2023-10-11 13:00:47,816 ---------------------------------------------------------------------------------------------------- 2023-10-11 13:00:47,818 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 13:00:47,818 ---------------------------------------------------------------------------------------------------- 2023-10-11 13:00:47,818 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 13:00:47,818 ---------------------------------------------------------------------------------------------------- 2023-10-11 13:00:47,818 Train: 7142 sentences 2023-10-11 13:00:47,818 (train_with_dev=False, train_with_test=False) 2023-10-11 13:00:47,819 ---------------------------------------------------------------------------------------------------- 2023-10-11 13:00:47,819 Training Params: 2023-10-11 13:00:47,819 - learning_rate: "0.00016" 2023-10-11 13:00:47,819 - mini_batch_size: "4" 2023-10-11 13:00:47,819 - max_epochs: "10" 2023-10-11 13:00:47,819 - shuffle: "True" 2023-10-11 13:00:47,819 ---------------------------------------------------------------------------------------------------- 2023-10-11 13:00:47,819 Plugins: 2023-10-11 13:00:47,819 - TensorboardLogger 2023-10-11 13:00:47,819 - LinearScheduler | warmup_fraction: '0.1' 2023-10-11 13:00:47,819 ---------------------------------------------------------------------------------------------------- 2023-10-11 13:00:47,819 Final evaluation on model from best epoch (best-model.pt) 2023-10-11 13:00:47,819 - metric: "('micro avg', 'f1-score')" 2023-10-11 13:00:47,820 ---------------------------------------------------------------------------------------------------- 2023-10-11 13:00:47,820 Computation: 2023-10-11 13:00:47,820 - compute on device: cuda:0 2023-10-11 13:00:47,820 - embedding storage: none 2023-10-11 13:00:47,820 ---------------------------------------------------------------------------------------------------- 2023-10-11 13:00:47,820 Model training base path: "hmbench-newseye/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-3" 2023-10-11 13:00:47,820 ---------------------------------------------------------------------------------------------------- 2023-10-11 13:00:47,820 ---------------------------------------------------------------------------------------------------- 2023-10-11 13:00:47,820 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-11 13:01:44,208 epoch 1 - iter 178/1786 - loss 2.81170991 - time (sec): 56.38 - samples/sec: 477.56 - lr: 0.000016 - momentum: 0.000000 2023-10-11 13:02:39,090 epoch 1 - iter 356/1786 - loss 2.63745970 - time (sec): 111.27 - samples/sec: 455.58 - lr: 0.000032 - momentum: 0.000000 2023-10-11 13:03:34,033 epoch 1 - iter 534/1786 - loss 2.35715581 - time (sec): 166.21 - samples/sec: 448.32 - lr: 0.000048 - momentum: 0.000000 2023-10-11 13:04:29,519 epoch 1 - iter 712/1786 - loss 2.06074361 - time (sec): 221.70 - samples/sec: 445.36 - lr: 0.000064 - momentum: 0.000000 2023-10-11 13:05:26,211 epoch 1 - iter 890/1786 - loss 1.78883260 - time (sec): 278.39 - samples/sec: 448.55 - lr: 0.000080 - momentum: 0.000000 2023-10-11 13:06:20,110 epoch 1 - iter 1068/1786 - loss 1.58610501 - time (sec): 332.29 - samples/sec: 447.69 - lr: 0.000096 - momentum: 0.000000 2023-10-11 13:07:15,433 epoch 1 - iter 1246/1786 - loss 1.40913339 - time (sec): 387.61 - samples/sec: 449.86 - lr: 0.000112 - momentum: 0.000000 2023-10-11 13:08:13,637 epoch 1 - iter 1424/1786 - loss 1.27901313 - time (sec): 445.81 - samples/sec: 446.05 - lr: 0.000127 - momentum: 0.000000 2023-10-11 13:09:09,748 epoch 1 - iter 1602/1786 - loss 1.16574649 - time (sec): 501.92 - samples/sec: 446.37 - lr: 0.000143 - momentum: 0.000000 2023-10-11 13:10:05,183 epoch 1 - iter 1780/1786 - loss 1.07906676 - time (sec): 557.36 - samples/sec: 444.83 - lr: 0.000159 - momentum: 0.000000 2023-10-11 13:10:06,873 ---------------------------------------------------------------------------------------------------- 2023-10-11 13:10:06,874 EPOCH 1 done: loss 1.0763 - lr: 0.000159 2023-10-11 13:10:27,235 DEV : loss 0.17565929889678955 - f1-score (micro avg) 0.6003 2023-10-11 13:10:27,268 saving best model 2023-10-11 13:10:28,252 ---------------------------------------------------------------------------------------------------- 2023-10-11 13:11:24,075 epoch 2 - iter 178/1786 - loss 0.16255892 - time (sec): 55.82 - samples/sec: 470.03 - lr: 0.000158 - momentum: 0.000000 2023-10-11 13:12:20,568 epoch 2 - iter 356/1786 - loss 0.16881265 - time (sec): 112.31 - samples/sec: 458.36 - lr: 0.000156 - momentum: 0.000000 2023-10-11 13:13:19,344 epoch 2 - iter 534/1786 - loss 0.15873645 - time (sec): 171.09 - samples/sec: 440.45 - lr: 0.000155 - momentum: 0.000000 2023-10-11 13:14:17,771 epoch 2 - iter 712/1786 - loss 0.14765138 - time (sec): 229.52 - samples/sec: 438.98 - lr: 0.000153 - momentum: 0.000000 2023-10-11 13:15:13,377 epoch 2 - iter 890/1786 - loss 0.14254566 - time (sec): 285.12 - samples/sec: 436.39 - lr: 0.000151 - momentum: 0.000000 2023-10-11 13:16:14,773 epoch 2 - iter 1068/1786 - loss 0.13991538 - time (sec): 346.52 - samples/sec: 433.54 - lr: 0.000149 - momentum: 0.000000 2023-10-11 13:17:16,141 epoch 2 - iter 1246/1786 - loss 0.13851750 - time (sec): 407.89 - samples/sec: 428.23 - lr: 0.000148 - momentum: 0.000000 2023-10-11 13:18:15,089 epoch 2 - iter 1424/1786 - loss 0.13580485 - time (sec): 466.83 - samples/sec: 424.05 - lr: 0.000146 - momentum: 0.000000 2023-10-11 13:19:09,450 epoch 2 - iter 1602/1786 - loss 0.13348564 - time (sec): 521.20 - samples/sec: 426.97 - lr: 0.000144 - momentum: 0.000000 2023-10-11 13:20:06,575 epoch 2 - iter 1780/1786 - loss 0.13055574 - time (sec): 578.32 - samples/sec: 428.70 - lr: 0.000142 - momentum: 0.000000 2023-10-11 13:20:08,493 ---------------------------------------------------------------------------------------------------- 2023-10-11 13:20:08,493 EPOCH 2 done: loss 0.1303 - lr: 0.000142 2023-10-11 13:20:30,672 DEV : loss 0.10548630356788635 - f1-score (micro avg) 0.7578 2023-10-11 13:20:30,703 saving best model 2023-10-11 13:20:33,274 ---------------------------------------------------------------------------------------------------- 2023-10-11 13:21:25,594 epoch 3 - iter 178/1786 - loss 0.06430020 - time (sec): 52.32 - samples/sec: 455.81 - lr: 0.000140 - momentum: 0.000000 2023-10-11 13:22:18,793 epoch 3 - iter 356/1786 - loss 0.06283356 - time (sec): 105.51 - samples/sec: 463.47 - lr: 0.000139 - momentum: 0.000000 2023-10-11 13:23:10,053 epoch 3 - iter 534/1786 - loss 0.06385061 - time (sec): 156.77 - samples/sec: 466.82 - lr: 0.000137 - momentum: 0.000000 2023-10-11 13:24:04,302 epoch 3 - iter 712/1786 - loss 0.06659712 - time (sec): 211.02 - samples/sec: 465.28 - lr: 0.000135 - momentum: 0.000000 2023-10-11 13:24:56,648 epoch 3 - iter 890/1786 - loss 0.07114886 - time (sec): 263.37 - samples/sec: 466.58 - lr: 0.000133 - momentum: 0.000000 2023-10-11 13:25:48,940 epoch 3 - iter 1068/1786 - loss 0.07352806 - time (sec): 315.66 - samples/sec: 467.30 - lr: 0.000132 - momentum: 0.000000 2023-10-11 13:26:45,407 epoch 3 - iter 1246/1786 - loss 0.07514593 - time (sec): 372.13 - samples/sec: 468.01 - lr: 0.000130 - momentum: 0.000000 2023-10-11 13:27:42,174 epoch 3 - iter 1424/1786 - loss 0.07344231 - time (sec): 428.90 - samples/sec: 461.84 - lr: 0.000128 - momentum: 0.000000 2023-10-11 13:28:39,296 epoch 3 - iter 1602/1786 - loss 0.07210740 - time (sec): 486.02 - samples/sec: 459.31 - lr: 0.000126 - momentum: 0.000000 2023-10-11 13:29:36,366 epoch 3 - iter 1780/1786 - loss 0.07250482 - time (sec): 543.09 - samples/sec: 456.92 - lr: 0.000125 - momentum: 0.000000 2023-10-11 13:29:38,029 ---------------------------------------------------------------------------------------------------- 2023-10-11 13:29:38,029 EPOCH 3 done: loss 0.0726 - lr: 0.000125 2023-10-11 13:30:01,230 DEV : loss 0.10552459955215454 - f1-score (micro avg) 0.7957 2023-10-11 13:30:01,262 saving best model 2023-10-11 13:30:03,958 ---------------------------------------------------------------------------------------------------- 2023-10-11 13:30:59,030 epoch 4 - iter 178/1786 - loss 0.04332888 - time (sec): 55.07 - samples/sec: 436.30 - lr: 0.000123 - momentum: 0.000000 2023-10-11 13:31:55,067 epoch 4 - iter 356/1786 - loss 0.04883610 - time (sec): 111.10 - samples/sec: 441.19 - lr: 0.000121 - momentum: 0.000000 2023-10-11 13:32:50,737 epoch 4 - iter 534/1786 - loss 0.04847201 - time (sec): 166.77 - samples/sec: 453.36 - lr: 0.000119 - momentum: 0.000000 2023-10-11 13:33:44,707 epoch 4 - iter 712/1786 - loss 0.05062238 - time (sec): 220.74 - samples/sec: 453.88 - lr: 0.000117 - momentum: 0.000000 2023-10-11 13:34:38,564 epoch 4 - iter 890/1786 - loss 0.05112561 - time (sec): 274.60 - samples/sec: 458.03 - lr: 0.000116 - momentum: 0.000000 2023-10-11 13:35:31,790 epoch 4 - iter 1068/1786 - loss 0.05210222 - time (sec): 327.83 - samples/sec: 453.38 - lr: 0.000114 - momentum: 0.000000 2023-10-11 13:36:30,302 epoch 4 - iter 1246/1786 - loss 0.05253056 - time (sec): 386.34 - samples/sec: 449.69 - lr: 0.000112 - momentum: 0.000000 2023-10-11 13:37:25,466 epoch 4 - iter 1424/1786 - loss 0.05198458 - time (sec): 441.50 - samples/sec: 448.36 - lr: 0.000110 - momentum: 0.000000 2023-10-11 13:38:23,731 epoch 4 - iter 1602/1786 - loss 0.05147654 - time (sec): 499.77 - samples/sec: 446.38 - lr: 0.000109 - momentum: 0.000000 2023-10-11 13:39:18,931 epoch 4 - iter 1780/1786 - loss 0.05054278 - time (sec): 554.97 - samples/sec: 446.88 - lr: 0.000107 - momentum: 0.000000 2023-10-11 13:39:20,569 ---------------------------------------------------------------------------------------------------- 2023-10-11 13:39:20,569 EPOCH 4 done: loss 0.0505 - lr: 0.000107 2023-10-11 13:39:43,214 DEV : loss 0.156653493642807 - f1-score (micro avg) 0.7909 2023-10-11 13:39:43,255 ---------------------------------------------------------------------------------------------------- 2023-10-11 13:40:43,029 epoch 5 - iter 178/1786 - loss 0.03328650 - time (sec): 59.77 - samples/sec: 420.20 - lr: 0.000105 - momentum: 0.000000 2023-10-11 13:41:39,262 epoch 5 - iter 356/1786 - loss 0.04050878 - time (sec): 116.00 - samples/sec: 436.17 - lr: 0.000103 - momentum: 0.000000 2023-10-11 13:42:33,938 epoch 5 - iter 534/1786 - loss 0.03621945 - time (sec): 170.68 - samples/sec: 443.11 - lr: 0.000101 - momentum: 0.000000 2023-10-11 13:43:27,857 epoch 5 - iter 712/1786 - loss 0.03475165 - time (sec): 224.60 - samples/sec: 444.06 - lr: 0.000100 - momentum: 0.000000 2023-10-11 13:44:20,262 epoch 5 - iter 890/1786 - loss 0.03461616 - time (sec): 277.00 - samples/sec: 446.22 - lr: 0.000098 - momentum: 0.000000 2023-10-11 13:45:14,874 epoch 5 - iter 1068/1786 - loss 0.03386874 - time (sec): 331.62 - samples/sec: 445.15 - lr: 0.000096 - momentum: 0.000000 2023-10-11 13:46:12,877 epoch 5 - iter 1246/1786 - loss 0.03430305 - time (sec): 389.62 - samples/sec: 443.88 - lr: 0.000094 - momentum: 0.000000 2023-10-11 13:47:13,910 epoch 5 - iter 1424/1786 - loss 0.03506913 - time (sec): 450.65 - samples/sec: 437.96 - lr: 0.000093 - momentum: 0.000000 2023-10-11 13:48:13,048 epoch 5 - iter 1602/1786 - loss 0.03525813 - time (sec): 509.79 - samples/sec: 436.07 - lr: 0.000091 - momentum: 0.000000 2023-10-11 13:49:06,991 epoch 5 - iter 1780/1786 - loss 0.03807119 - time (sec): 563.73 - samples/sec: 440.04 - lr: 0.000089 - momentum: 0.000000 2023-10-11 13:49:08,616 ---------------------------------------------------------------------------------------------------- 2023-10-11 13:49:08,616 EPOCH 5 done: loss 0.0381 - lr: 0.000089 2023-10-11 13:49:30,953 DEV : loss 0.17025883495807648 - f1-score (micro avg) 0.788 2023-10-11 13:49:30,986 ---------------------------------------------------------------------------------------------------- 2023-10-11 13:50:30,448 epoch 6 - iter 178/1786 - loss 0.03037012 - time (sec): 59.46 - samples/sec: 437.48 - lr: 0.000087 - momentum: 0.000000 2023-10-11 13:51:23,893 epoch 6 - iter 356/1786 - loss 0.02735029 - time (sec): 112.90 - samples/sec: 439.65 - lr: 0.000085 - momentum: 0.000000 2023-10-11 13:52:18,715 epoch 6 - iter 534/1786 - loss 0.02908184 - time (sec): 167.73 - samples/sec: 439.17 - lr: 0.000084 - momentum: 0.000000 2023-10-11 13:53:17,235 epoch 6 - iter 712/1786 - loss 0.02904645 - time (sec): 226.25 - samples/sec: 438.56 - lr: 0.000082 - momentum: 0.000000 2023-10-11 13:54:11,434 epoch 6 - iter 890/1786 - loss 0.02790749 - time (sec): 280.45 - samples/sec: 441.20 - lr: 0.000080 - momentum: 0.000000 2023-10-11 13:55:03,988 epoch 6 - iter 1068/1786 - loss 0.02701632 - time (sec): 333.00 - samples/sec: 441.61 - lr: 0.000078 - momentum: 0.000000 2023-10-11 13:55:57,385 epoch 6 - iter 1246/1786 - loss 0.02648261 - time (sec): 386.40 - samples/sec: 444.67 - lr: 0.000077 - momentum: 0.000000 2023-10-11 13:56:56,380 epoch 6 - iter 1424/1786 - loss 0.02759848 - time (sec): 445.39 - samples/sec: 444.90 - lr: 0.000075 - momentum: 0.000000 2023-10-11 13:57:51,975 epoch 6 - iter 1602/1786 - loss 0.02741452 - time (sec): 500.99 - samples/sec: 445.51 - lr: 0.000073 - momentum: 0.000000 2023-10-11 13:58:44,618 epoch 6 - iter 1780/1786 - loss 0.02735127 - time (sec): 553.63 - samples/sec: 448.07 - lr: 0.000071 - momentum: 0.000000 2023-10-11 13:58:46,201 ---------------------------------------------------------------------------------------------------- 2023-10-11 13:58:46,201 EPOCH 6 done: loss 0.0273 - lr: 0.000071 2023-10-11 13:59:06,894 DEV : loss 0.18334950506687164 - f1-score (micro avg) 0.8033 2023-10-11 13:59:06,940 saving best model 2023-10-11 13:59:09,574 ---------------------------------------------------------------------------------------------------- 2023-10-11 14:00:03,722 epoch 7 - iter 178/1786 - loss 0.02116711 - time (sec): 54.14 - samples/sec: 452.76 - lr: 0.000069 - momentum: 0.000000 2023-10-11 14:00:56,097 epoch 7 - iter 356/1786 - loss 0.02358540 - time (sec): 106.52 - samples/sec: 448.60 - lr: 0.000068 - momentum: 0.000000 2023-10-11 14:01:50,382 epoch 7 - iter 534/1786 - loss 0.02380570 - time (sec): 160.80 - samples/sec: 454.36 - lr: 0.000066 - momentum: 0.000000 2023-10-11 14:02:44,369 epoch 7 - iter 712/1786 - loss 0.02389820 - time (sec): 214.79 - samples/sec: 452.33 - lr: 0.000064 - momentum: 0.000000 2023-10-11 14:03:38,758 epoch 7 - iter 890/1786 - loss 0.02394287 - time (sec): 269.18 - samples/sec: 454.52 - lr: 0.000062 - momentum: 0.000000 2023-10-11 14:04:33,458 epoch 7 - iter 1068/1786 - loss 0.02260999 - time (sec): 323.88 - samples/sec: 456.84 - lr: 0.000061 - momentum: 0.000000 2023-10-11 14:05:28,679 epoch 7 - iter 1246/1786 - loss 0.02182956 - time (sec): 379.10 - samples/sec: 456.41 - lr: 0.000059 - momentum: 0.000000 2023-10-11 14:06:22,977 epoch 7 - iter 1424/1786 - loss 0.02105207 - time (sec): 433.40 - samples/sec: 457.13 - lr: 0.000057 - momentum: 0.000000 2023-10-11 14:07:16,767 epoch 7 - iter 1602/1786 - loss 0.02145575 - time (sec): 487.19 - samples/sec: 458.26 - lr: 0.000055 - momentum: 0.000000 2023-10-11 14:08:09,598 epoch 7 - iter 1780/1786 - loss 0.02106435 - time (sec): 540.02 - samples/sec: 459.55 - lr: 0.000053 - momentum: 0.000000 2023-10-11 14:08:11,066 ---------------------------------------------------------------------------------------------------- 2023-10-11 14:08:11,067 EPOCH 7 done: loss 0.0211 - lr: 0.000053 2023-10-11 14:08:32,426 DEV : loss 0.194308340549469 - f1-score (micro avg) 0.7928 2023-10-11 14:08:32,455 ---------------------------------------------------------------------------------------------------- 2023-10-11 14:09:23,421 epoch 8 - iter 178/1786 - loss 0.01641532 - time (sec): 50.96 - samples/sec: 484.58 - lr: 0.000052 - momentum: 0.000000 2023-10-11 14:10:14,666 epoch 8 - iter 356/1786 - loss 0.01440783 - time (sec): 102.21 - samples/sec: 483.68 - lr: 0.000050 - momentum: 0.000000 2023-10-11 14:11:06,282 epoch 8 - iter 534/1786 - loss 0.01285935 - time (sec): 153.83 - samples/sec: 473.03 - lr: 0.000048 - momentum: 0.000000 2023-10-11 14:11:58,394 epoch 8 - iter 712/1786 - loss 0.01220440 - time (sec): 205.94 - samples/sec: 467.19 - lr: 0.000046 - momentum: 0.000000 2023-10-11 14:12:50,304 epoch 8 - iter 890/1786 - loss 0.01374698 - time (sec): 257.85 - samples/sec: 465.93 - lr: 0.000044 - momentum: 0.000000 2023-10-11 14:13:44,353 epoch 8 - iter 1068/1786 - loss 0.01485753 - time (sec): 311.90 - samples/sec: 469.85 - lr: 0.000043 - momentum: 0.000000 2023-10-11 14:14:37,502 epoch 8 - iter 1246/1786 - loss 0.01514008 - time (sec): 365.05 - samples/sec: 472.32 - lr: 0.000041 - momentum: 0.000000 2023-10-11 14:15:30,709 epoch 8 - iter 1424/1786 - loss 0.01557502 - time (sec): 418.25 - samples/sec: 475.69 - lr: 0.000039 - momentum: 0.000000 2023-10-11 14:16:22,878 epoch 8 - iter 1602/1786 - loss 0.01641034 - time (sec): 470.42 - samples/sec: 477.11 - lr: 0.000037 - momentum: 0.000000 2023-10-11 14:17:13,034 epoch 8 - iter 1780/1786 - loss 0.01592313 - time (sec): 520.58 - samples/sec: 476.58 - lr: 0.000036 - momentum: 0.000000 2023-10-11 14:17:14,558 ---------------------------------------------------------------------------------------------------- 2023-10-11 14:17:14,559 EPOCH 8 done: loss 0.0159 - lr: 0.000036 2023-10-11 14:17:36,426 DEV : loss 0.2070261836051941 - f1-score (micro avg) 0.7869 2023-10-11 14:17:36,457 ---------------------------------------------------------------------------------------------------- 2023-10-11 14:18:28,783 epoch 9 - iter 178/1786 - loss 0.01279919 - time (sec): 52.32 - samples/sec: 455.83 - lr: 0.000034 - momentum: 0.000000 2023-10-11 14:19:20,233 epoch 9 - iter 356/1786 - loss 0.01083550 - time (sec): 103.77 - samples/sec: 450.34 - lr: 0.000032 - momentum: 0.000000 2023-10-11 14:20:11,070 epoch 9 - iter 534/1786 - loss 0.01073281 - time (sec): 154.61 - samples/sec: 445.00 - lr: 0.000030 - momentum: 0.000000 2023-10-11 14:21:04,072 epoch 9 - iter 712/1786 - loss 0.01005165 - time (sec): 207.61 - samples/sec: 457.10 - lr: 0.000028 - momentum: 0.000000 2023-10-11 14:21:56,520 epoch 9 - iter 890/1786 - loss 0.01008754 - time (sec): 260.06 - samples/sec: 462.46 - lr: 0.000027 - momentum: 0.000000 2023-10-11 14:22:50,929 epoch 9 - iter 1068/1786 - loss 0.01062970 - time (sec): 314.47 - samples/sec: 464.76 - lr: 0.000025 - momentum: 0.000000 2023-10-11 14:23:44,599 epoch 9 - iter 1246/1786 - loss 0.01091575 - time (sec): 368.14 - samples/sec: 468.87 - lr: 0.000023 - momentum: 0.000000 2023-10-11 14:24:38,320 epoch 9 - iter 1424/1786 - loss 0.01047999 - time (sec): 421.86 - samples/sec: 470.87 - lr: 0.000021 - momentum: 0.000000 2023-10-11 14:25:31,677 epoch 9 - iter 1602/1786 - loss 0.01056976 - time (sec): 475.22 - samples/sec: 470.42 - lr: 0.000020 - momentum: 0.000000 2023-10-11 14:26:25,182 epoch 9 - iter 1780/1786 - loss 0.01085308 - time (sec): 528.72 - samples/sec: 468.44 - lr: 0.000018 - momentum: 0.000000 2023-10-11 14:26:27,252 ---------------------------------------------------------------------------------------------------- 2023-10-11 14:26:27,252 EPOCH 9 done: loss 0.0109 - lr: 0.000018 2023-10-11 14:26:49,598 DEV : loss 0.21609559655189514 - f1-score (micro avg) 0.8054 2023-10-11 14:26:49,638 saving best model 2023-10-11 14:26:52,383 ---------------------------------------------------------------------------------------------------- 2023-10-11 14:27:45,526 epoch 10 - iter 178/1786 - loss 0.00795703 - time (sec): 53.14 - samples/sec: 469.48 - lr: 0.000016 - momentum: 0.000000 2023-10-11 14:28:38,444 epoch 10 - iter 356/1786 - loss 0.00911006 - time (sec): 106.06 - samples/sec: 453.61 - lr: 0.000014 - momentum: 0.000000 2023-10-11 14:29:32,746 epoch 10 - iter 534/1786 - loss 0.00807375 - time (sec): 160.36 - samples/sec: 453.72 - lr: 0.000012 - momentum: 0.000000 2023-10-11 14:30:25,481 epoch 10 - iter 712/1786 - loss 0.00794053 - time (sec): 213.09 - samples/sec: 461.38 - lr: 0.000011 - momentum: 0.000000 2023-10-11 14:31:18,756 epoch 10 - iter 890/1786 - loss 0.00822094 - time (sec): 266.37 - samples/sec: 466.30 - lr: 0.000009 - momentum: 0.000000 2023-10-11 14:32:11,427 epoch 10 - iter 1068/1786 - loss 0.00765007 - time (sec): 319.04 - samples/sec: 466.07 - lr: 0.000007 - momentum: 0.000000 2023-10-11 14:33:02,846 epoch 10 - iter 1246/1786 - loss 0.00782413 - time (sec): 370.46 - samples/sec: 466.16 - lr: 0.000005 - momentum: 0.000000 2023-10-11 14:33:55,039 epoch 10 - iter 1424/1786 - loss 0.00755323 - time (sec): 422.65 - samples/sec: 467.52 - lr: 0.000004 - momentum: 0.000000 2023-10-11 14:34:48,378 epoch 10 - iter 1602/1786 - loss 0.00741746 - time (sec): 475.99 - samples/sec: 467.39 - lr: 0.000002 - momentum: 0.000000 2023-10-11 14:35:41,429 epoch 10 - iter 1780/1786 - loss 0.00748365 - time (sec): 529.04 - samples/sec: 469.23 - lr: 0.000000 - momentum: 0.000000 2023-10-11 14:35:42,901 ---------------------------------------------------------------------------------------------------- 2023-10-11 14:35:42,901 EPOCH 10 done: loss 0.0075 - lr: 0.000000 2023-10-11 14:36:04,553 DEV : loss 0.22351031005382538 - f1-score (micro avg) 0.7973 2023-10-11 14:36:05,510 ---------------------------------------------------------------------------------------------------- 2023-10-11 14:36:05,512 Loading model from best epoch ... 2023-10-11 14:36:09,777 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 14:37:17,218 Results: - F-score (micro) 0.7051 - F-score (macro) 0.6521 - Accuracy 0.5611 By class: precision recall f1-score support LOC 0.6985 0.7342 0.7159 1095 PER 0.7796 0.7688 0.7741 1012 ORG 0.4577 0.5910 0.5159 357 HumanProd 0.5000 0.7576 0.6024 33 micro avg 0.6835 0.7281 0.7051 2497 macro avg 0.6089 0.7129 0.6521 2497 weighted avg 0.6943 0.7281 0.7094 2497 2023-10-11 14:37:17,218 ----------------------------------------------------------------------------------------------------