2023-10-11 00:23:05,794 ---------------------------------------------------------------------------------------------------- 2023-10-11 00:23:05,796 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 00:23:05,796 ---------------------------------------------------------------------------------------------------- 2023-10-11 00:23:05,796 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 00:23:05,796 ---------------------------------------------------------------------------------------------------- 2023-10-11 00:23:05,796 Train: 7142 sentences 2023-10-11 00:23:05,797 (train_with_dev=False, train_with_test=False) 2023-10-11 00:23:05,797 ---------------------------------------------------------------------------------------------------- 2023-10-11 00:23:05,797 Training Params: 2023-10-11 00:23:05,797 - learning_rate: "0.00016" 2023-10-11 00:23:05,797 - mini_batch_size: "4" 2023-10-11 00:23:05,797 - max_epochs: "10" 2023-10-11 00:23:05,797 - shuffle: "True" 2023-10-11 00:23:05,797 ---------------------------------------------------------------------------------------------------- 2023-10-11 00:23:05,797 Plugins: 2023-10-11 00:23:05,797 - TensorboardLogger 2023-10-11 00:23:05,797 - LinearScheduler | warmup_fraction: '0.1' 2023-10-11 00:23:05,797 ---------------------------------------------------------------------------------------------------- 2023-10-11 00:23:05,797 Final evaluation on model from best epoch (best-model.pt) 2023-10-11 00:23:05,797 - metric: "('micro avg', 'f1-score')" 2023-10-11 00:23:05,797 ---------------------------------------------------------------------------------------------------- 2023-10-11 00:23:05,798 Computation: 2023-10-11 00:23:05,798 - compute on device: cuda:0 2023-10-11 00:23:05,798 - embedding storage: none 2023-10-11 00:23:05,798 ---------------------------------------------------------------------------------------------------- 2023-10-11 00:23:05,798 Model training base path: "hmbench-newseye/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-1" 2023-10-11 00:23:05,798 ---------------------------------------------------------------------------------------------------- 2023-10-11 00:23:05,798 ---------------------------------------------------------------------------------------------------- 2023-10-11 00:23:05,798 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-11 00:24:00,729 epoch 1 - iter 178/1786 - loss 2.82172262 - time (sec): 54.93 - samples/sec: 463.83 - lr: 0.000016 - momentum: 0.000000 2023-10-11 00:24:54,757 epoch 1 - iter 356/1786 - loss 2.68438929 - time (sec): 108.96 - samples/sec: 462.67 - lr: 0.000032 - momentum: 0.000000 2023-10-11 00:25:49,197 epoch 1 - iter 534/1786 - loss 2.40556334 - time (sec): 163.40 - samples/sec: 459.28 - lr: 0.000048 - momentum: 0.000000 2023-10-11 00:26:43,891 epoch 1 - iter 712/1786 - loss 2.09683094 - time (sec): 218.09 - samples/sec: 456.19 - lr: 0.000064 - momentum: 0.000000 2023-10-11 00:27:39,073 epoch 1 - iter 890/1786 - loss 1.79374454 - time (sec): 273.27 - samples/sec: 459.99 - lr: 0.000080 - momentum: 0.000000 2023-10-11 00:28:31,677 epoch 1 - iter 1068/1786 - loss 1.59687181 - time (sec): 325.88 - samples/sec: 456.68 - lr: 0.000096 - momentum: 0.000000 2023-10-11 00:29:23,556 epoch 1 - iter 1246/1786 - loss 1.43709470 - time (sec): 377.76 - samples/sec: 456.38 - lr: 0.000112 - momentum: 0.000000 2023-10-11 00:30:17,433 epoch 1 - iter 1424/1786 - loss 1.30135011 - time (sec): 431.63 - samples/sec: 457.91 - lr: 0.000127 - momentum: 0.000000 2023-10-11 00:31:10,923 epoch 1 - iter 1602/1786 - loss 1.19051970 - time (sec): 485.12 - samples/sec: 460.52 - lr: 0.000143 - momentum: 0.000000 2023-10-11 00:32:04,890 epoch 1 - iter 1780/1786 - loss 1.10265404 - time (sec): 539.09 - samples/sec: 459.99 - lr: 0.000159 - momentum: 0.000000 2023-10-11 00:32:06,577 ---------------------------------------------------------------------------------------------------- 2023-10-11 00:32:06,577 EPOCH 1 done: loss 1.1001 - lr: 0.000159 2023-10-11 00:32:26,381 DEV : loss 0.21496298909187317 - f1-score (micro avg) 0.4516 2023-10-11 00:32:26,413 saving best model 2023-10-11 00:32:27,258 ---------------------------------------------------------------------------------------------------- 2023-10-11 00:33:22,790 epoch 2 - iter 178/1786 - loss 0.22116459 - time (sec): 55.53 - samples/sec: 476.30 - lr: 0.000158 - momentum: 0.000000 2023-10-11 00:34:16,303 epoch 2 - iter 356/1786 - loss 0.21373852 - time (sec): 109.04 - samples/sec: 462.61 - lr: 0.000156 - momentum: 0.000000 2023-10-11 00:35:09,232 epoch 2 - iter 534/1786 - loss 0.19894140 - time (sec): 161.97 - samples/sec: 458.65 - lr: 0.000155 - momentum: 0.000000 2023-10-11 00:36:03,022 epoch 2 - iter 712/1786 - loss 0.18566527 - time (sec): 215.76 - samples/sec: 460.41 - lr: 0.000153 - momentum: 0.000000 2023-10-11 00:36:56,171 epoch 2 - iter 890/1786 - loss 0.17473385 - time (sec): 268.91 - samples/sec: 461.71 - lr: 0.000151 - momentum: 0.000000 2023-10-11 00:37:48,589 epoch 2 - iter 1068/1786 - loss 0.16918782 - time (sec): 321.33 - samples/sec: 461.01 - lr: 0.000149 - momentum: 0.000000 2023-10-11 00:38:43,484 epoch 2 - iter 1246/1786 - loss 0.16287825 - time (sec): 376.22 - samples/sec: 461.21 - lr: 0.000148 - momentum: 0.000000 2023-10-11 00:39:39,645 epoch 2 - iter 1424/1786 - loss 0.15775704 - time (sec): 432.38 - samples/sec: 461.75 - lr: 0.000146 - momentum: 0.000000 2023-10-11 00:40:34,151 epoch 2 - iter 1602/1786 - loss 0.15320043 - time (sec): 486.89 - samples/sec: 459.43 - lr: 0.000144 - momentum: 0.000000 2023-10-11 00:41:28,520 epoch 2 - iter 1780/1786 - loss 0.14884221 - time (sec): 541.26 - samples/sec: 458.41 - lr: 0.000142 - momentum: 0.000000 2023-10-11 00:41:30,090 ---------------------------------------------------------------------------------------------------- 2023-10-11 00:41:30,091 EPOCH 2 done: loss 0.1488 - lr: 0.000142 2023-10-11 00:41:51,592 DEV : loss 0.10961832106113434 - f1-score (micro avg) 0.7418 2023-10-11 00:41:51,627 saving best model 2023-10-11 00:42:03,657 ---------------------------------------------------------------------------------------------------- 2023-10-11 00:42:59,110 epoch 3 - iter 178/1786 - loss 0.07855565 - time (sec): 55.45 - samples/sec: 430.85 - lr: 0.000140 - momentum: 0.000000 2023-10-11 00:43:55,091 epoch 3 - iter 356/1786 - loss 0.07506546 - time (sec): 111.43 - samples/sec: 443.18 - lr: 0.000139 - momentum: 0.000000 2023-10-11 00:44:52,205 epoch 3 - iter 534/1786 - loss 0.07999800 - time (sec): 168.54 - samples/sec: 439.31 - lr: 0.000137 - momentum: 0.000000 2023-10-11 00:45:47,516 epoch 3 - iter 712/1786 - loss 0.08377628 - time (sec): 223.86 - samples/sec: 435.06 - lr: 0.000135 - momentum: 0.000000 2023-10-11 00:46:45,191 epoch 3 - iter 890/1786 - loss 0.08368157 - time (sec): 281.53 - samples/sec: 437.88 - lr: 0.000133 - momentum: 0.000000 2023-10-11 00:47:41,048 epoch 3 - iter 1068/1786 - loss 0.08196451 - time (sec): 337.39 - samples/sec: 439.01 - lr: 0.000132 - momentum: 0.000000 2023-10-11 00:48:37,387 epoch 3 - iter 1246/1786 - loss 0.07913681 - time (sec): 393.73 - samples/sec: 439.17 - lr: 0.000130 - momentum: 0.000000 2023-10-11 00:49:34,302 epoch 3 - iter 1424/1786 - loss 0.07927759 - time (sec): 450.64 - samples/sec: 440.23 - lr: 0.000128 - momentum: 0.000000 2023-10-11 00:50:31,342 epoch 3 - iter 1602/1786 - loss 0.07914526 - time (sec): 507.68 - samples/sec: 443.41 - lr: 0.000126 - momentum: 0.000000 2023-10-11 00:51:25,788 epoch 3 - iter 1780/1786 - loss 0.07955825 - time (sec): 562.13 - samples/sec: 441.21 - lr: 0.000125 - momentum: 0.000000 2023-10-11 00:51:27,478 ---------------------------------------------------------------------------------------------------- 2023-10-11 00:51:27,479 EPOCH 3 done: loss 0.0795 - lr: 0.000125 2023-10-11 00:51:50,488 DEV : loss 0.13271120190620422 - f1-score (micro avg) 0.7694 2023-10-11 00:51:50,522 saving best model 2023-10-11 00:51:59,738 ---------------------------------------------------------------------------------------------------- 2023-10-11 00:52:53,740 epoch 4 - iter 178/1786 - loss 0.05116751 - time (sec): 54.00 - samples/sec: 461.44 - lr: 0.000123 - momentum: 0.000000 2023-10-11 00:53:48,364 epoch 4 - iter 356/1786 - loss 0.05373327 - time (sec): 108.62 - samples/sec: 453.07 - lr: 0.000121 - momentum: 0.000000 2023-10-11 00:54:42,706 epoch 4 - iter 534/1786 - loss 0.05528519 - time (sec): 162.96 - samples/sec: 452.82 - lr: 0.000119 - momentum: 0.000000 2023-10-11 00:55:39,034 epoch 4 - iter 712/1786 - loss 0.05948955 - time (sec): 219.29 - samples/sec: 452.44 - lr: 0.000117 - momentum: 0.000000 2023-10-11 00:56:35,623 epoch 4 - iter 890/1786 - loss 0.05774224 - time (sec): 275.88 - samples/sec: 455.07 - lr: 0.000116 - momentum: 0.000000 2023-10-11 00:57:31,802 epoch 4 - iter 1068/1786 - loss 0.05684100 - time (sec): 332.06 - samples/sec: 453.93 - lr: 0.000114 - momentum: 0.000000 2023-10-11 00:58:28,188 epoch 4 - iter 1246/1786 - loss 0.05589756 - time (sec): 388.45 - samples/sec: 455.82 - lr: 0.000112 - momentum: 0.000000 2023-10-11 00:59:21,599 epoch 4 - iter 1424/1786 - loss 0.05592335 - time (sec): 441.86 - samples/sec: 456.10 - lr: 0.000110 - momentum: 0.000000 2023-10-11 01:00:15,044 epoch 4 - iter 1602/1786 - loss 0.05630542 - time (sec): 495.30 - samples/sec: 454.18 - lr: 0.000109 - momentum: 0.000000 2023-10-11 01:01:08,426 epoch 4 - iter 1780/1786 - loss 0.05643132 - time (sec): 548.68 - samples/sec: 452.21 - lr: 0.000107 - momentum: 0.000000 2023-10-11 01:01:09,965 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:01:09,966 EPOCH 4 done: loss 0.0565 - lr: 0.000107 2023-10-11 01:01:32,933 DEV : loss 0.1545310765504837 - f1-score (micro avg) 0.754 2023-10-11 01:01:32,964 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:02:28,970 epoch 5 - iter 178/1786 - loss 0.03780263 - time (sec): 56.00 - samples/sec: 454.04 - lr: 0.000105 - momentum: 0.000000 2023-10-11 01:03:22,450 epoch 5 - iter 356/1786 - loss 0.04158210 - time (sec): 109.48 - samples/sec: 444.41 - lr: 0.000103 - momentum: 0.000000 2023-10-11 01:04:19,870 epoch 5 - iter 534/1786 - loss 0.03924275 - time (sec): 166.90 - samples/sec: 446.20 - lr: 0.000101 - momentum: 0.000000 2023-10-11 01:05:16,274 epoch 5 - iter 712/1786 - loss 0.04250127 - time (sec): 223.31 - samples/sec: 450.83 - lr: 0.000100 - momentum: 0.000000 2023-10-11 01:06:10,674 epoch 5 - iter 890/1786 - loss 0.04170514 - time (sec): 277.71 - samples/sec: 444.62 - lr: 0.000098 - momentum: 0.000000 2023-10-11 01:07:04,253 epoch 5 - iter 1068/1786 - loss 0.04064150 - time (sec): 331.29 - samples/sec: 444.86 - lr: 0.000096 - momentum: 0.000000 2023-10-11 01:07:58,515 epoch 5 - iter 1246/1786 - loss 0.04058142 - time (sec): 385.55 - samples/sec: 447.34 - lr: 0.000094 - momentum: 0.000000 2023-10-11 01:08:52,190 epoch 5 - iter 1424/1786 - loss 0.04122103 - time (sec): 439.22 - samples/sec: 451.22 - lr: 0.000093 - momentum: 0.000000 2023-10-11 01:09:46,923 epoch 5 - iter 1602/1786 - loss 0.04009365 - time (sec): 493.96 - samples/sec: 451.63 - lr: 0.000091 - momentum: 0.000000 2023-10-11 01:10:40,618 epoch 5 - iter 1780/1786 - loss 0.04017012 - time (sec): 547.65 - samples/sec: 452.91 - lr: 0.000089 - momentum: 0.000000 2023-10-11 01:10:42,299 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:10:42,299 EPOCH 5 done: loss 0.0402 - lr: 0.000089 2023-10-11 01:11:05,576 DEV : loss 0.17520776391029358 - f1-score (micro avg) 0.793 2023-10-11 01:11:05,607 saving best model 2023-10-11 01:11:16,215 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:12:11,344 epoch 6 - iter 178/1786 - loss 0.03032200 - time (sec): 55.12 - samples/sec: 452.24 - lr: 0.000087 - momentum: 0.000000 2023-10-11 01:13:06,283 epoch 6 - iter 356/1786 - loss 0.03095283 - time (sec): 110.06 - samples/sec: 450.08 - lr: 0.000085 - momentum: 0.000000 2023-10-11 01:14:01,870 epoch 6 - iter 534/1786 - loss 0.03211693 - time (sec): 165.65 - samples/sec: 453.72 - lr: 0.000084 - momentum: 0.000000 2023-10-11 01:14:56,756 epoch 6 - iter 712/1786 - loss 0.03217743 - time (sec): 220.54 - samples/sec: 449.82 - lr: 0.000082 - momentum: 0.000000 2023-10-11 01:15:49,861 epoch 6 - iter 890/1786 - loss 0.03143026 - time (sec): 273.64 - samples/sec: 449.60 - lr: 0.000080 - momentum: 0.000000 2023-10-11 01:16:43,783 epoch 6 - iter 1068/1786 - loss 0.03104840 - time (sec): 327.56 - samples/sec: 451.43 - lr: 0.000078 - momentum: 0.000000 2023-10-11 01:17:38,716 epoch 6 - iter 1246/1786 - loss 0.03042104 - time (sec): 382.50 - samples/sec: 455.49 - lr: 0.000077 - momentum: 0.000000 2023-10-11 01:18:33,149 epoch 6 - iter 1424/1786 - loss 0.03050314 - time (sec): 436.93 - samples/sec: 455.70 - lr: 0.000075 - momentum: 0.000000 2023-10-11 01:19:26,347 epoch 6 - iter 1602/1786 - loss 0.03080150 - time (sec): 490.13 - samples/sec: 458.40 - lr: 0.000073 - momentum: 0.000000 2023-10-11 01:20:20,368 epoch 6 - iter 1780/1786 - loss 0.03128837 - time (sec): 544.15 - samples/sec: 455.81 - lr: 0.000071 - momentum: 0.000000 2023-10-11 01:20:22,030 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:20:22,030 EPOCH 6 done: loss 0.0312 - lr: 0.000071 2023-10-11 01:20:43,987 DEV : loss 0.1924738883972168 - f1-score (micro avg) 0.7722 2023-10-11 01:20:44,018 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:21:37,787 epoch 7 - iter 178/1786 - loss 0.01587708 - time (sec): 53.77 - samples/sec: 471.93 - lr: 0.000069 - momentum: 0.000000 2023-10-11 01:22:29,472 epoch 7 - iter 356/1786 - loss 0.01825781 - time (sec): 105.45 - samples/sec: 461.97 - lr: 0.000068 - momentum: 0.000000 2023-10-11 01:23:22,449 epoch 7 - iter 534/1786 - loss 0.01732356 - time (sec): 158.43 - samples/sec: 468.28 - lr: 0.000066 - momentum: 0.000000 2023-10-11 01:24:16,983 epoch 7 - iter 712/1786 - loss 0.01837650 - time (sec): 212.96 - samples/sec: 464.65 - lr: 0.000064 - momentum: 0.000000 2023-10-11 01:25:10,070 epoch 7 - iter 890/1786 - loss 0.01927023 - time (sec): 266.05 - samples/sec: 462.18 - lr: 0.000062 - momentum: 0.000000 2023-10-11 01:26:04,003 epoch 7 - iter 1068/1786 - loss 0.01865808 - time (sec): 319.98 - samples/sec: 463.41 - lr: 0.000061 - momentum: 0.000000 2023-10-11 01:26:59,688 epoch 7 - iter 1246/1786 - loss 0.01941833 - time (sec): 375.67 - samples/sec: 460.91 - lr: 0.000059 - momentum: 0.000000 2023-10-11 01:27:51,703 epoch 7 - iter 1424/1786 - loss 0.01972491 - time (sec): 427.68 - samples/sec: 459.36 - lr: 0.000057 - momentum: 0.000000 2023-10-11 01:28:45,857 epoch 7 - iter 1602/1786 - loss 0.02012094 - time (sec): 481.84 - samples/sec: 462.62 - lr: 0.000055 - momentum: 0.000000 2023-10-11 01:29:38,522 epoch 7 - iter 1780/1786 - loss 0.02068942 - time (sec): 534.50 - samples/sec: 464.17 - lr: 0.000053 - momentum: 0.000000 2023-10-11 01:29:40,057 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:29:40,057 EPOCH 7 done: loss 0.0206 - lr: 0.000053 2023-10-11 01:30:01,229 DEV : loss 0.21992838382720947 - f1-score (micro avg) 0.7884 2023-10-11 01:30:01,263 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:30:53,920 epoch 8 - iter 178/1786 - loss 0.01299467 - time (sec): 52.65 - samples/sec: 466.34 - lr: 0.000052 - momentum: 0.000000 2023-10-11 01:31:45,976 epoch 8 - iter 356/1786 - loss 0.01441319 - time (sec): 104.71 - samples/sec: 463.25 - lr: 0.000050 - momentum: 0.000000 2023-10-11 01:32:38,715 epoch 8 - iter 534/1786 - loss 0.01543810 - time (sec): 157.45 - samples/sec: 457.64 - lr: 0.000048 - momentum: 0.000000 2023-10-11 01:33:34,520 epoch 8 - iter 712/1786 - loss 0.01558449 - time (sec): 213.26 - samples/sec: 460.35 - lr: 0.000046 - momentum: 0.000000 2023-10-11 01:34:29,192 epoch 8 - iter 890/1786 - loss 0.01606204 - time (sec): 267.93 - samples/sec: 457.95 - lr: 0.000044 - momentum: 0.000000 2023-10-11 01:35:23,050 epoch 8 - iter 1068/1786 - loss 0.01737134 - time (sec): 321.78 - samples/sec: 453.47 - lr: 0.000043 - momentum: 0.000000 2023-10-11 01:36:18,022 epoch 8 - iter 1246/1786 - loss 0.01717115 - time (sec): 376.76 - samples/sec: 453.38 - lr: 0.000041 - momentum: 0.000000 2023-10-11 01:37:12,230 epoch 8 - iter 1424/1786 - loss 0.01719750 - time (sec): 430.96 - samples/sec: 452.83 - lr: 0.000039 - momentum: 0.000000 2023-10-11 01:38:07,659 epoch 8 - iter 1602/1786 - loss 0.01716691 - time (sec): 486.39 - samples/sec: 454.73 - lr: 0.000037 - momentum: 0.000000 2023-10-11 01:39:04,440 epoch 8 - iter 1780/1786 - loss 0.01671130 - time (sec): 543.18 - samples/sec: 456.11 - lr: 0.000036 - momentum: 0.000000 2023-10-11 01:39:06,365 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:39:06,365 EPOCH 8 done: loss 0.0169 - lr: 0.000036 2023-10-11 01:39:29,230 DEV : loss 0.22382444143295288 - f1-score (micro avg) 0.7783 2023-10-11 01:39:29,261 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:40:24,277 epoch 9 - iter 178/1786 - loss 0.01495584 - time (sec): 55.01 - samples/sec: 452.58 - lr: 0.000034 - momentum: 0.000000 2023-10-11 01:41:18,298 epoch 9 - iter 356/1786 - loss 0.01430322 - time (sec): 109.03 - samples/sec: 447.87 - lr: 0.000032 - momentum: 0.000000 2023-10-11 01:42:13,747 epoch 9 - iter 534/1786 - loss 0.01462408 - time (sec): 164.48 - samples/sec: 454.99 - lr: 0.000030 - momentum: 0.000000 2023-10-11 01:43:06,569 epoch 9 - iter 712/1786 - loss 0.01383529 - time (sec): 217.31 - samples/sec: 449.99 - lr: 0.000028 - momentum: 0.000000 2023-10-11 01:44:00,263 epoch 9 - iter 890/1786 - loss 0.01279666 - time (sec): 271.00 - samples/sec: 449.54 - lr: 0.000027 - momentum: 0.000000 2023-10-11 01:44:55,264 epoch 9 - iter 1068/1786 - loss 0.01221011 - time (sec): 326.00 - samples/sec: 448.49 - lr: 0.000025 - momentum: 0.000000 2023-10-11 01:45:49,115 epoch 9 - iter 1246/1786 - loss 0.01201829 - time (sec): 379.85 - samples/sec: 448.44 - lr: 0.000023 - momentum: 0.000000 2023-10-11 01:46:43,117 epoch 9 - iter 1424/1786 - loss 0.01119229 - time (sec): 433.85 - samples/sec: 451.22 - lr: 0.000021 - momentum: 0.000000 2023-10-11 01:47:37,095 epoch 9 - iter 1602/1786 - loss 0.01109461 - time (sec): 487.83 - samples/sec: 453.37 - lr: 0.000020 - momentum: 0.000000 2023-10-11 01:48:31,386 epoch 9 - iter 1780/1786 - loss 0.01142929 - time (sec): 542.12 - samples/sec: 457.32 - lr: 0.000018 - momentum: 0.000000 2023-10-11 01:48:33,130 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:48:33,131 EPOCH 9 done: loss 0.0115 - lr: 0.000018 2023-10-11 01:48:54,395 DEV : loss 0.23885728418827057 - f1-score (micro avg) 0.7909 2023-10-11 01:48:54,425 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:49:47,365 epoch 10 - iter 178/1786 - loss 0.00824939 - time (sec): 52.94 - samples/sec: 476.73 - lr: 0.000016 - momentum: 0.000000 2023-10-11 01:50:40,024 epoch 10 - iter 356/1786 - loss 0.00754479 - time (sec): 105.60 - samples/sec: 466.45 - lr: 0.000014 - momentum: 0.000000 2023-10-11 01:51:31,760 epoch 10 - iter 534/1786 - loss 0.00886912 - time (sec): 157.33 - samples/sec: 459.24 - lr: 0.000012 - momentum: 0.000000 2023-10-11 01:52:26,703 epoch 10 - iter 712/1786 - loss 0.00857059 - time (sec): 212.28 - samples/sec: 464.60 - lr: 0.000011 - momentum: 0.000000 2023-10-11 01:53:21,373 epoch 10 - iter 890/1786 - loss 0.00828606 - time (sec): 266.95 - samples/sec: 467.58 - lr: 0.000009 - momentum: 0.000000 2023-10-11 01:54:16,053 epoch 10 - iter 1068/1786 - loss 0.00847447 - time (sec): 321.63 - samples/sec: 462.32 - lr: 0.000007 - momentum: 0.000000 2023-10-11 01:55:13,521 epoch 10 - iter 1246/1786 - loss 0.00884080 - time (sec): 379.09 - samples/sec: 461.95 - lr: 0.000005 - momentum: 0.000000 2023-10-11 01:56:09,216 epoch 10 - iter 1424/1786 - loss 0.00980399 - time (sec): 434.79 - samples/sec: 456.91 - lr: 0.000004 - momentum: 0.000000 2023-10-11 01:57:05,107 epoch 10 - iter 1602/1786 - loss 0.00968986 - time (sec): 490.68 - samples/sec: 453.59 - lr: 0.000002 - momentum: 0.000000 2023-10-11 01:58:03,236 epoch 10 - iter 1780/1786 - loss 0.00921449 - time (sec): 548.81 - samples/sec: 451.98 - lr: 0.000000 - momentum: 0.000000 2023-10-11 01:58:05,045 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:58:05,046 EPOCH 10 done: loss 0.0092 - lr: 0.000000 2023-10-11 01:58:26,928 DEV : loss 0.24146509170532227 - f1-score (micro avg) 0.7818 2023-10-11 01:58:27,995 ---------------------------------------------------------------------------------------------------- 2023-10-11 01:58:27,997 Loading model from best epoch ... 2023-10-11 01:58:32,001 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 01:59:39,506 Results: - F-score (micro) 0.6929 - F-score (macro) 0.6294 - Accuracy 0.5446 By class: precision recall f1-score support LOC 0.7365 0.6740 0.7039 1095 PER 0.7930 0.7569 0.7745 1012 ORG 0.4201 0.5742 0.4852 357 HumanProd 0.5625 0.5455 0.5538 33 micro avg 0.6941 0.6916 0.6929 2497 macro avg 0.6280 0.6376 0.6294 2497 weighted avg 0.7119 0.6916 0.6993 2497 2023-10-11 01:59:39,506 ----------------------------------------------------------------------------------------------------