2023-10-13 01:11:03,794 ---------------------------------------------------------------------------------------------------- 2023-10-13 01:11:03,796 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=13, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-13 01:11:03,796 ---------------------------------------------------------------------------------------------------- 2023-10-13 01:11:03,796 MultiCorpus: 7936 train + 992 dev + 992 test sentences - NER_ICDAR_EUROPEANA Corpus: 7936 train + 992 dev + 992 test sentences - /root/.flair/datasets/ner_icdar_europeana/fr 2023-10-13 01:11:03,797 ---------------------------------------------------------------------------------------------------- 2023-10-13 01:11:03,797 Train: 7936 sentences 2023-10-13 01:11:03,797 (train_with_dev=False, train_with_test=False) 2023-10-13 01:11:03,797 ---------------------------------------------------------------------------------------------------- 2023-10-13 01:11:03,797 Training Params: 2023-10-13 01:11:03,797 - learning_rate: "0.00016" 2023-10-13 01:11:03,797 - mini_batch_size: "4" 2023-10-13 01:11:03,797 - max_epochs: "10" 2023-10-13 01:11:03,797 - shuffle: "True" 2023-10-13 01:11:03,797 ---------------------------------------------------------------------------------------------------- 2023-10-13 01:11:03,797 Plugins: 2023-10-13 01:11:03,797 - TensorboardLogger 2023-10-13 01:11:03,797 - LinearScheduler | warmup_fraction: '0.1' 2023-10-13 01:11:03,797 ---------------------------------------------------------------------------------------------------- 2023-10-13 01:11:03,797 Final evaluation on model from best epoch (best-model.pt) 2023-10-13 01:11:03,798 - metric: "('micro avg', 'f1-score')" 2023-10-13 01:11:03,798 ---------------------------------------------------------------------------------------------------- 2023-10-13 01:11:03,798 Computation: 2023-10-13 01:11:03,798 - compute on device: cuda:0 2023-10-13 01:11:03,798 - embedding storage: none 2023-10-13 01:11:03,798 ---------------------------------------------------------------------------------------------------- 2023-10-13 01:11:03,798 Model training base path: "hmbench-icdar/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-3" 2023-10-13 01:11:03,798 ---------------------------------------------------------------------------------------------------- 2023-10-13 01:11:03,798 ---------------------------------------------------------------------------------------------------- 2023-10-13 01:11:03,798 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-13 01:11:57,074 epoch 1 - iter 198/1984 - loss 2.52843731 - time (sec): 53.27 - samples/sec: 301.54 - lr: 0.000016 - momentum: 0.000000 2023-10-13 01:12:49,896 epoch 1 - iter 396/1984 - loss 2.33063285 - time (sec): 106.10 - samples/sec: 299.47 - lr: 0.000032 - momentum: 0.000000 2023-10-13 01:13:43,828 epoch 1 - iter 594/1984 - loss 2.00558938 - time (sec): 160.03 - samples/sec: 304.60 - lr: 0.000048 - momentum: 0.000000 2023-10-13 01:14:37,623 epoch 1 - iter 792/1984 - loss 1.68962898 - time (sec): 213.82 - samples/sec: 305.19 - lr: 0.000064 - momentum: 0.000000 2023-10-13 01:15:32,254 epoch 1 - iter 990/1984 - loss 1.42576251 - time (sec): 268.45 - samples/sec: 305.85 - lr: 0.000080 - momentum: 0.000000 2023-10-13 01:16:30,374 epoch 1 - iter 1188/1984 - loss 1.22765584 - time (sec): 326.57 - samples/sec: 301.16 - lr: 0.000096 - momentum: 0.000000 2023-10-13 01:17:25,376 epoch 1 - iter 1386/1984 - loss 1.09207250 - time (sec): 381.58 - samples/sec: 298.82 - lr: 0.000112 - momentum: 0.000000 2023-10-13 01:18:19,155 epoch 1 - iter 1584/1984 - loss 0.98238438 - time (sec): 435.36 - samples/sec: 299.79 - lr: 0.000128 - momentum: 0.000000 2023-10-13 01:19:13,147 epoch 1 - iter 1782/1984 - loss 0.89132216 - time (sec): 489.35 - samples/sec: 301.35 - lr: 0.000144 - momentum: 0.000000 2023-10-13 01:20:07,717 epoch 1 - iter 1980/1984 - loss 0.81919035 - time (sec): 543.92 - samples/sec: 300.96 - lr: 0.000160 - momentum: 0.000000 2023-10-13 01:20:08,902 ---------------------------------------------------------------------------------------------------- 2023-10-13 01:20:08,902 EPOCH 1 done: loss 0.8182 - lr: 0.000160 2023-10-13 01:20:33,569 DEV : loss 0.13464027643203735 - f1-score (micro avg) 0.6822 2023-10-13 01:20:33,608 saving best model 2023-10-13 01:20:34,548 ---------------------------------------------------------------------------------------------------- 2023-10-13 01:21:27,518 epoch 2 - iter 198/1984 - loss 0.13398518 - time (sec): 52.97 - samples/sec: 317.95 - lr: 0.000158 - momentum: 0.000000 2023-10-13 01:22:20,320 epoch 2 - iter 396/1984 - loss 0.13653987 - time (sec): 105.77 - samples/sec: 312.54 - lr: 0.000156 - momentum: 0.000000 2023-10-13 01:23:13,784 epoch 2 - iter 594/1984 - loss 0.13426706 - time (sec): 159.23 - samples/sec: 309.94 - lr: 0.000155 - momentum: 0.000000 2023-10-13 01:24:07,847 epoch 2 - iter 792/1984 - loss 0.12961703 - time (sec): 213.30 - samples/sec: 310.15 - lr: 0.000153 - momentum: 0.000000 2023-10-13 01:25:00,723 epoch 2 - iter 990/1984 - loss 0.12845842 - time (sec): 266.17 - samples/sec: 308.58 - lr: 0.000151 - momentum: 0.000000 2023-10-13 01:25:57,368 epoch 2 - iter 1188/1984 - loss 0.12621724 - time (sec): 322.82 - samples/sec: 306.90 - lr: 0.000149 - momentum: 0.000000 2023-10-13 01:26:53,050 epoch 2 - iter 1386/1984 - loss 0.12382592 - time (sec): 378.50 - samples/sec: 306.09 - lr: 0.000148 - momentum: 0.000000 2023-10-13 01:27:47,084 epoch 2 - iter 1584/1984 - loss 0.12243932 - time (sec): 432.53 - samples/sec: 304.10 - lr: 0.000146 - momentum: 0.000000 2023-10-13 01:28:40,225 epoch 2 - iter 1782/1984 - loss 0.12044255 - time (sec): 485.67 - samples/sec: 303.81 - lr: 0.000144 - momentum: 0.000000 2023-10-13 01:29:33,733 epoch 2 - iter 1980/1984 - loss 0.11911028 - time (sec): 539.18 - samples/sec: 303.69 - lr: 0.000142 - momentum: 0.000000 2023-10-13 01:29:34,760 ---------------------------------------------------------------------------------------------------- 2023-10-13 01:29:34,760 EPOCH 2 done: loss 0.1191 - lr: 0.000142 2023-10-13 01:30:00,644 DEV : loss 0.09056346118450165 - f1-score (micro avg) 0.7464 2023-10-13 01:30:00,691 saving best model 2023-10-13 01:30:03,443 ---------------------------------------------------------------------------------------------------- 2023-10-13 01:30:58,918 epoch 3 - iter 198/1984 - loss 0.06739931 - time (sec): 55.47 - samples/sec: 309.34 - lr: 0.000140 - momentum: 0.000000 2023-10-13 01:31:51,844 epoch 3 - iter 396/1984 - loss 0.07425965 - time (sec): 108.40 - samples/sec: 306.04 - lr: 0.000139 - momentum: 0.000000 2023-10-13 01:32:45,346 epoch 3 - iter 594/1984 - loss 0.07267235 - time (sec): 161.90 - samples/sec: 302.08 - lr: 0.000137 - momentum: 0.000000 2023-10-13 01:33:40,251 epoch 3 - iter 792/1984 - loss 0.07060065 - time (sec): 216.80 - samples/sec: 298.37 - lr: 0.000135 - momentum: 0.000000 2023-10-13 01:34:34,939 epoch 3 - iter 990/1984 - loss 0.07092729 - time (sec): 271.49 - samples/sec: 299.52 - lr: 0.000133 - momentum: 0.000000 2023-10-13 01:35:31,126 epoch 3 - iter 1188/1984 - loss 0.07276706 - time (sec): 327.68 - samples/sec: 297.65 - lr: 0.000132 - momentum: 0.000000 2023-10-13 01:36:27,297 epoch 3 - iter 1386/1984 - loss 0.07298504 - time (sec): 383.85 - samples/sec: 295.50 - lr: 0.000130 - momentum: 0.000000 2023-10-13 01:37:24,493 epoch 3 - iter 1584/1984 - loss 0.07331926 - time (sec): 441.05 - samples/sec: 294.46 - lr: 0.000128 - momentum: 0.000000 2023-10-13 01:38:20,383 epoch 3 - iter 1782/1984 - loss 0.07273039 - time (sec): 496.94 - samples/sec: 294.70 - lr: 0.000126 - momentum: 0.000000 2023-10-13 01:39:13,609 epoch 3 - iter 1980/1984 - loss 0.07386661 - time (sec): 550.16 - samples/sec: 297.68 - lr: 0.000125 - momentum: 0.000000 2023-10-13 01:39:14,597 ---------------------------------------------------------------------------------------------------- 2023-10-13 01:39:14,597 EPOCH 3 done: loss 0.0738 - lr: 0.000125 2023-10-13 01:39:39,918 DEV : loss 0.10628753900527954 - f1-score (micro avg) 0.7431 2023-10-13 01:39:39,957 ---------------------------------------------------------------------------------------------------- 2023-10-13 01:40:34,507 epoch 4 - iter 198/1984 - loss 0.05002684 - time (sec): 54.55 - samples/sec: 319.28 - lr: 0.000123 - momentum: 0.000000 2023-10-13 01:41:28,212 epoch 4 - iter 396/1984 - loss 0.04642258 - time (sec): 108.25 - samples/sec: 306.78 - lr: 0.000121 - momentum: 0.000000 2023-10-13 01:42:24,209 epoch 4 - iter 594/1984 - loss 0.04922220 - time (sec): 164.25 - samples/sec: 300.82 - lr: 0.000119 - momentum: 0.000000 2023-10-13 01:43:20,372 epoch 4 - iter 792/1984 - loss 0.04903358 - time (sec): 220.41 - samples/sec: 298.20 - lr: 0.000117 - momentum: 0.000000 2023-10-13 01:44:16,017 epoch 4 - iter 990/1984 - loss 0.05035239 - time (sec): 276.06 - samples/sec: 297.50 - lr: 0.000116 - momentum: 0.000000 2023-10-13 01:45:10,220 epoch 4 - iter 1188/1984 - loss 0.05166238 - time (sec): 330.26 - samples/sec: 297.17 - lr: 0.000114 - momentum: 0.000000 2023-10-13 01:46:04,647 epoch 4 - iter 1386/1984 - loss 0.05299053 - time (sec): 384.69 - samples/sec: 295.60 - lr: 0.000112 - momentum: 0.000000 2023-10-13 01:46:57,739 epoch 4 - iter 1584/1984 - loss 0.05360553 - time (sec): 437.78 - samples/sec: 298.38 - lr: 0.000110 - momentum: 0.000000 2023-10-13 01:47:51,321 epoch 4 - iter 1782/1984 - loss 0.05404042 - time (sec): 491.36 - samples/sec: 298.68 - lr: 0.000109 - momentum: 0.000000 2023-10-13 01:48:46,615 epoch 4 - iter 1980/1984 - loss 0.05469939 - time (sec): 546.66 - samples/sec: 299.43 - lr: 0.000107 - momentum: 0.000000 2023-10-13 01:48:47,680 ---------------------------------------------------------------------------------------------------- 2023-10-13 01:48:47,680 EPOCH 4 done: loss 0.0547 - lr: 0.000107 2023-10-13 01:49:13,494 DEV : loss 0.12986594438552856 - f1-score (micro avg) 0.7629 2023-10-13 01:49:13,542 saving best model 2023-10-13 01:49:16,237 ---------------------------------------------------------------------------------------------------- 2023-10-13 01:50:11,842 epoch 5 - iter 198/1984 - loss 0.03937109 - time (sec): 55.60 - samples/sec: 279.25 - lr: 0.000105 - momentum: 0.000000 2023-10-13 01:51:04,893 epoch 5 - iter 396/1984 - loss 0.03149180 - time (sec): 108.65 - samples/sec: 293.70 - lr: 0.000103 - momentum: 0.000000 2023-10-13 01:51:57,276 epoch 5 - iter 594/1984 - loss 0.03526855 - time (sec): 161.03 - samples/sec: 300.86 - lr: 0.000101 - momentum: 0.000000 2023-10-13 01:52:48,839 epoch 5 - iter 792/1984 - loss 0.04002136 - time (sec): 212.59 - samples/sec: 299.54 - lr: 0.000100 - momentum: 0.000000 2023-10-13 01:53:41,046 epoch 5 - iter 990/1984 - loss 0.04019836 - time (sec): 264.80 - samples/sec: 299.44 - lr: 0.000098 - momentum: 0.000000 2023-10-13 01:54:33,468 epoch 5 - iter 1188/1984 - loss 0.03965078 - time (sec): 317.22 - samples/sec: 302.34 - lr: 0.000096 - momentum: 0.000000 2023-10-13 01:55:27,904 epoch 5 - iter 1386/1984 - loss 0.03976973 - time (sec): 371.66 - samples/sec: 306.05 - lr: 0.000094 - momentum: 0.000000 2023-10-13 01:56:21,617 epoch 5 - iter 1584/1984 - loss 0.03952830 - time (sec): 425.37 - samples/sec: 306.42 - lr: 0.000093 - momentum: 0.000000 2023-10-13 01:57:14,389 epoch 5 - iter 1782/1984 - loss 0.04052637 - time (sec): 478.14 - samples/sec: 306.45 - lr: 0.000091 - momentum: 0.000000 2023-10-13 01:58:06,695 epoch 5 - iter 1980/1984 - loss 0.04037425 - time (sec): 530.45 - samples/sec: 308.53 - lr: 0.000089 - momentum: 0.000000 2023-10-13 01:58:07,772 ---------------------------------------------------------------------------------------------------- 2023-10-13 01:58:07,772 EPOCH 5 done: loss 0.0403 - lr: 0.000089 2023-10-13 01:58:31,953 DEV : loss 0.15890514850616455 - f1-score (micro avg) 0.7578 2023-10-13 01:58:31,995 ---------------------------------------------------------------------------------------------------- 2023-10-13 01:59:24,863 epoch 6 - iter 198/1984 - loss 0.03383294 - time (sec): 52.87 - samples/sec: 307.40 - lr: 0.000087 - momentum: 0.000000 2023-10-13 02:00:17,092 epoch 6 - iter 396/1984 - loss 0.03126055 - time (sec): 105.09 - samples/sec: 306.52 - lr: 0.000085 - momentum: 0.000000 2023-10-13 02:01:09,827 epoch 6 - iter 594/1984 - loss 0.03079949 - time (sec): 157.83 - samples/sec: 309.23 - lr: 0.000084 - momentum: 0.000000 2023-10-13 02:02:02,146 epoch 6 - iter 792/1984 - loss 0.03162031 - time (sec): 210.15 - samples/sec: 311.53 - lr: 0.000082 - momentum: 0.000000 2023-10-13 02:02:54,554 epoch 6 - iter 990/1984 - loss 0.03146134 - time (sec): 262.56 - samples/sec: 312.79 - lr: 0.000080 - momentum: 0.000000 2023-10-13 02:03:46,567 epoch 6 - iter 1188/1984 - loss 0.03128278 - time (sec): 314.57 - samples/sec: 311.49 - lr: 0.000078 - momentum: 0.000000 2023-10-13 02:04:39,641 epoch 6 - iter 1386/1984 - loss 0.03149198 - time (sec): 367.64 - samples/sec: 313.24 - lr: 0.000077 - momentum: 0.000000 2023-10-13 02:05:33,650 epoch 6 - iter 1584/1984 - loss 0.02992630 - time (sec): 421.65 - samples/sec: 312.39 - lr: 0.000075 - momentum: 0.000000 2023-10-13 02:06:26,646 epoch 6 - iter 1782/1984 - loss 0.02996741 - time (sec): 474.65 - samples/sec: 312.51 - lr: 0.000073 - momentum: 0.000000 2023-10-13 02:07:18,444 epoch 6 - iter 1980/1984 - loss 0.02958364 - time (sec): 526.45 - samples/sec: 311.08 - lr: 0.000071 - momentum: 0.000000 2023-10-13 02:07:19,461 ---------------------------------------------------------------------------------------------------- 2023-10-13 02:07:19,461 EPOCH 6 done: loss 0.0296 - lr: 0.000071 2023-10-13 02:07:43,565 DEV : loss 0.16922588646411896 - f1-score (micro avg) 0.7613 2023-10-13 02:07:43,608 ---------------------------------------------------------------------------------------------------- 2023-10-13 02:08:34,306 epoch 7 - iter 198/1984 - loss 0.01507566 - time (sec): 50.70 - samples/sec: 318.70 - lr: 0.000069 - momentum: 0.000000 2023-10-13 02:09:25,504 epoch 7 - iter 396/1984 - loss 0.01573958 - time (sec): 101.89 - samples/sec: 323.20 - lr: 0.000068 - momentum: 0.000000 2023-10-13 02:10:18,310 epoch 7 - iter 594/1984 - loss 0.01664591 - time (sec): 154.70 - samples/sec: 316.50 - lr: 0.000066 - momentum: 0.000000 2023-10-13 02:11:10,262 epoch 7 - iter 792/1984 - loss 0.01785772 - time (sec): 206.65 - samples/sec: 316.65 - lr: 0.000064 - momentum: 0.000000 2023-10-13 02:12:01,198 epoch 7 - iter 990/1984 - loss 0.01876625 - time (sec): 257.59 - samples/sec: 316.49 - lr: 0.000062 - momentum: 0.000000 2023-10-13 02:12:52,573 epoch 7 - iter 1188/1984 - loss 0.02003134 - time (sec): 308.96 - samples/sec: 318.37 - lr: 0.000061 - momentum: 0.000000 2023-10-13 02:13:43,862 epoch 7 - iter 1386/1984 - loss 0.01967015 - time (sec): 360.25 - samples/sec: 318.42 - lr: 0.000059 - momentum: 0.000000 2023-10-13 02:14:37,995 epoch 7 - iter 1584/1984 - loss 0.02069974 - time (sec): 414.38 - samples/sec: 316.77 - lr: 0.000057 - momentum: 0.000000 2023-10-13 02:15:35,237 epoch 7 - iter 1782/1984 - loss 0.02041747 - time (sec): 471.63 - samples/sec: 312.32 - lr: 0.000055 - momentum: 0.000000 2023-10-13 02:16:29,103 epoch 7 - iter 1980/1984 - loss 0.02160503 - time (sec): 525.49 - samples/sec: 311.51 - lr: 0.000053 - momentum: 0.000000 2023-10-13 02:16:30,210 ---------------------------------------------------------------------------------------------------- 2023-10-13 02:16:30,210 EPOCH 7 done: loss 0.0217 - lr: 0.000053 2023-10-13 02:16:54,770 DEV : loss 0.19541531801223755 - f1-score (micro avg) 0.7587 2023-10-13 02:16:54,810 ---------------------------------------------------------------------------------------------------- 2023-10-13 02:17:51,202 epoch 8 - iter 198/1984 - loss 0.01913314 - time (sec): 56.39 - samples/sec: 293.35 - lr: 0.000052 - momentum: 0.000000 2023-10-13 02:18:47,470 epoch 8 - iter 396/1984 - loss 0.01758474 - time (sec): 112.66 - samples/sec: 283.41 - lr: 0.000050 - momentum: 0.000000 2023-10-13 02:19:44,479 epoch 8 - iter 594/1984 - loss 0.01501596 - time (sec): 169.67 - samples/sec: 287.41 - lr: 0.000048 - momentum: 0.000000 2023-10-13 02:20:38,244 epoch 8 - iter 792/1984 - loss 0.01524608 - time (sec): 223.43 - samples/sec: 293.67 - lr: 0.000046 - momentum: 0.000000 2023-10-13 02:21:28,379 epoch 8 - iter 990/1984 - loss 0.01572225 - time (sec): 273.57 - samples/sec: 299.81 - lr: 0.000045 - momentum: 0.000000 2023-10-13 02:22:18,130 epoch 8 - iter 1188/1984 - loss 0.01543476 - time (sec): 323.32 - samples/sec: 303.46 - lr: 0.000043 - momentum: 0.000000 2023-10-13 02:23:07,803 epoch 8 - iter 1386/1984 - loss 0.01514254 - time (sec): 372.99 - samples/sec: 306.00 - lr: 0.000041 - momentum: 0.000000 2023-10-13 02:23:58,125 epoch 8 - iter 1584/1984 - loss 0.01446280 - time (sec): 423.31 - samples/sec: 309.49 - lr: 0.000039 - momentum: 0.000000 2023-10-13 02:24:47,909 epoch 8 - iter 1782/1984 - loss 0.01453879 - time (sec): 473.10 - samples/sec: 310.57 - lr: 0.000037 - momentum: 0.000000 2023-10-13 02:25:39,611 epoch 8 - iter 1980/1984 - loss 0.01586744 - time (sec): 524.80 - samples/sec: 311.79 - lr: 0.000036 - momentum: 0.000000 2023-10-13 02:25:40,640 ---------------------------------------------------------------------------------------------------- 2023-10-13 02:25:40,641 EPOCH 8 done: loss 0.0159 - lr: 0.000036 2023-10-13 02:26:07,441 DEV : loss 0.20716699957847595 - f1-score (micro avg) 0.7565 2023-10-13 02:26:07,484 ---------------------------------------------------------------------------------------------------- 2023-10-13 02:27:01,686 epoch 9 - iter 198/1984 - loss 0.00740229 - time (sec): 54.20 - samples/sec: 285.24 - lr: 0.000034 - momentum: 0.000000 2023-10-13 02:27:55,342 epoch 9 - iter 396/1984 - loss 0.00741304 - time (sec): 107.86 - samples/sec: 286.26 - lr: 0.000032 - momentum: 0.000000 2023-10-13 02:28:48,271 epoch 9 - iter 594/1984 - loss 0.00953282 - time (sec): 160.78 - samples/sec: 293.98 - lr: 0.000030 - momentum: 0.000000 2023-10-13 02:29:41,888 epoch 9 - iter 792/1984 - loss 0.00987460 - time (sec): 214.40 - samples/sec: 299.90 - lr: 0.000029 - momentum: 0.000000 2023-10-13 02:30:34,733 epoch 9 - iter 990/1984 - loss 0.00972689 - time (sec): 267.25 - samples/sec: 302.63 - lr: 0.000027 - momentum: 0.000000 2023-10-13 02:31:26,785 epoch 9 - iter 1188/1984 - loss 0.00991641 - time (sec): 319.30 - samples/sec: 308.78 - lr: 0.000025 - momentum: 0.000000 2023-10-13 02:32:18,304 epoch 9 - iter 1386/1984 - loss 0.01028782 - time (sec): 370.82 - samples/sec: 311.33 - lr: 0.000023 - momentum: 0.000000 2023-10-13 02:33:10,088 epoch 9 - iter 1584/1984 - loss 0.01070081 - time (sec): 422.60 - samples/sec: 312.62 - lr: 0.000021 - momentum: 0.000000 2023-10-13 02:34:02,588 epoch 9 - iter 1782/1984 - loss 0.01055651 - time (sec): 475.10 - samples/sec: 312.75 - lr: 0.000020 - momentum: 0.000000 2023-10-13 02:34:54,501 epoch 9 - iter 1980/1984 - loss 0.01028953 - time (sec): 527.01 - samples/sec: 310.49 - lr: 0.000018 - momentum: 0.000000 2023-10-13 02:34:55,569 ---------------------------------------------------------------------------------------------------- 2023-10-13 02:34:55,570 EPOCH 9 done: loss 0.0103 - lr: 0.000018 2023-10-13 02:35:21,677 DEV : loss 0.2197086364030838 - f1-score (micro avg) 0.7675 2023-10-13 02:35:21,722 saving best model 2023-10-13 02:35:24,408 ---------------------------------------------------------------------------------------------------- 2023-10-13 02:36:16,693 epoch 10 - iter 198/1984 - loss 0.00313690 - time (sec): 52.28 - samples/sec: 315.59 - lr: 0.000016 - momentum: 0.000000 2023-10-13 02:37:09,099 epoch 10 - iter 396/1984 - loss 0.00558899 - time (sec): 104.69 - samples/sec: 317.42 - lr: 0.000014 - momentum: 0.000000 2023-10-13 02:38:01,060 epoch 10 - iter 594/1984 - loss 0.00556445 - time (sec): 156.65 - samples/sec: 319.86 - lr: 0.000013 - momentum: 0.000000 2023-10-13 02:38:53,915 epoch 10 - iter 792/1984 - loss 0.00552369 - time (sec): 209.50 - samples/sec: 315.53 - lr: 0.000011 - momentum: 0.000000 2023-10-13 02:39:46,052 epoch 10 - iter 990/1984 - loss 0.00607246 - time (sec): 261.64 - samples/sec: 315.00 - lr: 0.000009 - momentum: 0.000000 2023-10-13 02:40:39,710 epoch 10 - iter 1188/1984 - loss 0.00683260 - time (sec): 315.30 - samples/sec: 311.80 - lr: 0.000007 - momentum: 0.000000 2023-10-13 02:41:32,898 epoch 10 - iter 1386/1984 - loss 0.00736140 - time (sec): 368.49 - samples/sec: 312.14 - lr: 0.000005 - momentum: 0.000000 2023-10-13 02:42:26,307 epoch 10 - iter 1584/1984 - loss 0.00744783 - time (sec): 421.89 - samples/sec: 310.11 - lr: 0.000004 - momentum: 0.000000 2023-10-13 02:43:19,462 epoch 10 - iter 1782/1984 - loss 0.00740743 - time (sec): 475.05 - samples/sec: 311.13 - lr: 0.000002 - momentum: 0.000000 2023-10-13 02:44:11,558 epoch 10 - iter 1980/1984 - loss 0.00755919 - time (sec): 527.14 - samples/sec: 310.50 - lr: 0.000000 - momentum: 0.000000 2023-10-13 02:44:12,617 ---------------------------------------------------------------------------------------------------- 2023-10-13 02:44:12,618 EPOCH 10 done: loss 0.0076 - lr: 0.000000 2023-10-13 02:44:38,438 DEV : loss 0.22709771990776062 - f1-score (micro avg) 0.769 2023-10-13 02:44:38,479 saving best model 2023-10-13 02:44:42,054 ---------------------------------------------------------------------------------------------------- 2023-10-13 02:44:42,057 Loading model from best epoch ... 2023-10-13 02:44:47,171 SequenceTagger predicts: Dictionary with 13 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 2023-10-13 02:45:11,248 Results: - F-score (micro) 0.7627 - F-score (macro) 0.6811 - Accuracy 0.6449 By class: precision recall f1-score support LOC 0.7997 0.8351 0.8170 655 PER 0.6965 0.8027 0.7458 223 ORG 0.5392 0.4331 0.4803 127 micro avg 0.7488 0.7771 0.7627 1005 macro avg 0.6785 0.6903 0.6811 1005 weighted avg 0.7439 0.7771 0.7587 1005 2023-10-13 02:45:11,248 ----------------------------------------------------------------------------------------------------