2023-10-13 07:09:19,789 ---------------------------------------------------------------------------------------------------- 2023-10-13 07:09:19,791 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 07:09:19,791 ---------------------------------------------------------------------------------------------------- 2023-10-13 07:09:19,791 MultiCorpus: 14465 train + 1392 dev + 2432 test sentences - NER_HIPE_2022 Corpus: 14465 train + 1392 dev + 2432 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/letemps/fr/with_doc_seperator 2023-10-13 07:09:19,791 ---------------------------------------------------------------------------------------------------- 2023-10-13 07:09:19,791 Train: 14465 sentences 2023-10-13 07:09:19,791 (train_with_dev=False, train_with_test=False) 2023-10-13 07:09:19,792 ---------------------------------------------------------------------------------------------------- 2023-10-13 07:09:19,792 Training Params: 2023-10-13 07:09:19,792 - learning_rate: "0.00016" 2023-10-13 07:09:19,792 - mini_batch_size: "4" 2023-10-13 07:09:19,792 - max_epochs: "10" 2023-10-13 07:09:19,792 - shuffle: "True" 2023-10-13 07:09:19,792 ---------------------------------------------------------------------------------------------------- 2023-10-13 07:09:19,792 Plugins: 2023-10-13 07:09:19,792 - TensorboardLogger 2023-10-13 07:09:19,792 - LinearScheduler | warmup_fraction: '0.1' 2023-10-13 07:09:19,792 ---------------------------------------------------------------------------------------------------- 2023-10-13 07:09:19,792 Final evaluation on model from best epoch (best-model.pt) 2023-10-13 07:09:19,792 - metric: "('micro avg', 'f1-score')" 2023-10-13 07:09:19,792 ---------------------------------------------------------------------------------------------------- 2023-10-13 07:09:19,792 Computation: 2023-10-13 07:09:19,793 - compute on device: cuda:0 2023-10-13 07:09:19,793 - embedding storage: none 2023-10-13 07:09:19,793 ---------------------------------------------------------------------------------------------------- 2023-10-13 07:09:19,793 Model training base path: "hmbench-letemps/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-1" 2023-10-13 07:09:19,793 ---------------------------------------------------------------------------------------------------- 2023-10-13 07:09:19,793 ---------------------------------------------------------------------------------------------------- 2023-10-13 07:09:19,793 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-13 07:11:03,127 epoch 1 - iter 361/3617 - loss 2.54672490 - time (sec): 103.33 - samples/sec: 372.78 - lr: 0.000016 - momentum: 0.000000 2023-10-13 07:12:44,471 epoch 1 - iter 722/3617 - loss 2.14481700 - time (sec): 204.68 - samples/sec: 372.30 - lr: 0.000032 - momentum: 0.000000 2023-10-13 07:14:23,374 epoch 1 - iter 1083/3617 - loss 1.68863182 - time (sec): 303.58 - samples/sec: 372.84 - lr: 0.000048 - momentum: 0.000000 2023-10-13 07:16:07,432 epoch 1 - iter 1444/3617 - loss 1.33089974 - time (sec): 407.64 - samples/sec: 373.61 - lr: 0.000064 - momentum: 0.000000 2023-10-13 07:17:50,485 epoch 1 - iter 1805/3617 - loss 1.10545660 - time (sec): 510.69 - samples/sec: 371.79 - lr: 0.000080 - momentum: 0.000000 2023-10-13 07:19:32,500 epoch 1 - iter 2166/3617 - loss 0.95603286 - time (sec): 612.71 - samples/sec: 370.31 - lr: 0.000096 - momentum: 0.000000 2023-10-13 07:21:14,877 epoch 1 - iter 2527/3617 - loss 0.84785682 - time (sec): 715.08 - samples/sec: 368.44 - lr: 0.000112 - momentum: 0.000000 2023-10-13 07:22:59,295 epoch 1 - iter 2888/3617 - loss 0.75882497 - time (sec): 819.50 - samples/sec: 368.18 - lr: 0.000128 - momentum: 0.000000 2023-10-13 07:24:39,030 epoch 1 - iter 3249/3617 - loss 0.69053264 - time (sec): 919.23 - samples/sec: 369.76 - lr: 0.000144 - momentum: 0.000000 2023-10-13 07:26:17,579 epoch 1 - iter 3610/3617 - loss 0.62979015 - time (sec): 1017.78 - samples/sec: 372.72 - lr: 0.000160 - momentum: 0.000000 2023-10-13 07:26:19,360 ---------------------------------------------------------------------------------------------------- 2023-10-13 07:26:19,361 EPOCH 1 done: loss 0.6290 - lr: 0.000160 2023-10-13 07:26:58,014 DEV : loss 0.12302889674901962 - f1-score (micro avg) 0.5399 2023-10-13 07:26:58,075 saving best model 2023-10-13 07:26:58,931 ---------------------------------------------------------------------------------------------------- 2023-10-13 07:28:41,476 epoch 2 - iter 361/3617 - loss 0.09184357 - time (sec): 102.54 - samples/sec: 379.17 - lr: 0.000158 - momentum: 0.000000 2023-10-13 07:30:19,808 epoch 2 - iter 722/3617 - loss 0.09724116 - time (sec): 200.88 - samples/sec: 385.81 - lr: 0.000156 - momentum: 0.000000 2023-10-13 07:31:58,762 epoch 2 - iter 1083/3617 - loss 0.09717697 - time (sec): 299.83 - samples/sec: 385.45 - lr: 0.000155 - momentum: 0.000000 2023-10-13 07:33:40,743 epoch 2 - iter 1444/3617 - loss 0.09928124 - time (sec): 401.81 - samples/sec: 379.33 - lr: 0.000153 - momentum: 0.000000 2023-10-13 07:35:19,332 epoch 2 - iter 1805/3617 - loss 0.09824636 - time (sec): 500.40 - samples/sec: 378.12 - lr: 0.000151 - momentum: 0.000000 2023-10-13 07:36:59,984 epoch 2 - iter 2166/3617 - loss 0.09948203 - time (sec): 601.05 - samples/sec: 378.45 - lr: 0.000149 - momentum: 0.000000 2023-10-13 07:38:39,721 epoch 2 - iter 2527/3617 - loss 0.09758575 - time (sec): 700.79 - samples/sec: 378.47 - lr: 0.000148 - momentum: 0.000000 2023-10-13 07:40:20,216 epoch 2 - iter 2888/3617 - loss 0.09631999 - time (sec): 801.28 - samples/sec: 378.70 - lr: 0.000146 - momentum: 0.000000 2023-10-13 07:42:00,600 epoch 2 - iter 3249/3617 - loss 0.09404083 - time (sec): 901.67 - samples/sec: 379.47 - lr: 0.000144 - momentum: 0.000000 2023-10-13 07:43:37,048 epoch 2 - iter 3610/3617 - loss 0.09413353 - time (sec): 998.11 - samples/sec: 379.70 - lr: 0.000142 - momentum: 0.000000 2023-10-13 07:43:38,928 ---------------------------------------------------------------------------------------------------- 2023-10-13 07:43:38,928 EPOCH 2 done: loss 0.0940 - lr: 0.000142 2023-10-13 07:44:17,984 DEV : loss 0.11269791424274445 - f1-score (micro avg) 0.6141 2023-10-13 07:44:18,043 saving best model 2023-10-13 07:44:20,630 ---------------------------------------------------------------------------------------------------- 2023-10-13 07:45:59,384 epoch 3 - iter 361/3617 - loss 0.05810261 - time (sec): 98.75 - samples/sec: 387.16 - lr: 0.000140 - momentum: 0.000000 2023-10-13 07:47:42,320 epoch 3 - iter 722/3617 - loss 0.06040886 - time (sec): 201.69 - samples/sec: 380.77 - lr: 0.000139 - momentum: 0.000000 2023-10-13 07:49:20,365 epoch 3 - iter 1083/3617 - loss 0.06342105 - time (sec): 299.73 - samples/sec: 379.17 - lr: 0.000137 - momentum: 0.000000 2023-10-13 07:50:58,968 epoch 3 - iter 1444/3617 - loss 0.06293381 - time (sec): 398.33 - samples/sec: 378.53 - lr: 0.000135 - momentum: 0.000000 2023-10-13 07:52:35,727 epoch 3 - iter 1805/3617 - loss 0.06498947 - time (sec): 495.09 - samples/sec: 381.36 - lr: 0.000133 - momentum: 0.000000 2023-10-13 07:54:12,518 epoch 3 - iter 2166/3617 - loss 0.06553587 - time (sec): 591.88 - samples/sec: 383.44 - lr: 0.000132 - momentum: 0.000000 2023-10-13 07:55:49,481 epoch 3 - iter 2527/3617 - loss 0.06591706 - time (sec): 688.85 - samples/sec: 385.80 - lr: 0.000130 - momentum: 0.000000 2023-10-13 07:57:29,045 epoch 3 - iter 2888/3617 - loss 0.06544978 - time (sec): 788.41 - samples/sec: 384.44 - lr: 0.000128 - momentum: 0.000000 2023-10-13 07:59:12,481 epoch 3 - iter 3249/3617 - loss 0.06554704 - time (sec): 891.85 - samples/sec: 382.70 - lr: 0.000126 - momentum: 0.000000 2023-10-13 08:00:54,592 epoch 3 - iter 3610/3617 - loss 0.06594083 - time (sec): 993.96 - samples/sec: 381.46 - lr: 0.000124 - momentum: 0.000000 2023-10-13 08:00:56,337 ---------------------------------------------------------------------------------------------------- 2023-10-13 08:00:56,337 EPOCH 3 done: loss 0.0659 - lr: 0.000124 2023-10-13 08:01:36,047 DEV : loss 0.17262127995491028 - f1-score (micro avg) 0.6187 2023-10-13 08:01:36,109 saving best model 2023-10-13 08:01:38,726 ---------------------------------------------------------------------------------------------------- 2023-10-13 08:03:17,667 epoch 4 - iter 361/3617 - loss 0.04571932 - time (sec): 98.94 - samples/sec: 372.96 - lr: 0.000123 - momentum: 0.000000 2023-10-13 08:04:59,641 epoch 4 - iter 722/3617 - loss 0.04768170 - time (sec): 200.91 - samples/sec: 380.44 - lr: 0.000121 - momentum: 0.000000 2023-10-13 08:06:45,305 epoch 4 - iter 1083/3617 - loss 0.04760406 - time (sec): 306.57 - samples/sec: 371.61 - lr: 0.000119 - momentum: 0.000000 2023-10-13 08:08:29,628 epoch 4 - iter 1444/3617 - loss 0.04660555 - time (sec): 410.90 - samples/sec: 367.32 - lr: 0.000117 - momentum: 0.000000 2023-10-13 08:10:11,625 epoch 4 - iter 1805/3617 - loss 0.04686239 - time (sec): 512.90 - samples/sec: 366.00 - lr: 0.000116 - momentum: 0.000000 2023-10-13 08:11:51,533 epoch 4 - iter 2166/3617 - loss 0.04882203 - time (sec): 612.80 - samples/sec: 369.00 - lr: 0.000114 - momentum: 0.000000 2023-10-13 08:13:32,397 epoch 4 - iter 2527/3617 - loss 0.04981878 - time (sec): 713.67 - samples/sec: 369.31 - lr: 0.000112 - momentum: 0.000000 2023-10-13 08:15:15,443 epoch 4 - iter 2888/3617 - loss 0.04947922 - time (sec): 816.71 - samples/sec: 369.59 - lr: 0.000110 - momentum: 0.000000 2023-10-13 08:16:59,452 epoch 4 - iter 3249/3617 - loss 0.04870494 - time (sec): 920.72 - samples/sec: 370.52 - lr: 0.000108 - momentum: 0.000000 2023-10-13 08:18:43,800 epoch 4 - iter 3610/3617 - loss 0.04931029 - time (sec): 1025.07 - samples/sec: 369.92 - lr: 0.000107 - momentum: 0.000000 2023-10-13 08:18:45,673 ---------------------------------------------------------------------------------------------------- 2023-10-13 08:18:45,673 EPOCH 4 done: loss 0.0493 - lr: 0.000107 2023-10-13 08:19:25,861 DEV : loss 0.21019545197486877 - f1-score (micro avg) 0.6423 2023-10-13 08:19:25,919 saving best model 2023-10-13 08:19:28,524 ---------------------------------------------------------------------------------------------------- 2023-10-13 08:21:10,444 epoch 5 - iter 361/3617 - loss 0.03375571 - time (sec): 101.91 - samples/sec: 377.94 - lr: 0.000105 - momentum: 0.000000 2023-10-13 08:22:55,141 epoch 5 - iter 722/3617 - loss 0.03056571 - time (sec): 206.61 - samples/sec: 372.05 - lr: 0.000103 - momentum: 0.000000 2023-10-13 08:24:42,670 epoch 5 - iter 1083/3617 - loss 0.03134028 - time (sec): 314.14 - samples/sec: 361.24 - lr: 0.000101 - momentum: 0.000000 2023-10-13 08:26:25,328 epoch 5 - iter 1444/3617 - loss 0.03200139 - time (sec): 416.80 - samples/sec: 360.14 - lr: 0.000100 - momentum: 0.000000 2023-10-13 08:28:05,490 epoch 5 - iter 1805/3617 - loss 0.03332061 - time (sec): 516.96 - samples/sec: 363.22 - lr: 0.000098 - momentum: 0.000000 2023-10-13 08:29:47,849 epoch 5 - iter 2166/3617 - loss 0.03328642 - time (sec): 619.32 - samples/sec: 364.98 - lr: 0.000096 - momentum: 0.000000 2023-10-13 08:31:31,345 epoch 5 - iter 2527/3617 - loss 0.03444733 - time (sec): 722.81 - samples/sec: 364.33 - lr: 0.000094 - momentum: 0.000000 2023-10-13 08:33:13,022 epoch 5 - iter 2888/3617 - loss 0.03503322 - time (sec): 824.49 - samples/sec: 364.33 - lr: 0.000092 - momentum: 0.000000 2023-10-13 08:34:58,383 epoch 5 - iter 3249/3617 - loss 0.03543944 - time (sec): 929.85 - samples/sec: 366.65 - lr: 0.000091 - momentum: 0.000000 2023-10-13 08:36:40,423 epoch 5 - iter 3610/3617 - loss 0.03600676 - time (sec): 1031.89 - samples/sec: 367.67 - lr: 0.000089 - momentum: 0.000000 2023-10-13 08:36:42,044 ---------------------------------------------------------------------------------------------------- 2023-10-13 08:36:42,045 EPOCH 5 done: loss 0.0360 - lr: 0.000089 2023-10-13 08:37:22,806 DEV : loss 0.2428499311208725 - f1-score (micro avg) 0.6353 2023-10-13 08:37:22,864 ---------------------------------------------------------------------------------------------------- 2023-10-13 08:39:03,567 epoch 6 - iter 361/3617 - loss 0.02475435 - time (sec): 100.70 - samples/sec: 374.38 - lr: 0.000087 - momentum: 0.000000 2023-10-13 08:40:46,107 epoch 6 - iter 722/3617 - loss 0.02240235 - time (sec): 203.24 - samples/sec: 373.84 - lr: 0.000085 - momentum: 0.000000 2023-10-13 08:42:26,885 epoch 6 - iter 1083/3617 - loss 0.02242435 - time (sec): 304.02 - samples/sec: 373.28 - lr: 0.000084 - momentum: 0.000000 2023-10-13 08:44:07,737 epoch 6 - iter 1444/3617 - loss 0.02215591 - time (sec): 404.87 - samples/sec: 374.59 - lr: 0.000082 - momentum: 0.000000 2023-10-13 08:45:50,341 epoch 6 - iter 1805/3617 - loss 0.02397631 - time (sec): 507.47 - samples/sec: 374.23 - lr: 0.000080 - momentum: 0.000000 2023-10-13 08:47:34,746 epoch 6 - iter 2166/3617 - loss 0.02419269 - time (sec): 611.88 - samples/sec: 372.36 - lr: 0.000078 - momentum: 0.000000 2023-10-13 08:49:19,195 epoch 6 - iter 2527/3617 - loss 0.02490330 - time (sec): 716.33 - samples/sec: 370.90 - lr: 0.000076 - momentum: 0.000000 2023-10-13 08:51:05,354 epoch 6 - iter 2888/3617 - loss 0.02459970 - time (sec): 822.49 - samples/sec: 367.63 - lr: 0.000075 - momentum: 0.000000 2023-10-13 08:52:48,651 epoch 6 - iter 3249/3617 - loss 0.02441037 - time (sec): 925.78 - samples/sec: 367.52 - lr: 0.000073 - momentum: 0.000000 2023-10-13 08:54:31,614 epoch 6 - iter 3610/3617 - loss 0.02410353 - time (sec): 1028.75 - samples/sec: 368.80 - lr: 0.000071 - momentum: 0.000000 2023-10-13 08:54:33,301 ---------------------------------------------------------------------------------------------------- 2023-10-13 08:54:33,302 EPOCH 6 done: loss 0.0241 - lr: 0.000071 2023-10-13 08:55:13,801 DEV : loss 0.30240556597709656 - f1-score (micro avg) 0.6237 2023-10-13 08:55:13,880 ---------------------------------------------------------------------------------------------------- 2023-10-13 08:56:59,065 epoch 7 - iter 361/3617 - loss 0.01445564 - time (sec): 105.18 - samples/sec: 353.40 - lr: 0.000069 - momentum: 0.000000 2023-10-13 08:58:43,231 epoch 7 - iter 722/3617 - loss 0.01467506 - time (sec): 209.35 - samples/sec: 362.74 - lr: 0.000068 - momentum: 0.000000 2023-10-13 09:00:28,382 epoch 7 - iter 1083/3617 - loss 0.01578205 - time (sec): 314.50 - samples/sec: 361.82 - lr: 0.000066 - momentum: 0.000000 2023-10-13 09:02:13,449 epoch 7 - iter 1444/3617 - loss 0.01703422 - time (sec): 419.57 - samples/sec: 360.77 - lr: 0.000064 - momentum: 0.000000 2023-10-13 09:03:56,020 epoch 7 - iter 1805/3617 - loss 0.01763882 - time (sec): 522.14 - samples/sec: 362.88 - lr: 0.000062 - momentum: 0.000000 2023-10-13 09:05:39,265 epoch 7 - iter 2166/3617 - loss 0.01724767 - time (sec): 625.38 - samples/sec: 362.10 - lr: 0.000060 - momentum: 0.000000 2023-10-13 09:07:26,199 epoch 7 - iter 2527/3617 - loss 0.01745495 - time (sec): 732.32 - samples/sec: 361.47 - lr: 0.000059 - momentum: 0.000000 2023-10-13 09:09:12,338 epoch 7 - iter 2888/3617 - loss 0.01801878 - time (sec): 838.46 - samples/sec: 360.57 - lr: 0.000057 - momentum: 0.000000 2023-10-13 09:10:58,550 epoch 7 - iter 3249/3617 - loss 0.01838554 - time (sec): 944.67 - samples/sec: 360.61 - lr: 0.000055 - momentum: 0.000000 2023-10-13 09:12:45,236 epoch 7 - iter 3610/3617 - loss 0.01799741 - time (sec): 1051.35 - samples/sec: 360.80 - lr: 0.000053 - momentum: 0.000000 2023-10-13 09:12:46,988 ---------------------------------------------------------------------------------------------------- 2023-10-13 09:12:46,988 EPOCH 7 done: loss 0.0180 - lr: 0.000053 2023-10-13 09:13:29,166 DEV : loss 0.32392749190330505 - f1-score (micro avg) 0.6152 2023-10-13 09:13:29,227 ---------------------------------------------------------------------------------------------------- 2023-10-13 09:15:11,286 epoch 8 - iter 361/3617 - loss 0.00831627 - time (sec): 102.06 - samples/sec: 368.88 - lr: 0.000052 - momentum: 0.000000 2023-10-13 09:16:52,695 epoch 8 - iter 722/3617 - loss 0.00914200 - time (sec): 203.46 - samples/sec: 373.90 - lr: 0.000050 - momentum: 0.000000 2023-10-13 09:18:32,396 epoch 8 - iter 1083/3617 - loss 0.00912467 - time (sec): 303.17 - samples/sec: 376.89 - lr: 0.000048 - momentum: 0.000000 2023-10-13 09:20:14,373 epoch 8 - iter 1444/3617 - loss 0.00993485 - time (sec): 405.14 - samples/sec: 378.39 - lr: 0.000046 - momentum: 0.000000 2023-10-13 09:21:55,557 epoch 8 - iter 1805/3617 - loss 0.01122757 - time (sec): 506.33 - samples/sec: 379.34 - lr: 0.000044 - momentum: 0.000000 2023-10-13 09:23:35,609 epoch 8 - iter 2166/3617 - loss 0.01152886 - time (sec): 606.38 - samples/sec: 376.14 - lr: 0.000043 - momentum: 0.000000 2023-10-13 09:25:14,099 epoch 8 - iter 2527/3617 - loss 0.01150134 - time (sec): 704.87 - samples/sec: 376.75 - lr: 0.000041 - momentum: 0.000000 2023-10-13 09:26:54,414 epoch 8 - iter 2888/3617 - loss 0.01131531 - time (sec): 805.18 - samples/sec: 377.48 - lr: 0.000039 - momentum: 0.000000 2023-10-13 09:28:34,512 epoch 8 - iter 3249/3617 - loss 0.01111063 - time (sec): 905.28 - samples/sec: 377.77 - lr: 0.000037 - momentum: 0.000000 2023-10-13 09:30:14,589 epoch 8 - iter 3610/3617 - loss 0.01137858 - time (sec): 1005.36 - samples/sec: 377.30 - lr: 0.000036 - momentum: 0.000000 2023-10-13 09:30:16,327 ---------------------------------------------------------------------------------------------------- 2023-10-13 09:30:16,328 EPOCH 8 done: loss 0.0114 - lr: 0.000036 2023-10-13 09:30:58,519 DEV : loss 0.3561553359031677 - f1-score (micro avg) 0.636 2023-10-13 09:30:58,579 ---------------------------------------------------------------------------------------------------- 2023-10-13 09:32:40,941 epoch 9 - iter 361/3617 - loss 0.00723439 - time (sec): 102.36 - samples/sec: 364.00 - lr: 0.000034 - momentum: 0.000000 2023-10-13 09:34:22,522 epoch 9 - iter 722/3617 - loss 0.00867757 - time (sec): 203.94 - samples/sec: 367.26 - lr: 0.000032 - momentum: 0.000000 2023-10-13 09:36:01,585 epoch 9 - iter 1083/3617 - loss 0.00720214 - time (sec): 303.00 - samples/sec: 371.87 - lr: 0.000030 - momentum: 0.000000 2023-10-13 09:37:46,141 epoch 9 - iter 1444/3617 - loss 0.00884133 - time (sec): 407.56 - samples/sec: 374.63 - lr: 0.000028 - momentum: 0.000000 2023-10-13 09:39:27,863 epoch 9 - iter 1805/3617 - loss 0.00912211 - time (sec): 509.28 - samples/sec: 373.69 - lr: 0.000027 - momentum: 0.000000 2023-10-13 09:41:06,474 epoch 9 - iter 2166/3617 - loss 0.00902114 - time (sec): 607.89 - samples/sec: 375.21 - lr: 0.000025 - momentum: 0.000000 2023-10-13 09:42:44,579 epoch 9 - iter 2527/3617 - loss 0.00876123 - time (sec): 706.00 - samples/sec: 375.58 - lr: 0.000023 - momentum: 0.000000 2023-10-13 09:44:26,686 epoch 9 - iter 2888/3617 - loss 0.00893095 - time (sec): 808.10 - samples/sec: 375.68 - lr: 0.000021 - momentum: 0.000000 2023-10-13 09:46:08,982 epoch 9 - iter 3249/3617 - loss 0.00856866 - time (sec): 910.40 - samples/sec: 375.95 - lr: 0.000020 - momentum: 0.000000 2023-10-13 09:47:53,550 epoch 9 - iter 3610/3617 - loss 0.00830378 - time (sec): 1014.97 - samples/sec: 373.82 - lr: 0.000018 - momentum: 0.000000 2023-10-13 09:47:55,260 ---------------------------------------------------------------------------------------------------- 2023-10-13 09:47:55,260 EPOCH 9 done: loss 0.0083 - lr: 0.000018 2023-10-13 09:48:36,184 DEV : loss 0.38502201437950134 - f1-score (micro avg) 0.6408 2023-10-13 09:48:36,253 ---------------------------------------------------------------------------------------------------- 2023-10-13 09:50:20,200 epoch 10 - iter 361/3617 - loss 0.00438003 - time (sec): 103.94 - samples/sec: 366.40 - lr: 0.000016 - momentum: 0.000000 2023-10-13 09:52:04,672 epoch 10 - iter 722/3617 - loss 0.00381567 - time (sec): 208.42 - samples/sec: 361.18 - lr: 0.000014 - momentum: 0.000000 2023-10-13 09:53:50,255 epoch 10 - iter 1083/3617 - loss 0.00444144 - time (sec): 314.00 - samples/sec: 360.25 - lr: 0.000012 - momentum: 0.000000 2023-10-13 09:55:35,823 epoch 10 - iter 1444/3617 - loss 0.00443475 - time (sec): 419.57 - samples/sec: 358.31 - lr: 0.000011 - momentum: 0.000000 2023-10-13 09:57:23,187 epoch 10 - iter 1805/3617 - loss 0.00403389 - time (sec): 526.93 - samples/sec: 357.77 - lr: 0.000009 - momentum: 0.000000 2023-10-13 09:59:06,440 epoch 10 - iter 2166/3617 - loss 0.00409837 - time (sec): 630.18 - samples/sec: 359.97 - lr: 0.000007 - momentum: 0.000000 2023-10-13 10:00:50,582 epoch 10 - iter 2527/3617 - loss 0.00436368 - time (sec): 734.33 - samples/sec: 360.70 - lr: 0.000005 - momentum: 0.000000 2023-10-13 10:02:33,802 epoch 10 - iter 2888/3617 - loss 0.00446848 - time (sec): 837.55 - samples/sec: 361.82 - lr: 0.000004 - momentum: 0.000000 2023-10-13 10:04:19,545 epoch 10 - iter 3249/3617 - loss 0.00433966 - time (sec): 943.29 - samples/sec: 361.99 - lr: 0.000002 - momentum: 0.000000 2023-10-13 10:06:04,901 epoch 10 - iter 3610/3617 - loss 0.00416053 - time (sec): 1048.65 - samples/sec: 361.81 - lr: 0.000000 - momentum: 0.000000 2023-10-13 10:06:06,582 ---------------------------------------------------------------------------------------------------- 2023-10-13 10:06:06,582 EPOCH 10 done: loss 0.0042 - lr: 0.000000 2023-10-13 10:06:48,722 DEV : loss 0.3937250077724457 - f1-score (micro avg) 0.6409 2023-10-13 10:06:49,678 ---------------------------------------------------------------------------------------------------- 2023-10-13 10:06:49,680 Loading model from best epoch ... 2023-10-13 10:06:53,669 SequenceTagger predicts: Dictionary with 13 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org 2023-10-13 10:07:56,286 Results: - F-score (micro) 0.6366 - F-score (macro) 0.4589 - Accuracy 0.4766 By class: precision recall f1-score support loc 0.6312 0.8020 0.7064 591 pers 0.5514 0.6611 0.6013 357 org 0.1081 0.0506 0.0690 79 micro avg 0.5872 0.6952 0.6366 1027 macro avg 0.4302 0.5046 0.4589 1027 weighted avg 0.5632 0.6952 0.6208 1027 2023-10-13 10:07:56,287 ----------------------------------------------------------------------------------------------------