2023-10-12 17:47:39,484 ---------------------------------------------------------------------------------------------------- 2023-10-12 17:47:39,486 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() )" 2023-10-12 17:47:39,486 ---------------------------------------------------------------------------------------------------- 2023-10-12 17:47:39,486 MultiCorpus: 5777 train + 722 dev + 723 test sentences - NER_ICDAR_EUROPEANA Corpus: 5777 train + 722 dev + 723 test sentences - /root/.flair/datasets/ner_icdar_europeana/nl 2023-10-12 17:47:39,487 ---------------------------------------------------------------------------------------------------- 2023-10-12 17:47:39,487 Train: 5777 sentences 2023-10-12 17:47:39,487 (train_with_dev=False, train_with_test=False) 2023-10-12 17:47:39,487 ---------------------------------------------------------------------------------------------------- 2023-10-12 17:47:39,487 Training Params: 2023-10-12 17:47:39,487 - learning_rate: "0.00015" 2023-10-12 17:47:39,487 - mini_batch_size: "8" 2023-10-12 17:47:39,487 - max_epochs: "10" 2023-10-12 17:47:39,487 - shuffle: "True" 2023-10-12 17:47:39,487 ---------------------------------------------------------------------------------------------------- 2023-10-12 17:47:39,487 Plugins: 2023-10-12 17:47:39,487 - TensorboardLogger 2023-10-12 17:47:39,487 - LinearScheduler | warmup_fraction: '0.1' 2023-10-12 17:47:39,487 ---------------------------------------------------------------------------------------------------- 2023-10-12 17:47:39,488 Final evaluation on model from best epoch (best-model.pt) 2023-10-12 17:47:39,488 - metric: "('micro avg', 'f1-score')" 2023-10-12 17:47:39,488 ---------------------------------------------------------------------------------------------------- 2023-10-12 17:47:39,488 Computation: 2023-10-12 17:47:39,488 - compute on device: cuda:0 2023-10-12 17:47:39,488 - embedding storage: none 2023-10-12 17:47:39,488 ---------------------------------------------------------------------------------------------------- 2023-10-12 17:47:39,488 Model training base path: "hmbench-icdar/nl-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-5" 2023-10-12 17:47:39,488 ---------------------------------------------------------------------------------------------------- 2023-10-12 17:47:39,488 ---------------------------------------------------------------------------------------------------- 2023-10-12 17:47:39,488 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-12 17:48:18,952 epoch 1 - iter 72/723 - loss 2.54426887 - time (sec): 39.46 - samples/sec: 457.63 - lr: 0.000015 - momentum: 0.000000 2023-10-12 17:48:58,124 epoch 1 - iter 144/723 - loss 2.48027812 - time (sec): 78.63 - samples/sec: 459.08 - lr: 0.000030 - momentum: 0.000000 2023-10-12 17:49:36,310 epoch 1 - iter 216/723 - loss 2.32386957 - time (sec): 116.82 - samples/sec: 448.35 - lr: 0.000045 - momentum: 0.000000 2023-10-12 17:50:14,644 epoch 1 - iter 288/723 - loss 2.11630297 - time (sec): 155.15 - samples/sec: 451.57 - lr: 0.000060 - momentum: 0.000000 2023-10-12 17:50:53,387 epoch 1 - iter 360/723 - loss 1.89885572 - time (sec): 193.90 - samples/sec: 452.96 - lr: 0.000074 - momentum: 0.000000 2023-10-12 17:51:31,686 epoch 1 - iter 432/723 - loss 1.68800148 - time (sec): 232.20 - samples/sec: 450.91 - lr: 0.000089 - momentum: 0.000000 2023-10-12 17:52:11,717 epoch 1 - iter 504/723 - loss 1.48803245 - time (sec): 272.23 - samples/sec: 449.72 - lr: 0.000104 - momentum: 0.000000 2023-10-12 17:52:51,394 epoch 1 - iter 576/723 - loss 1.34139358 - time (sec): 311.90 - samples/sec: 446.33 - lr: 0.000119 - momentum: 0.000000 2023-10-12 17:53:31,923 epoch 1 - iter 648/723 - loss 1.21633560 - time (sec): 352.43 - samples/sec: 445.27 - lr: 0.000134 - momentum: 0.000000 2023-10-12 17:54:12,600 epoch 1 - iter 720/723 - loss 1.10698770 - time (sec): 393.11 - samples/sec: 446.34 - lr: 0.000149 - momentum: 0.000000 2023-10-12 17:54:13,966 ---------------------------------------------------------------------------------------------------- 2023-10-12 17:54:13,967 EPOCH 1 done: loss 1.1029 - lr: 0.000149 2023-10-12 17:54:34,128 DEV : loss 0.23375172913074493 - f1-score (micro avg) 0.0 2023-10-12 17:54:34,158 ---------------------------------------------------------------------------------------------------- 2023-10-12 17:55:12,547 epoch 2 - iter 72/723 - loss 0.16731668 - time (sec): 38.39 - samples/sec: 454.44 - lr: 0.000148 - momentum: 0.000000 2023-10-12 17:55:51,979 epoch 2 - iter 144/723 - loss 0.15482755 - time (sec): 77.82 - samples/sec: 455.74 - lr: 0.000147 - momentum: 0.000000 2023-10-12 17:56:29,875 epoch 2 - iter 216/723 - loss 0.15195012 - time (sec): 115.71 - samples/sec: 449.25 - lr: 0.000145 - momentum: 0.000000 2023-10-12 17:57:09,651 epoch 2 - iter 288/723 - loss 0.14875477 - time (sec): 155.49 - samples/sec: 448.78 - lr: 0.000143 - momentum: 0.000000 2023-10-12 17:57:48,585 epoch 2 - iter 360/723 - loss 0.14422756 - time (sec): 194.42 - samples/sec: 448.71 - lr: 0.000142 - momentum: 0.000000 2023-10-12 17:58:27,126 epoch 2 - iter 432/723 - loss 0.14054029 - time (sec): 232.97 - samples/sec: 449.87 - lr: 0.000140 - momentum: 0.000000 2023-10-12 17:59:05,553 epoch 2 - iter 504/723 - loss 0.14009191 - time (sec): 271.39 - samples/sec: 449.88 - lr: 0.000138 - momentum: 0.000000 2023-10-12 17:59:44,287 epoch 2 - iter 576/723 - loss 0.13710606 - time (sec): 310.13 - samples/sec: 451.15 - lr: 0.000137 - momentum: 0.000000 2023-10-12 18:00:23,907 epoch 2 - iter 648/723 - loss 0.13415472 - time (sec): 349.75 - samples/sec: 452.30 - lr: 0.000135 - momentum: 0.000000 2023-10-12 18:01:03,365 epoch 2 - iter 720/723 - loss 0.13018803 - time (sec): 389.20 - samples/sec: 451.19 - lr: 0.000133 - momentum: 0.000000 2023-10-12 18:01:04,560 ---------------------------------------------------------------------------------------------------- 2023-10-12 18:01:04,560 EPOCH 2 done: loss 0.1300 - lr: 0.000133 2023-10-12 18:01:25,127 DEV : loss 0.11187870055437088 - f1-score (micro avg) 0.7651 2023-10-12 18:01:25,159 saving best model 2023-10-12 18:01:26,051 ---------------------------------------------------------------------------------------------------- 2023-10-12 18:02:05,492 epoch 3 - iter 72/723 - loss 0.08592561 - time (sec): 39.44 - samples/sec: 453.97 - lr: 0.000132 - momentum: 0.000000 2023-10-12 18:02:44,351 epoch 3 - iter 144/723 - loss 0.08297721 - time (sec): 78.30 - samples/sec: 459.33 - lr: 0.000130 - momentum: 0.000000 2023-10-12 18:03:22,483 epoch 3 - iter 216/723 - loss 0.08326043 - time (sec): 116.43 - samples/sec: 459.71 - lr: 0.000128 - momentum: 0.000000 2023-10-12 18:04:00,111 epoch 3 - iter 288/723 - loss 0.08082376 - time (sec): 154.06 - samples/sec: 461.41 - lr: 0.000127 - momentum: 0.000000 2023-10-12 18:04:38,650 epoch 3 - iter 360/723 - loss 0.08052324 - time (sec): 192.60 - samples/sec: 462.31 - lr: 0.000125 - momentum: 0.000000 2023-10-12 18:05:17,244 epoch 3 - iter 432/723 - loss 0.08037304 - time (sec): 231.19 - samples/sec: 466.15 - lr: 0.000123 - momentum: 0.000000 2023-10-12 18:05:56,024 epoch 3 - iter 504/723 - loss 0.07975153 - time (sec): 269.97 - samples/sec: 463.63 - lr: 0.000122 - momentum: 0.000000 2023-10-12 18:06:34,552 epoch 3 - iter 576/723 - loss 0.07988037 - time (sec): 308.50 - samples/sec: 460.86 - lr: 0.000120 - momentum: 0.000000 2023-10-12 18:07:12,746 epoch 3 - iter 648/723 - loss 0.07895931 - time (sec): 346.69 - samples/sec: 457.82 - lr: 0.000118 - momentum: 0.000000 2023-10-12 18:07:50,890 epoch 3 - iter 720/723 - loss 0.07794918 - time (sec): 384.84 - samples/sec: 456.08 - lr: 0.000117 - momentum: 0.000000 2023-10-12 18:07:52,195 ---------------------------------------------------------------------------------------------------- 2023-10-12 18:07:52,195 EPOCH 3 done: loss 0.0781 - lr: 0.000117 2023-10-12 18:08:14,169 DEV : loss 0.0785035490989685 - f1-score (micro avg) 0.8324 2023-10-12 18:08:14,200 saving best model 2023-10-12 18:08:16,769 ---------------------------------------------------------------------------------------------------- 2023-10-12 18:08:58,498 epoch 4 - iter 72/723 - loss 0.05759562 - time (sec): 41.72 - samples/sec: 429.84 - lr: 0.000115 - momentum: 0.000000 2023-10-12 18:09:39,681 epoch 4 - iter 144/723 - loss 0.05741867 - time (sec): 82.91 - samples/sec: 419.13 - lr: 0.000113 - momentum: 0.000000 2023-10-12 18:10:20,375 epoch 4 - iter 216/723 - loss 0.05415096 - time (sec): 123.60 - samples/sec: 424.18 - lr: 0.000112 - momentum: 0.000000 2023-10-12 18:11:00,926 epoch 4 - iter 288/723 - loss 0.05445295 - time (sec): 164.15 - samples/sec: 435.34 - lr: 0.000110 - momentum: 0.000000 2023-10-12 18:11:39,771 epoch 4 - iter 360/723 - loss 0.05261383 - time (sec): 203.00 - samples/sec: 439.52 - lr: 0.000108 - momentum: 0.000000 2023-10-12 18:12:18,198 epoch 4 - iter 432/723 - loss 0.05135342 - time (sec): 241.42 - samples/sec: 439.86 - lr: 0.000107 - momentum: 0.000000 2023-10-12 18:12:57,140 epoch 4 - iter 504/723 - loss 0.05062735 - time (sec): 280.37 - samples/sec: 439.37 - lr: 0.000105 - momentum: 0.000000 2023-10-12 18:13:36,398 epoch 4 - iter 576/723 - loss 0.04994851 - time (sec): 319.62 - samples/sec: 442.45 - lr: 0.000103 - momentum: 0.000000 2023-10-12 18:14:15,270 epoch 4 - iter 648/723 - loss 0.05141664 - time (sec): 358.50 - samples/sec: 443.12 - lr: 0.000102 - momentum: 0.000000 2023-10-12 18:14:53,270 epoch 4 - iter 720/723 - loss 0.05052236 - time (sec): 396.50 - samples/sec: 443.46 - lr: 0.000100 - momentum: 0.000000 2023-10-12 18:14:54,369 ---------------------------------------------------------------------------------------------------- 2023-10-12 18:14:54,369 EPOCH 4 done: loss 0.0506 - lr: 0.000100 2023-10-12 18:15:15,356 DEV : loss 0.07858217507600784 - f1-score (micro avg) 0.8518 2023-10-12 18:15:15,388 saving best model 2023-10-12 18:15:17,971 ---------------------------------------------------------------------------------------------------- 2023-10-12 18:15:59,176 epoch 5 - iter 72/723 - loss 0.03874655 - time (sec): 41.20 - samples/sec: 458.15 - lr: 0.000098 - momentum: 0.000000 2023-10-12 18:16:39,098 epoch 5 - iter 144/723 - loss 0.03347685 - time (sec): 81.12 - samples/sec: 447.35 - lr: 0.000097 - momentum: 0.000000 2023-10-12 18:17:16,677 epoch 5 - iter 216/723 - loss 0.03299884 - time (sec): 118.70 - samples/sec: 436.84 - lr: 0.000095 - momentum: 0.000000 2023-10-12 18:17:54,188 epoch 5 - iter 288/723 - loss 0.03250430 - time (sec): 156.21 - samples/sec: 435.65 - lr: 0.000093 - momentum: 0.000000 2023-10-12 18:18:33,812 epoch 5 - iter 360/723 - loss 0.03475104 - time (sec): 195.84 - samples/sec: 442.33 - lr: 0.000092 - momentum: 0.000000 2023-10-12 18:19:12,566 epoch 5 - iter 432/723 - loss 0.03403611 - time (sec): 234.59 - samples/sec: 443.37 - lr: 0.000090 - momentum: 0.000000 2023-10-12 18:19:52,453 epoch 5 - iter 504/723 - loss 0.03407641 - time (sec): 274.48 - samples/sec: 446.40 - lr: 0.000088 - momentum: 0.000000 2023-10-12 18:20:31,367 epoch 5 - iter 576/723 - loss 0.03399255 - time (sec): 313.39 - samples/sec: 447.69 - lr: 0.000087 - momentum: 0.000000 2023-10-12 18:21:10,192 epoch 5 - iter 648/723 - loss 0.03424309 - time (sec): 352.22 - samples/sec: 448.30 - lr: 0.000085 - momentum: 0.000000 2023-10-12 18:21:48,915 epoch 5 - iter 720/723 - loss 0.03511790 - time (sec): 390.94 - samples/sec: 448.58 - lr: 0.000083 - momentum: 0.000000 2023-10-12 18:21:50,307 ---------------------------------------------------------------------------------------------------- 2023-10-12 18:21:50,308 EPOCH 5 done: loss 0.0353 - lr: 0.000083 2023-10-12 18:22:12,540 DEV : loss 0.0762009546160698 - f1-score (micro avg) 0.8552 2023-10-12 18:22:12,572 saving best model 2023-10-12 18:22:15,194 ---------------------------------------------------------------------------------------------------- 2023-10-12 18:22:54,690 epoch 6 - iter 72/723 - loss 0.02292044 - time (sec): 39.49 - samples/sec: 435.70 - lr: 0.000082 - momentum: 0.000000 2023-10-12 18:23:33,953 epoch 6 - iter 144/723 - loss 0.02437413 - time (sec): 78.75 - samples/sec: 438.41 - lr: 0.000080 - momentum: 0.000000 2023-10-12 18:24:13,910 epoch 6 - iter 216/723 - loss 0.02690559 - time (sec): 118.71 - samples/sec: 441.44 - lr: 0.000078 - momentum: 0.000000 2023-10-12 18:24:54,865 epoch 6 - iter 288/723 - loss 0.02616528 - time (sec): 159.67 - samples/sec: 440.99 - lr: 0.000077 - momentum: 0.000000 2023-10-12 18:25:34,380 epoch 6 - iter 360/723 - loss 0.02600927 - time (sec): 199.18 - samples/sec: 442.46 - lr: 0.000075 - momentum: 0.000000 2023-10-12 18:26:14,694 epoch 6 - iter 432/723 - loss 0.02342678 - time (sec): 239.50 - samples/sec: 444.95 - lr: 0.000073 - momentum: 0.000000 2023-10-12 18:26:55,637 epoch 6 - iter 504/723 - loss 0.02526975 - time (sec): 280.44 - samples/sec: 443.58 - lr: 0.000072 - momentum: 0.000000 2023-10-12 18:27:35,655 epoch 6 - iter 576/723 - loss 0.02479113 - time (sec): 320.46 - samples/sec: 439.85 - lr: 0.000070 - momentum: 0.000000 2023-10-12 18:28:15,107 epoch 6 - iter 648/723 - loss 0.02441460 - time (sec): 359.91 - samples/sec: 438.05 - lr: 0.000068 - momentum: 0.000000 2023-10-12 18:28:56,224 epoch 6 - iter 720/723 - loss 0.02520439 - time (sec): 401.03 - samples/sec: 438.06 - lr: 0.000067 - momentum: 0.000000 2023-10-12 18:28:57,424 ---------------------------------------------------------------------------------------------------- 2023-10-12 18:28:57,425 EPOCH 6 done: loss 0.0252 - lr: 0.000067 2023-10-12 18:29:18,375 DEV : loss 0.08811366558074951 - f1-score (micro avg) 0.8673 2023-10-12 18:29:18,405 saving best model 2023-10-12 18:29:20,978 ---------------------------------------------------------------------------------------------------- 2023-10-12 18:30:00,020 epoch 7 - iter 72/723 - loss 0.02375220 - time (sec): 39.04 - samples/sec: 451.86 - lr: 0.000065 - momentum: 0.000000 2023-10-12 18:30:39,390 epoch 7 - iter 144/723 - loss 0.02317751 - time (sec): 78.41 - samples/sec: 453.74 - lr: 0.000063 - momentum: 0.000000 2023-10-12 18:31:17,750 epoch 7 - iter 216/723 - loss 0.02191170 - time (sec): 116.77 - samples/sec: 446.14 - lr: 0.000062 - momentum: 0.000000 2023-10-12 18:31:56,342 epoch 7 - iter 288/723 - loss 0.02214116 - time (sec): 155.36 - samples/sec: 441.87 - lr: 0.000060 - momentum: 0.000000 2023-10-12 18:32:35,819 epoch 7 - iter 360/723 - loss 0.02354976 - time (sec): 194.84 - samples/sec: 445.48 - lr: 0.000058 - momentum: 0.000000 2023-10-12 18:33:15,315 epoch 7 - iter 432/723 - loss 0.02157574 - time (sec): 234.33 - samples/sec: 444.74 - lr: 0.000057 - momentum: 0.000000 2023-10-12 18:33:54,392 epoch 7 - iter 504/723 - loss 0.02134366 - time (sec): 273.41 - samples/sec: 445.97 - lr: 0.000055 - momentum: 0.000000 2023-10-12 18:34:33,498 epoch 7 - iter 576/723 - loss 0.02118785 - time (sec): 312.51 - samples/sec: 444.59 - lr: 0.000053 - momentum: 0.000000 2023-10-12 18:35:12,523 epoch 7 - iter 648/723 - loss 0.02048925 - time (sec): 351.54 - samples/sec: 445.23 - lr: 0.000052 - momentum: 0.000000 2023-10-12 18:35:52,164 epoch 7 - iter 720/723 - loss 0.02077841 - time (sec): 391.18 - samples/sec: 448.61 - lr: 0.000050 - momentum: 0.000000 2023-10-12 18:35:53,549 ---------------------------------------------------------------------------------------------------- 2023-10-12 18:35:53,550 EPOCH 7 done: loss 0.0207 - lr: 0.000050 2023-10-12 18:36:15,769 DEV : loss 0.09639902412891388 - f1-score (micro avg) 0.8597 2023-10-12 18:36:15,806 ---------------------------------------------------------------------------------------------------- 2023-10-12 18:36:57,015 epoch 8 - iter 72/723 - loss 0.01788093 - time (sec): 41.21 - samples/sec: 450.42 - lr: 0.000048 - momentum: 0.000000 2023-10-12 18:37:36,612 epoch 8 - iter 144/723 - loss 0.01763665 - time (sec): 80.80 - samples/sec: 444.40 - lr: 0.000047 - momentum: 0.000000 2023-10-12 18:38:16,755 epoch 8 - iter 216/723 - loss 0.01670632 - time (sec): 120.95 - samples/sec: 436.75 - lr: 0.000045 - momentum: 0.000000 2023-10-12 18:38:57,813 epoch 8 - iter 288/723 - loss 0.01563682 - time (sec): 162.00 - samples/sec: 444.00 - lr: 0.000043 - momentum: 0.000000 2023-10-12 18:39:37,387 epoch 8 - iter 360/723 - loss 0.01520499 - time (sec): 201.58 - samples/sec: 443.01 - lr: 0.000042 - momentum: 0.000000 2023-10-12 18:40:16,403 epoch 8 - iter 432/723 - loss 0.01554144 - time (sec): 240.59 - samples/sec: 439.58 - lr: 0.000040 - momentum: 0.000000 2023-10-12 18:40:55,597 epoch 8 - iter 504/723 - loss 0.01575127 - time (sec): 279.79 - samples/sec: 439.79 - lr: 0.000038 - momentum: 0.000000 2023-10-12 18:41:34,231 epoch 8 - iter 576/723 - loss 0.01547145 - time (sec): 318.42 - samples/sec: 438.17 - lr: 0.000037 - momentum: 0.000000 2023-10-12 18:42:14,603 epoch 8 - iter 648/723 - loss 0.01719346 - time (sec): 358.79 - samples/sec: 440.09 - lr: 0.000035 - momentum: 0.000000 2023-10-12 18:42:54,038 epoch 8 - iter 720/723 - loss 0.01672576 - time (sec): 398.23 - samples/sec: 441.26 - lr: 0.000033 - momentum: 0.000000 2023-10-12 18:42:55,161 ---------------------------------------------------------------------------------------------------- 2023-10-12 18:42:55,161 EPOCH 8 done: loss 0.0167 - lr: 0.000033 2023-10-12 18:43:18,087 DEV : loss 0.10770395398139954 - f1-score (micro avg) 0.855 2023-10-12 18:43:18,120 ---------------------------------------------------------------------------------------------------- 2023-10-12 18:43:58,068 epoch 9 - iter 72/723 - loss 0.00592187 - time (sec): 39.95 - samples/sec: 460.70 - lr: 0.000032 - momentum: 0.000000 2023-10-12 18:44:37,416 epoch 9 - iter 144/723 - loss 0.01701284 - time (sec): 79.29 - samples/sec: 469.13 - lr: 0.000030 - momentum: 0.000000 2023-10-12 18:45:16,392 epoch 9 - iter 216/723 - loss 0.01704092 - time (sec): 118.27 - samples/sec: 465.48 - lr: 0.000028 - momentum: 0.000000 2023-10-12 18:45:54,208 epoch 9 - iter 288/723 - loss 0.01622595 - time (sec): 156.09 - samples/sec: 455.77 - lr: 0.000027 - momentum: 0.000000 2023-10-12 18:46:32,450 epoch 9 - iter 360/723 - loss 0.01556455 - time (sec): 194.33 - samples/sec: 447.69 - lr: 0.000025 - momentum: 0.000000 2023-10-12 18:47:11,398 epoch 9 - iter 432/723 - loss 0.01494819 - time (sec): 233.28 - samples/sec: 449.75 - lr: 0.000023 - momentum: 0.000000 2023-10-12 18:47:50,162 epoch 9 - iter 504/723 - loss 0.01504915 - time (sec): 272.04 - samples/sec: 450.43 - lr: 0.000022 - momentum: 0.000000 2023-10-12 18:48:29,909 epoch 9 - iter 576/723 - loss 0.01463468 - time (sec): 311.79 - samples/sec: 453.71 - lr: 0.000020 - momentum: 0.000000 2023-10-12 18:49:08,405 epoch 9 - iter 648/723 - loss 0.01388108 - time (sec): 350.28 - samples/sec: 452.53 - lr: 0.000018 - momentum: 0.000000 2023-10-12 18:49:50,026 epoch 9 - iter 720/723 - loss 0.01368723 - time (sec): 391.90 - samples/sec: 448.25 - lr: 0.000017 - momentum: 0.000000 2023-10-12 18:49:51,219 ---------------------------------------------------------------------------------------------------- 2023-10-12 18:49:51,220 EPOCH 9 done: loss 0.0136 - lr: 0.000017 2023-10-12 18:50:13,188 DEV : loss 0.11267418414354324 - f1-score (micro avg) 0.8553 2023-10-12 18:50:13,220 ---------------------------------------------------------------------------------------------------- 2023-10-12 18:50:52,654 epoch 10 - iter 72/723 - loss 0.00826221 - time (sec): 39.43 - samples/sec: 456.87 - lr: 0.000015 - momentum: 0.000000 2023-10-12 18:51:29,673 epoch 10 - iter 144/723 - loss 0.00855540 - time (sec): 76.45 - samples/sec: 441.22 - lr: 0.000013 - momentum: 0.000000 2023-10-12 18:52:07,710 epoch 10 - iter 216/723 - loss 0.00947687 - time (sec): 114.49 - samples/sec: 442.76 - lr: 0.000012 - momentum: 0.000000 2023-10-12 18:52:47,939 epoch 10 - iter 288/723 - loss 0.01177807 - time (sec): 154.71 - samples/sec: 446.72 - lr: 0.000010 - momentum: 0.000000 2023-10-12 18:53:27,103 epoch 10 - iter 360/723 - loss 0.01109256 - time (sec): 193.88 - samples/sec: 445.15 - lr: 0.000008 - momentum: 0.000000 2023-10-12 18:54:07,126 epoch 10 - iter 432/723 - loss 0.01009327 - time (sec): 233.90 - samples/sec: 447.66 - lr: 0.000007 - momentum: 0.000000 2023-10-12 18:54:48,429 epoch 10 - iter 504/723 - loss 0.01055010 - time (sec): 275.20 - samples/sec: 447.94 - lr: 0.000005 - momentum: 0.000000 2023-10-12 18:55:28,401 epoch 10 - iter 576/723 - loss 0.01002322 - time (sec): 315.18 - samples/sec: 444.09 - lr: 0.000003 - momentum: 0.000000 2023-10-12 18:56:07,902 epoch 10 - iter 648/723 - loss 0.01138945 - time (sec): 354.68 - samples/sec: 444.59 - lr: 0.000002 - momentum: 0.000000 2023-10-12 18:56:48,588 epoch 10 - iter 720/723 - loss 0.01160600 - time (sec): 395.36 - samples/sec: 444.50 - lr: 0.000000 - momentum: 0.000000 2023-10-12 18:56:49,706 ---------------------------------------------------------------------------------------------------- 2023-10-12 18:56:49,707 EPOCH 10 done: loss 0.0116 - lr: 0.000000 2023-10-12 18:57:10,758 DEV : loss 0.11588139832019806 - f1-score (micro avg) 0.8573 2023-10-12 18:57:11,656 ---------------------------------------------------------------------------------------------------- 2023-10-12 18:57:11,658 Loading model from best epoch ... 2023-10-12 18:57:15,259 SequenceTagger predicts: Dictionary with 13 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-ORG, B-ORG, E-ORG, I-ORG 2023-10-12 18:57:35,839 Results: - F-score (micro) 0.858 - F-score (macro) 0.7492 - Accuracy 0.7592 By class: precision recall f1-score support PER 0.8376 0.8880 0.8620 482 LOC 0.9273 0.8908 0.9087 458 ORG 0.5082 0.4493 0.4769 69 micro avg 0.8567 0.8593 0.8580 1009 macro avg 0.7577 0.7427 0.7492 1009 weighted avg 0.8558 0.8593 0.8569 1009 2023-10-12 18:57:35,840 ----------------------------------------------------------------------------------------------------