2023-10-10 01:25:46,775 ---------------------------------------------------------------------------------------------------- 2023-10-10 01:25:46,778 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() )" 2023-10-10 01:25:46,778 ---------------------------------------------------------------------------------------------------- 2023-10-10 01:25:46,778 MultiCorpus: 20847 train + 1123 dev + 3350 test sentences - NER_HIPE_2022 Corpus: 20847 train + 1123 dev + 3350 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/de/with_doc_seperator 2023-10-10 01:25:46,778 ---------------------------------------------------------------------------------------------------- 2023-10-10 01:25:46,778 Train: 20847 sentences 2023-10-10 01:25:46,778 (train_with_dev=False, train_with_test=False) 2023-10-10 01:25:46,778 ---------------------------------------------------------------------------------------------------- 2023-10-10 01:25:46,779 Training Params: 2023-10-10 01:25:46,779 - learning_rate: "0.00015" 2023-10-10 01:25:46,779 - mini_batch_size: "4" 2023-10-10 01:25:46,779 - max_epochs: "10" 2023-10-10 01:25:46,779 - shuffle: "True" 2023-10-10 01:25:46,779 ---------------------------------------------------------------------------------------------------- 2023-10-10 01:25:46,779 Plugins: 2023-10-10 01:25:46,779 - TensorboardLogger 2023-10-10 01:25:46,779 - LinearScheduler | warmup_fraction: '0.1' 2023-10-10 01:25:46,779 ---------------------------------------------------------------------------------------------------- 2023-10-10 01:25:46,779 Final evaluation on model from best epoch (best-model.pt) 2023-10-10 01:25:46,779 - metric: "('micro avg', 'f1-score')" 2023-10-10 01:25:46,779 ---------------------------------------------------------------------------------------------------- 2023-10-10 01:25:46,780 Computation: 2023-10-10 01:25:46,780 - compute on device: cuda:0 2023-10-10 01:25:46,780 - embedding storage: none 2023-10-10 01:25:46,780 ---------------------------------------------------------------------------------------------------- 2023-10-10 01:25:46,780 Model training base path: "hmbench-newseye/de-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-1" 2023-10-10 01:25:46,780 ---------------------------------------------------------------------------------------------------- 2023-10-10 01:25:46,780 ---------------------------------------------------------------------------------------------------- 2023-10-10 01:25:46,780 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-10 01:28:19,901 epoch 1 - iter 521/5212 - loss 2.78721271 - time (sec): 153.12 - samples/sec: 258.19 - lr: 0.000015 - momentum: 0.000000 2023-10-10 01:30:47,683 epoch 1 - iter 1042/5212 - loss 2.40394453 - time (sec): 300.90 - samples/sec: 246.08 - lr: 0.000030 - momentum: 0.000000 2023-10-10 01:33:20,332 epoch 1 - iter 1563/5212 - loss 1.88420497 - time (sec): 453.55 - samples/sec: 243.15 - lr: 0.000045 - momentum: 0.000000 2023-10-10 01:35:50,068 epoch 1 - iter 2084/5212 - loss 1.56398813 - time (sec): 603.29 - samples/sec: 238.94 - lr: 0.000060 - momentum: 0.000000 2023-10-10 01:38:28,783 epoch 1 - iter 2605/5212 - loss 1.32922150 - time (sec): 762.00 - samples/sec: 239.39 - lr: 0.000075 - momentum: 0.000000 2023-10-10 01:41:01,466 epoch 1 - iter 3126/5212 - loss 1.16575348 - time (sec): 914.68 - samples/sec: 240.69 - lr: 0.000090 - momentum: 0.000000 2023-10-10 01:43:35,466 epoch 1 - iter 3647/5212 - loss 1.04771990 - time (sec): 1068.68 - samples/sec: 239.12 - lr: 0.000105 - momentum: 0.000000 2023-10-10 01:46:05,151 epoch 1 - iter 4168/5212 - loss 0.94981352 - time (sec): 1218.37 - samples/sec: 240.08 - lr: 0.000120 - momentum: 0.000000 2023-10-10 01:48:42,725 epoch 1 - iter 4689/5212 - loss 0.86675606 - time (sec): 1375.94 - samples/sec: 240.43 - lr: 0.000135 - momentum: 0.000000 2023-10-10 01:51:13,055 epoch 1 - iter 5210/5212 - loss 0.80285395 - time (sec): 1526.27 - samples/sec: 240.62 - lr: 0.000150 - momentum: 0.000000 2023-10-10 01:51:13,587 ---------------------------------------------------------------------------------------------------- 2023-10-10 01:51:13,588 EPOCH 1 done: loss 0.8026 - lr: 0.000150 2023-10-10 01:51:50,130 DEV : loss 0.13000161945819855 - f1-score (micro avg) 0.2937 2023-10-10 01:51:50,181 saving best model 2023-10-10 01:51:51,139 ---------------------------------------------------------------------------------------------------- 2023-10-10 01:54:26,458 epoch 2 - iter 521/5212 - loss 0.17727772 - time (sec): 155.32 - samples/sec: 256.21 - lr: 0.000148 - momentum: 0.000000 2023-10-10 01:57:04,108 epoch 2 - iter 1042/5212 - loss 0.18552352 - time (sec): 312.97 - samples/sec: 251.81 - lr: 0.000147 - momentum: 0.000000 2023-10-10 01:59:40,241 epoch 2 - iter 1563/5212 - loss 0.17656777 - time (sec): 469.10 - samples/sec: 247.34 - lr: 0.000145 - momentum: 0.000000 2023-10-10 02:02:10,466 epoch 2 - iter 2084/5212 - loss 0.17347903 - time (sec): 619.32 - samples/sec: 243.06 - lr: 0.000143 - momentum: 0.000000 2023-10-10 02:04:44,612 epoch 2 - iter 2605/5212 - loss 0.17247510 - time (sec): 773.47 - samples/sec: 241.26 - lr: 0.000142 - momentum: 0.000000 2023-10-10 02:07:16,051 epoch 2 - iter 3126/5212 - loss 0.17019344 - time (sec): 924.91 - samples/sec: 240.53 - lr: 0.000140 - momentum: 0.000000 2023-10-10 02:09:48,370 epoch 2 - iter 3647/5212 - loss 0.16742648 - time (sec): 1077.23 - samples/sec: 240.03 - lr: 0.000138 - momentum: 0.000000 2023-10-10 02:12:23,147 epoch 2 - iter 4168/5212 - loss 0.16216771 - time (sec): 1232.01 - samples/sec: 240.31 - lr: 0.000137 - momentum: 0.000000 2023-10-10 02:15:01,539 epoch 2 - iter 4689/5212 - loss 0.15863022 - time (sec): 1390.40 - samples/sec: 239.62 - lr: 0.000135 - momentum: 0.000000 2023-10-10 02:17:30,737 epoch 2 - iter 5210/5212 - loss 0.15545238 - time (sec): 1539.60 - samples/sec: 238.55 - lr: 0.000133 - momentum: 0.000000 2023-10-10 02:17:31,246 ---------------------------------------------------------------------------------------------------- 2023-10-10 02:17:31,247 EPOCH 2 done: loss 0.1554 - lr: 0.000133 2023-10-10 02:18:13,857 DEV : loss 0.1568019837141037 - f1-score (micro avg) 0.3643 2023-10-10 02:18:13,912 saving best model 2023-10-10 02:18:16,652 ---------------------------------------------------------------------------------------------------- 2023-10-10 02:20:47,238 epoch 3 - iter 521/5212 - loss 0.09227579 - time (sec): 150.58 - samples/sec: 242.13 - lr: 0.000132 - momentum: 0.000000 2023-10-10 02:23:18,633 epoch 3 - iter 1042/5212 - loss 0.10115948 - time (sec): 301.98 - samples/sec: 236.47 - lr: 0.000130 - momentum: 0.000000 2023-10-10 02:25:54,573 epoch 3 - iter 1563/5212 - loss 0.10224575 - time (sec): 457.92 - samples/sec: 241.87 - lr: 0.000128 - momentum: 0.000000 2023-10-10 02:28:26,479 epoch 3 - iter 2084/5212 - loss 0.10442613 - time (sec): 609.82 - samples/sec: 236.77 - lr: 0.000127 - momentum: 0.000000 2023-10-10 02:30:55,098 epoch 3 - iter 2605/5212 - loss 0.10550793 - time (sec): 758.44 - samples/sec: 234.00 - lr: 0.000125 - momentum: 0.000000 2023-10-10 02:33:31,679 epoch 3 - iter 3126/5212 - loss 0.10488651 - time (sec): 915.02 - samples/sec: 237.84 - lr: 0.000123 - momentum: 0.000000 2023-10-10 02:36:05,961 epoch 3 - iter 3647/5212 - loss 0.10697333 - time (sec): 1069.30 - samples/sec: 239.46 - lr: 0.000122 - momentum: 0.000000 2023-10-10 02:38:37,130 epoch 3 - iter 4168/5212 - loss 0.10672325 - time (sec): 1220.47 - samples/sec: 240.95 - lr: 0.000120 - momentum: 0.000000 2023-10-10 02:41:12,922 epoch 3 - iter 4689/5212 - loss 0.10596268 - time (sec): 1376.27 - samples/sec: 241.08 - lr: 0.000118 - momentum: 0.000000 2023-10-10 02:43:44,745 epoch 3 - iter 5210/5212 - loss 0.10482979 - time (sec): 1528.09 - samples/sec: 240.42 - lr: 0.000117 - momentum: 0.000000 2023-10-10 02:43:45,188 ---------------------------------------------------------------------------------------------------- 2023-10-10 02:43:45,188 EPOCH 3 done: loss 0.1048 - lr: 0.000117 2023-10-10 02:44:26,403 DEV : loss 0.26355189085006714 - f1-score (micro avg) 0.3544 2023-10-10 02:44:26,460 ---------------------------------------------------------------------------------------------------- 2023-10-10 02:47:01,262 epoch 4 - iter 521/5212 - loss 0.06438853 - time (sec): 154.80 - samples/sec: 238.52 - lr: 0.000115 - momentum: 0.000000 2023-10-10 02:49:38,517 epoch 4 - iter 1042/5212 - loss 0.06254566 - time (sec): 312.05 - samples/sec: 231.26 - lr: 0.000113 - momentum: 0.000000 2023-10-10 02:52:14,492 epoch 4 - iter 1563/5212 - loss 0.06423763 - time (sec): 468.03 - samples/sec: 230.64 - lr: 0.000112 - momentum: 0.000000 2023-10-10 02:54:44,857 epoch 4 - iter 2084/5212 - loss 0.06688261 - time (sec): 618.39 - samples/sec: 232.20 - lr: 0.000110 - momentum: 0.000000 2023-10-10 02:57:20,257 epoch 4 - iter 2605/5212 - loss 0.07115747 - time (sec): 773.79 - samples/sec: 234.30 - lr: 0.000108 - momentum: 0.000000 2023-10-10 02:59:54,457 epoch 4 - iter 3126/5212 - loss 0.06965197 - time (sec): 927.99 - samples/sec: 238.97 - lr: 0.000107 - momentum: 0.000000 2023-10-10 03:02:33,869 epoch 4 - iter 3647/5212 - loss 0.06885936 - time (sec): 1087.41 - samples/sec: 235.99 - lr: 0.000105 - momentum: 0.000000 2023-10-10 03:05:03,715 epoch 4 - iter 4168/5212 - loss 0.07019996 - time (sec): 1237.25 - samples/sec: 236.43 - lr: 0.000103 - momentum: 0.000000 2023-10-10 03:07:37,272 epoch 4 - iter 4689/5212 - loss 0.07063262 - time (sec): 1390.81 - samples/sec: 237.63 - lr: 0.000102 - momentum: 0.000000 2023-10-10 03:10:08,879 epoch 4 - iter 5210/5212 - loss 0.07185046 - time (sec): 1542.42 - samples/sec: 238.20 - lr: 0.000100 - momentum: 0.000000 2023-10-10 03:10:09,314 ---------------------------------------------------------------------------------------------------- 2023-10-10 03:10:09,314 EPOCH 4 done: loss 0.0719 - lr: 0.000100 2023-10-10 03:10:57,812 DEV : loss 0.328808069229126 - f1-score (micro avg) 0.3675 2023-10-10 03:10:57,879 saving best model 2023-10-10 03:11:09,277 ---------------------------------------------------------------------------------------------------- 2023-10-10 03:13:41,434 epoch 5 - iter 521/5212 - loss 0.05178094 - time (sec): 152.15 - samples/sec: 227.02 - lr: 0.000098 - momentum: 0.000000 2023-10-10 03:16:13,044 epoch 5 - iter 1042/5212 - loss 0.05498718 - time (sec): 303.76 - samples/sec: 234.98 - lr: 0.000097 - momentum: 0.000000 2023-10-10 03:18:45,690 epoch 5 - iter 1563/5212 - loss 0.05426323 - time (sec): 456.41 - samples/sec: 241.01 - lr: 0.000095 - momentum: 0.000000 2023-10-10 03:21:23,172 epoch 5 - iter 2084/5212 - loss 0.05256449 - time (sec): 613.89 - samples/sec: 242.02 - lr: 0.000093 - momentum: 0.000000 2023-10-10 03:23:56,778 epoch 5 - iter 2605/5212 - loss 0.05299146 - time (sec): 767.50 - samples/sec: 238.88 - lr: 0.000092 - momentum: 0.000000 2023-10-10 03:26:30,782 epoch 5 - iter 3126/5212 - loss 0.05454252 - time (sec): 921.50 - samples/sec: 239.64 - lr: 0.000090 - momentum: 0.000000 2023-10-10 03:29:07,368 epoch 5 - iter 3647/5212 - loss 0.05456496 - time (sec): 1078.09 - samples/sec: 240.64 - lr: 0.000088 - momentum: 0.000000 2023-10-10 03:31:43,290 epoch 5 - iter 4168/5212 - loss 0.05333736 - time (sec): 1234.01 - samples/sec: 239.80 - lr: 0.000087 - momentum: 0.000000 2023-10-10 03:34:15,429 epoch 5 - iter 4689/5212 - loss 0.05223842 - time (sec): 1386.15 - samples/sec: 238.43 - lr: 0.000085 - momentum: 0.000000 2023-10-10 03:36:49,643 epoch 5 - iter 5210/5212 - loss 0.05325230 - time (sec): 1540.36 - samples/sec: 238.48 - lr: 0.000083 - momentum: 0.000000 2023-10-10 03:36:50,135 ---------------------------------------------------------------------------------------------------- 2023-10-10 03:36:50,135 EPOCH 5 done: loss 0.0532 - lr: 0.000083 2023-10-10 03:37:32,491 DEV : loss 0.3098498284816742 - f1-score (micro avg) 0.3954 2023-10-10 03:37:32,548 saving best model 2023-10-10 03:37:35,444 ---------------------------------------------------------------------------------------------------- 2023-10-10 03:40:16,955 epoch 6 - iter 521/5212 - loss 0.03216296 - time (sec): 161.51 - samples/sec: 227.66 - lr: 0.000082 - momentum: 0.000000 2023-10-10 03:42:49,627 epoch 6 - iter 1042/5212 - loss 0.03763248 - time (sec): 314.18 - samples/sec: 225.89 - lr: 0.000080 - momentum: 0.000000 2023-10-10 03:45:24,399 epoch 6 - iter 1563/5212 - loss 0.03476631 - time (sec): 468.95 - samples/sec: 234.51 - lr: 0.000078 - momentum: 0.000000 2023-10-10 03:48:01,947 epoch 6 - iter 2084/5212 - loss 0.03495034 - time (sec): 626.50 - samples/sec: 232.55 - lr: 0.000077 - momentum: 0.000000 2023-10-10 03:50:33,387 epoch 6 - iter 2605/5212 - loss 0.03541576 - time (sec): 777.94 - samples/sec: 234.85 - lr: 0.000075 - momentum: 0.000000 2023-10-10 03:53:03,173 epoch 6 - iter 3126/5212 - loss 0.03439095 - time (sec): 927.72 - samples/sec: 233.04 - lr: 0.000073 - momentum: 0.000000 2023-10-10 03:55:38,848 epoch 6 - iter 3647/5212 - loss 0.03483107 - time (sec): 1083.40 - samples/sec: 233.74 - lr: 0.000072 - momentum: 0.000000 2023-10-10 03:58:12,884 epoch 6 - iter 4168/5212 - loss 0.03532550 - time (sec): 1237.44 - samples/sec: 235.10 - lr: 0.000070 - momentum: 0.000000 2023-10-10 04:00:56,870 epoch 6 - iter 4689/5212 - loss 0.03575938 - time (sec): 1401.42 - samples/sec: 235.54 - lr: 0.000068 - momentum: 0.000000 2023-10-10 04:03:29,722 epoch 6 - iter 5210/5212 - loss 0.03599569 - time (sec): 1554.27 - samples/sec: 236.34 - lr: 0.000067 - momentum: 0.000000 2023-10-10 04:03:30,188 ---------------------------------------------------------------------------------------------------- 2023-10-10 04:03:30,188 EPOCH 6 done: loss 0.0360 - lr: 0.000067 2023-10-10 04:04:12,941 DEV : loss 0.3672010004520416 - f1-score (micro avg) 0.3823 2023-10-10 04:04:13,004 ---------------------------------------------------------------------------------------------------- 2023-10-10 04:06:49,841 epoch 7 - iter 521/5212 - loss 0.01901470 - time (sec): 156.83 - samples/sec: 245.25 - lr: 0.000065 - momentum: 0.000000 2023-10-10 04:09:26,255 epoch 7 - iter 1042/5212 - loss 0.02194304 - time (sec): 313.25 - samples/sec: 244.81 - lr: 0.000063 - momentum: 0.000000 2023-10-10 04:12:00,065 epoch 7 - iter 1563/5212 - loss 0.02292694 - time (sec): 467.06 - samples/sec: 239.09 - lr: 0.000062 - momentum: 0.000000 2023-10-10 04:14:34,985 epoch 7 - iter 2084/5212 - loss 0.02286490 - time (sec): 621.98 - samples/sec: 239.67 - lr: 0.000060 - momentum: 0.000000 2023-10-10 04:17:07,483 epoch 7 - iter 2605/5212 - loss 0.02158969 - time (sec): 774.48 - samples/sec: 241.38 - lr: 0.000058 - momentum: 0.000000 2023-10-10 04:19:42,460 epoch 7 - iter 3126/5212 - loss 0.02437178 - time (sec): 929.45 - samples/sec: 241.11 - lr: 0.000057 - momentum: 0.000000 2023-10-10 04:22:14,067 epoch 7 - iter 3647/5212 - loss 0.02420860 - time (sec): 1081.06 - samples/sec: 240.27 - lr: 0.000055 - momentum: 0.000000 2023-10-10 04:24:44,274 epoch 7 - iter 4168/5212 - loss 0.02497424 - time (sec): 1231.27 - samples/sec: 240.58 - lr: 0.000053 - momentum: 0.000000 2023-10-10 04:27:18,230 epoch 7 - iter 4689/5212 - loss 0.02500752 - time (sec): 1385.22 - samples/sec: 240.06 - lr: 0.000052 - momentum: 0.000000 2023-10-10 04:29:46,445 epoch 7 - iter 5210/5212 - loss 0.02487125 - time (sec): 1533.44 - samples/sec: 239.57 - lr: 0.000050 - momentum: 0.000000 2023-10-10 04:29:46,901 ---------------------------------------------------------------------------------------------------- 2023-10-10 04:29:46,902 EPOCH 7 done: loss 0.0249 - lr: 0.000050 2023-10-10 04:30:31,180 DEV : loss 0.43393340706825256 - f1-score (micro avg) 0.3819 2023-10-10 04:30:31,244 ---------------------------------------------------------------------------------------------------- 2023-10-10 04:33:08,702 epoch 8 - iter 521/5212 - loss 0.02118548 - time (sec): 157.46 - samples/sec: 229.44 - lr: 0.000048 - momentum: 0.000000 2023-10-10 04:35:41,639 epoch 8 - iter 1042/5212 - loss 0.01953101 - time (sec): 310.39 - samples/sec: 235.18 - lr: 0.000047 - momentum: 0.000000 2023-10-10 04:38:18,307 epoch 8 - iter 1563/5212 - loss 0.01941108 - time (sec): 467.06 - samples/sec: 237.33 - lr: 0.000045 - momentum: 0.000000 2023-10-10 04:40:52,233 epoch 8 - iter 2084/5212 - loss 0.01883043 - time (sec): 620.99 - samples/sec: 236.34 - lr: 0.000043 - momentum: 0.000000 2023-10-10 04:43:26,171 epoch 8 - iter 2605/5212 - loss 0.01862250 - time (sec): 774.92 - samples/sec: 236.00 - lr: 0.000042 - momentum: 0.000000 2023-10-10 04:46:02,246 epoch 8 - iter 3126/5212 - loss 0.01937239 - time (sec): 931.00 - samples/sec: 236.09 - lr: 0.000040 - momentum: 0.000000 2023-10-10 04:48:33,038 epoch 8 - iter 3647/5212 - loss 0.01908805 - time (sec): 1081.79 - samples/sec: 235.79 - lr: 0.000038 - momentum: 0.000000 2023-10-10 04:51:12,633 epoch 8 - iter 4168/5212 - loss 0.01889720 - time (sec): 1241.39 - samples/sec: 236.80 - lr: 0.000037 - momentum: 0.000000 2023-10-10 04:53:47,905 epoch 8 - iter 4689/5212 - loss 0.01843401 - time (sec): 1396.66 - samples/sec: 236.95 - lr: 0.000035 - momentum: 0.000000 2023-10-10 04:56:21,396 epoch 8 - iter 5210/5212 - loss 0.01859840 - time (sec): 1550.15 - samples/sec: 236.93 - lr: 0.000033 - momentum: 0.000000 2023-10-10 04:56:21,934 ---------------------------------------------------------------------------------------------------- 2023-10-10 04:56:21,934 EPOCH 8 done: loss 0.0186 - lr: 0.000033 2023-10-10 04:57:03,167 DEV : loss 0.46257734298706055 - f1-score (micro avg) 0.3734 2023-10-10 04:57:03,238 ---------------------------------------------------------------------------------------------------- 2023-10-10 04:59:37,592 epoch 9 - iter 521/5212 - loss 0.01100351 - time (sec): 154.35 - samples/sec: 244.77 - lr: 0.000032 - momentum: 0.000000 2023-10-10 05:02:12,911 epoch 9 - iter 1042/5212 - loss 0.01142614 - time (sec): 309.67 - samples/sec: 244.21 - lr: 0.000030 - momentum: 0.000000 2023-10-10 05:04:45,707 epoch 9 - iter 1563/5212 - loss 0.01200825 - time (sec): 462.47 - samples/sec: 238.44 - lr: 0.000028 - momentum: 0.000000 2023-10-10 05:07:25,374 epoch 9 - iter 2084/5212 - loss 0.01167246 - time (sec): 622.13 - samples/sec: 236.70 - lr: 0.000027 - momentum: 0.000000 2023-10-10 05:10:00,448 epoch 9 - iter 2605/5212 - loss 0.01211194 - time (sec): 777.21 - samples/sec: 236.18 - lr: 0.000025 - momentum: 0.000000 2023-10-10 05:12:35,180 epoch 9 - iter 3126/5212 - loss 0.01227832 - time (sec): 931.94 - samples/sec: 236.70 - lr: 0.000023 - momentum: 0.000000 2023-10-10 05:15:13,447 epoch 9 - iter 3647/5212 - loss 0.01232401 - time (sec): 1090.21 - samples/sec: 235.04 - lr: 0.000022 - momentum: 0.000000 2023-10-10 05:17:46,800 epoch 9 - iter 4168/5212 - loss 0.01205408 - time (sec): 1243.56 - samples/sec: 234.71 - lr: 0.000020 - momentum: 0.000000 2023-10-10 05:20:19,024 epoch 9 - iter 4689/5212 - loss 0.01137052 - time (sec): 1395.78 - samples/sec: 235.39 - lr: 0.000018 - momentum: 0.000000 2023-10-10 05:23:03,373 epoch 9 - iter 5210/5212 - loss 0.01116459 - time (sec): 1560.13 - samples/sec: 235.41 - lr: 0.000017 - momentum: 0.000000 2023-10-10 05:23:03,935 ---------------------------------------------------------------------------------------------------- 2023-10-10 05:23:03,935 EPOCH 9 done: loss 0.0112 - lr: 0.000017 2023-10-10 05:23:45,741 DEV : loss 0.49680188298225403 - f1-score (micro avg) 0.386 2023-10-10 05:23:45,810 ---------------------------------------------------------------------------------------------------- 2023-10-10 05:26:19,459 epoch 10 - iter 521/5212 - loss 0.00618823 - time (sec): 153.65 - samples/sec: 239.16 - lr: 0.000015 - momentum: 0.000000 2023-10-10 05:28:50,815 epoch 10 - iter 1042/5212 - loss 0.00778641 - time (sec): 305.00 - samples/sec: 236.39 - lr: 0.000013 - momentum: 0.000000 2023-10-10 05:31:22,417 epoch 10 - iter 1563/5212 - loss 0.00880970 - time (sec): 456.60 - samples/sec: 231.53 - lr: 0.000012 - momentum: 0.000000 2023-10-10 05:34:00,744 epoch 10 - iter 2084/5212 - loss 0.00838074 - time (sec): 614.93 - samples/sec: 234.01 - lr: 0.000010 - momentum: 0.000000 2023-10-10 05:36:40,115 epoch 10 - iter 2605/5212 - loss 0.00876966 - time (sec): 774.30 - samples/sec: 238.88 - lr: 0.000008 - momentum: 0.000000 2023-10-10 05:39:13,455 epoch 10 - iter 3126/5212 - loss 0.00882349 - time (sec): 927.64 - samples/sec: 237.17 - lr: 0.000007 - momentum: 0.000000 2023-10-10 05:41:47,684 epoch 10 - iter 3647/5212 - loss 0.00895006 - time (sec): 1081.87 - samples/sec: 238.01 - lr: 0.000005 - momentum: 0.000000 2023-10-10 05:44:19,241 epoch 10 - iter 4168/5212 - loss 0.00888338 - time (sec): 1233.43 - samples/sec: 239.39 - lr: 0.000003 - momentum: 0.000000 2023-10-10 05:46:50,896 epoch 10 - iter 4689/5212 - loss 0.00858734 - time (sec): 1385.08 - samples/sec: 240.18 - lr: 0.000002 - momentum: 0.000000 2023-10-10 05:49:20,398 epoch 10 - iter 5210/5212 - loss 0.00861144 - time (sec): 1534.59 - samples/sec: 239.31 - lr: 0.000000 - momentum: 0.000000 2023-10-10 05:49:20,941 ---------------------------------------------------------------------------------------------------- 2023-10-10 05:49:20,942 EPOCH 10 done: loss 0.0086 - lr: 0.000000 2023-10-10 05:50:01,786 DEV : loss 0.507175087928772 - f1-score (micro avg) 0.382 2023-10-10 05:50:02,826 ---------------------------------------------------------------------------------------------------- 2023-10-10 05:50:02,828 Loading model from best epoch ... 2023-10-10 05:50:07,141 SequenceTagger predicts: Dictionary with 17 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, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd 2023-10-10 05:51:51,254 Results: - F-score (micro) 0.4873 - F-score (macro) 0.327 - Accuracy 0.3265 By class: precision recall f1-score support LOC 0.5032 0.6400 0.5635 1214 PER 0.4093 0.5025 0.4511 808 ORG 0.2982 0.2890 0.2935 353 HumanProd 0.0000 0.0000 0.0000 15 micro avg 0.4456 0.5377 0.4873 2390 macro avg 0.3027 0.3579 0.3270 2390 weighted avg 0.4380 0.5377 0.4821 2390 2023-10-10 05:51:51,255 ----------------------------------------------------------------------------------------------------