41,335 epoch 4 - iter 801/893 - loss 0.05019721 - time (sec): 431.92 - samples/sec: 518.39 - lr: 0.000102 - momentum: 0.000000 2023-10-11 21:26:29,456 epoch 4 - iter 890/893 - loss 0.05021542 - time (sec): 480.04 - samples/sec: 516.82 - lr: 0.000100 - momentum: 0.000000 2023-10-11 21:26:30,870 ---------------------------------------------------------------------------------------------------- 2023-10-11 21:26:30,870 EPOCH 4 done: loss 0.0502 - lr: 0.000100 2023-10-11 21:26:52,142 DEV : loss 0.12372089177370071 - f1-score (micro avg) 0.7877 2023-10-11 21:26:52,174 saving best model 2023-10-11 21:26:54,776 ---------------------------------------------------------------------------------------------------- 2023-10-11 21:27:42,615 epoch 5 - iter 89/893 - loss 0.03046838 - time (sec): 47.83 - samples/sec: 512.85 - lr: 0.000098 - momentum: 0.000000 2023-10-11 21:28:30,404 epoch 5 - iter 178/893 - loss 0.03085776 - time (sec): 95.62 - samples/sec: 512.12 - lr: 0.000097 - momentum: 0.000000 2023-10-11 21:29:19,488 epoch 5 - iter 267/893 - loss 0.03370752 - time (sec): 144.71 - samples/sec: 513.15 - lr: 0.000095 - momentum: 0.000000 2023-10-11 21:30:07,132 epoch 5 - iter 356/893 - loss 0.03329933 - time (sec): 192.35 - samples/sec: 508.19 - lr: 0.000093 - momentum: 0.000000 2023-10-11 21:30:54,518 epoch 5 - iter 445/893 - loss 0.03377639 - time (sec): 239.74 - samples/sec: 507.69 - lr: 0.000092 - momentum: 0.000000 2023-10-11 21:31:42,115 epoch 5 - iter 534/893 - loss 0.03379913 - time (sec): 287.33 - samples/sec: 509.40 - lr: 0.000090 - momentum: 0.000000 2023-10-11 21:32:31,921 epoch 5 - iter 623/893 - loss 0.03448817 - time (sec): 337.14 - samples/sec: 515.02 - lr: 0.000088 - momentum: 0.000000 2023-10-11 21:33:19,817 epoch 5 - iter 712/893 - loss 0.03635184 - time (sec): 385.04 - samples/sec: 515.41 - lr: 0.000087 - momentum: 0.000000 2023-10-11 21:34:08,207 epoch 5 - iter 801/893 - loss 0.03720106 - time (sec): 433.43 - samples/sec: 515.07 - lr: 0.000085 - momentum: 0.000000 2023-10-11 21:34:56,748 epoch 5 - iter 890/893 - loss 0.03706033 - time (sec): 481.97 - samples/sec: 514.77 - lr: 0.000083 - momentum: 0.000000 2023-10-11 21:34:58,157 ---------------------------------------------------------------------------------------------------- 2023-10-11 21:34:58,158 EPOCH 5 done: loss 0.0370 - lr: 0.000083 2023-10-11 21:35:19,477 DEV : loss 0.1354159116744995 - f1-score (micro avg) 0.8008 2023-10-11 21:35:19,508 saving best model 2023-10-11 21:35:22,255 ---------------------------------------------------------------------------------------------------- 2023-10-11 21:36:12,944 epoch 6 - iter 89/893 - loss 0.02532107 - time (sec): 50.69 - samples/sec: 506.97 - lr: 0.000082 - momentum: 0.000000 2023-10-11 21:37:02,222 epoch 6 - iter 178/893 - loss 0.02623938 - time (sec): 99.96 - samples/sec: 498.02 - lr: 0.000080 - momentum: 0.000000 2023-10-11 21:37:54,722 epoch 6 - iter 267/893 - loss 0.02667748 - time (sec): 152.46 - samples/sec: 505.98 - lr: 0.000078 - momentum: 0.000000 2023-10-11 21:38:44,103 epoch 6 - iter 356/893 - loss 0.02740607 - time (sec): 201.84 - samples/sec: 501.98 - lr: 0.000077 - momentum: 0.000000 2023-10-11 21:39:34,876 epoch 6 - iter 445/893 - loss 0.02812179 - time (sec): 252.62 - samples/sec: 504.13 - lr: 0.000075 - momentum: 0.000000 2023-10-11 21:40:23,868 epoch 6 - iter 534/893 - loss 0.02732688 - time (sec): 301.61 - samples/sec: 502.95 - lr: 0.000073 - momentum: 0.000000 2023-10-11 21:41:12,773 epoch 6 - iter 623/893 - loss 0.02745474 - time (sec): 350.51 - samples/sec: 501.52 - lr: 0.000072 - momentum: 0.000000 2023-10-11 21:42:02,849 epoch 6 - iter 712/893 - loss 0.02741235 - time (sec): 400.59 - samples/sec: 501.62 - lr: 0.000070 - momentum: 0.000000 2023-10-11 21:42:51,619 epoch 6 - iter 801/893 - loss 0.02707534 - time (sec): 449.36 - samples/sec: 499.01 - lr: 0.000068 - momentum: 0.000000 2023-10-11 21:43:40,632 epoch 6 - iter 890/893 - loss 0.02836216 - time (sec): 498.37 - samples/sec: 496.97 - lr: 0.000067 - momentum: 0.000000 2023-10-11 21:43:42,374 ---------------------------------------------------------------------------------------------------- 2023-10-11 21:43:42,375 EPOCH 6 done: loss 0.0283 - lr: 0.000067 2023-10-11 21:44:03,672 DEV : loss 0.15866339206695557 - f1-score (micro avg) 0.8064 2023-10-11 21:44:03,703 saving best model 2023-10-11 21:44:06,269 ---------------------------------------------------------------------------------------------------- 2023-10-11 21:44:55,120 epoch 7 - iter 89/893 - loss 0.02508843 - time (sec): 48.85 - samples/sec: 492.72 - lr: 0.000065 - momentum: 0.000000 2023-10-11 21:45:46,535 epoch 7 - iter 178/893 - loss 0.02218286 - time (sec): 100.26 - samples/sec: 496.71 - lr: 0.000063 - momentum: 0.000000 2023-10-11 21:46:37,443 epoch 7 - iter 267/893 - loss 0.02238652 - time (sec): 151.17 - samples/sec: 487.09 - lr: 0.000062 - momentum: 0.000000 2023-10-11 21:47:30,667 epoch 7 - iter 356/893 - loss 0.02043595 - time (sec): 204.39 - samples/sec: 486.68 - lr: 0.000060 - momentum: 0.000000 2023-10-11 21:48:22,320 epoch 7 - iter 445/893 - loss 0.02102355 - time (sec): 256.05 - samples/sec: 486.24 - lr: 0.000058 - momentum: 0.000000 2023-10-11 21:49:15,960 epoch 7 - iter 534/893 - loss 0.02018963 - time (sec): 309.69 - samples/sec: 482.13 - lr: 0.000057 - momentum: 0.000000 2023-10-11 21:50:07,792 epoch 7 - iter 623/893 - loss 0.02143349 - time (sec): 361.52 - samples/sec: 480.22 - lr: 0.000055 - momentum: 0.000000 2023-10-11 21:50:59,256 epoch 7 - iter 712/893 - loss 0.02184410 - time (sec): 412.98 - samples/sec: 479.36 - lr: 0.000053 - momentum: 0.000000 2023-10-11 21:51:50,666 epoch 7 - iter 801/893 - loss 0.02268425 - time (sec): 464.39 - samples/sec: 480.74 - lr: 0.000052 - momentum: 0.000000 2023-10-11 21:52:42,522 epoch 7 - iter 890/893 - loss 0.02266701 - time (sec): 516.25 - samples/sec: 480.13 - lr: 0.000050 - momentum: 0.000000 2023-10-11 21:52:44,260 ---------------------------------------------------------------------------------------------------- 2023-10-11 21:52:44,260 EPOCH 7 done: loss 0.0227 - lr: 0.000050 2023-10-11 21:53:07,308 DEV : loss 0.17522385716438293 - f1-score (micro avg) 0.8056 2023-10-11 21:53:07,340 ---------------------------------------------------------------------------------------------------- 2023-10-11 21:53:59,208 epoch 8 - iter 89/893 - loss 0.01896678 - time (sec): 51.87 - samples/sec: 482.91 - lr: 0.000048 - momentum: 0.000000 2023-10-11 21:54:51,807 epoch 8 - iter 178/893 - loss 0.01940469 - time (sec): 104.46 - samples/sec: 481.54 - lr: 0.000047 - momentum: 0.000000 2023-10-11 21:55:42,555 epoch 8 - iter 267/893 - loss 0.01877862 - time (sec): 155.21 - samples/sec: 484.20 - lr: 0.000045 - momentum: 0.000000 2023-10-11 21:56:32,597 epoch 8 - iter 356/893 - loss 0.01814048 - time (sec): 205.25 - samples/sec: 479.00 - lr: 0.000043 - momentum: 0.000000 2023-10-11 21:57:23,920 epoch 8 - iter 445/893 - loss 0.01757910 - time (sec): 256.58 - samples/sec: 474.93 - lr: 0.000042 - momentum: 0.000000 2023-10-11 21:58:17,174 epoch 8 - iter 534/893 - loss 0.01741564 - time (sec): 309.83 - samples/sec: 477.97 - lr: 0.000040 - momentum: 0.000000 2023-10-11 21:59:08,333 epoch 8 - iter 623/893 - loss 0.01792758 - time (sec): 360.99 - samples/sec: 473.96 - lr: 0.000038 - momentum: 0.000000 2023-10-11 22:00:01,125 epoch 8 - iter 712/893 - loss 0.01725608 - time (sec): 413.78 - samples/sec: 476.51 - lr: 0.000037 - momentum: 0.000000 2023-10-11 22:00:54,116 epoch 8 - iter 801/893 - loss 0.01742091 - time (sec): 466.77 - samples/sec: 478.74 - lr: 0.000035 - momentum: 0.000000 2023-10-11 22:01:45,063 epoch 8 - iter 890/893 - loss 0.01785493 - time (sec): 517.72 - samples/sec: 478.83 - lr: 0.000033 - momentum: 0.000000 2023-10-11 22:01:46,660 ---------------------------------------------------------------------------------------------------- 2023-10-11 22:01:46,661 EPOCH 8 done: loss 0.0178 - lr: 0.000033 2023-10-11 22:02:09,359 DEV : loss 0.18762467801570892 - f1-score (micro avg) 0.8003 2023-10-11 22:02:09,391 ---------------------------------------------------------------------------------------------------- 2023-10-11 22:02:59,516 epoch 9 - iter 89/893 - loss 0.01401632 - time (sec): 50.12 - samples/sec: 517.17 - lr: 0.000032 - momentum: 0.000000 2023-10-11 22:03:48,274 epoch 9 - iter 178/893 - loss 0.01214473 - time (sec): 98.88 - samples/sec: 505.06 - lr: 0.000030 - momentum: 0.000000 2023-10-11 22:04:37,160 epoch 9 - iter 267/893 - loss 0.01334299 - time (sec): 147.77 - samples/sec: 502.93 - lr: 0.000028 - momentum: 0.000000 2023-10-11 22:05:26,043 epoch 9 - iter 356/893 - loss 0.01321979 - time (sec): 196.65 - samples/sec: 499.97 - lr: 0.000027 - momentum: 0.000000 2023-10-11 22:06:14,819 epoch 9 - iter 445/893 - loss 0.01360501 - time (sec): 245.43 - samples/sec: 500.45 - lr: 0.000025 - momentum: 0.000000 2023-10-11 22:07:04,964 epoch 9 - iter 534/893 - loss 0.01382477 - time (sec): 295.57 - samples/sec: 504.43 - lr: 0.000023 - momentum: 0.000000 2023-10-11 22:07:55,514 epoch 9 - iter 623/893 - loss 0.01454438 - time (sec): 346.12 - samples/sec: 506.54 - lr: 0.000022 - momentum: 0.000000 2023-10-11 22:08:45,051 epoch 9 - iter 712/893 - loss 0.01451511 - time (sec): 395.66 - samples/sec: 506.42 - lr: 0.000020 - momentum: 0.000000 2023-10-11 22:09:34,492 epoch 9 - iter 801/893 - loss 0.01499387 - time (sec): 445.10 - samples/sec: 504.31 - lr: 0.000019 - momentum: 0.000000 2023-10-11 22:10:22,701 epoch 9 - iter 890/893 - loss 0.01501724 - time (sec): 493.31 - samples/sec: 502.86 - lr: 0.000017 - momentum: 0.000000 2023-10-11 22:10:24,180 ---------------------------------------------------------------------------------------------------- 2023-10-11 22:10:24,180 EPOCH 9 done: loss 0.0151 - lr: 0.000017 2023-10-11 22:10:45,263 DEV : loss 0.18942216038703918 - f1-score (micro avg) 0.8021 2023-10-11 22:10:45,292 ---------------------------------------------------------------------------------------------------- 2023-10-11 22:11:33,613 epoch 10 - iter 89/893 - loss 0.01255615 - time (sec): 48.32 - samples/sec: 522.34 - lr: 0.000015 - momentum: 0.000000 2023-10-11 22:12:22,743 epoch 10 - iter 178/893 - loss 0.01274093 - time (sec): 97.45 - samples/sec: 518.18 - lr: 0.000013 - momentum: 0.000000 2023-10-11 22:13:11,304 epoch 10 - iter 267/893 - loss 0.01151278 - time (sec): 146.01 - samples/sec: 519.59 - lr: 0.000012 - momentum: 0.000000 2023-10-11 22:13:59,597 epoch 10 - iter 356/893 - loss 0.01202932 - time (sec): 194.30 - samples/sec: 520.48 - lr: 0.000010 - momentum: 0.000000 2023-10-11 22:14:47,516 epoch 10 - iter 445/893 - loss 0.01209366 - time (sec): 242.22 - samples/sec: 522.14 - lr: 0.000008 - momentum: 0.000000 2023-10-11 22:15:34,392 epoch 10 - iter 534/893 - loss 0.01129065 - time (sec): 289.10 - samples/sec: 520.44 - lr: 0.000007 - momentum: 0.000000 2023-10-11 22:16:23,818 epoch 10 - iter 623/893 - loss 0.01226211 - time (sec): 338.52 - samples/sec: 520.53 - lr: 0.000005 - momentum: 0.000000 2023-10-11 22:17:10,864 epoch 10 - iter 712/893 - loss 0.01183880 - time (sec): 385.57 - samples/sec: 519.33 - lr: 0.000004 - momentum: 0.000000 2023-10-11 22:17:58,433 epoch 10 - iter 801/893 - loss 0.01204233 - time (sec): 433.14 - samples/sec: 517.95 - lr: 0.000002 - momentum: 0.000000 2023-10-11 22:18:45,661 epoch 10 - iter 890/893 - loss 0.01204849 - time (sec): 480.37 - samples/sec: 516.66 - lr: 0.000000 - momentum: 0.000000 2023-10-11 22:18:46,984 ---------------------------------------------------------------------------------------------------- 2023-10-11 22:18:46,984 EPOCH 10 done: loss 0.0120 - lr: 0.000000 2023-10-11 22:19:07,928 DEV : loss 0.19141535460948944 - f1-score (micro avg) 0.8029 2023-10-11 22:19:08,805 ---------------------------------------------------------------------------------------------------- 2023-10-11 22:19:08,807 Loading model from best epoch ... 2023-10-11 22:19:12,543 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 2023-10-11 22:20:19,954 Results: - F-score (micro) 0.7109 - F-score (macro) 0.6499 - Accuracy 0.5673 By class: precision recall f1-score support LOC 0.7284 0.7324 0.7304 1095 PER 0.7764 0.7925 0.7844 1012 ORG 0.4249 0.6022 0.4983 357 HumanProd 0.5238 0.6667 0.5867 33 micro avg 0.6864 0.7373 0.7109 2497 macro avg 0.6134 0.6985 0.6499 2497 weighted avg 0.7018 0.7373 0.7172 2497 2023-10-11 22:20:19,954 ----------------------------------------------------------------------------------------------------