2023-10-15 02:31:46,751 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:31:46,752 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=21, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-15 02:31:46,752 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:31:46,752 MultiCorpus: 3575 train + 1235 dev + 1266 test sentences - NER_HIPE_2022 Corpus: 3575 train + 1235 dev + 1266 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/de/with_doc_seperator 2023-10-15 02:31:46,752 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:31:46,752 Train: 3575 sentences 2023-10-15 02:31:46,752 (train_with_dev=False, train_with_test=False) 2023-10-15 02:31:46,752 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:31:46,752 Training Params: 2023-10-15 02:31:46,752 - learning_rate: "0.00016" 2023-10-15 02:31:46,752 - mini_batch_size: "4" 2023-10-15 02:31:46,752 - max_epochs: "10" 2023-10-15 02:31:46,752 - shuffle: "True" 2023-10-15 02:31:46,753 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:31:46,753 Plugins: 2023-10-15 02:31:46,753 - TensorboardLogger 2023-10-15 02:31:46,753 - LinearScheduler | warmup_fraction: '0.1' 2023-10-15 02:31:46,753 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:31:46,753 Final evaluation on model from best epoch (best-model.pt) 2023-10-15 02:31:46,753 - metric: "('micro avg', 'f1-score')" 2023-10-15 02:31:46,753 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:31:46,753 Computation: 2023-10-15 02:31:46,753 - compute on device: cuda:0 2023-10-15 02:31:46,753 - embedding storage: none 2023-10-15 02:31:46,753 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:31:46,753 Model training base path: "hmbench-hipe2020/de-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-4" 2023-10-15 02:31:46,753 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:31:46,753 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:31:46,753 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-15 02:32:03,162 epoch 1 - iter 89/894 - loss 3.02344183 - time (sec): 16.41 - samples/sec: 503.29 - lr: 0.000016 - momentum: 0.000000 2023-10-15 02:32:22,172 epoch 1 - iter 178/894 - loss 2.95643439 - time (sec): 35.42 - samples/sec: 517.95 - lr: 0.000032 - momentum: 0.000000 2023-10-15 02:32:39,335 epoch 1 - iter 267/894 - loss 2.78708734 - time (sec): 52.58 - samples/sec: 525.09 - lr: 0.000048 - momentum: 0.000000 2023-10-15 02:32:55,879 epoch 1 - iter 356/894 - loss 2.58160891 - time (sec): 69.13 - samples/sec: 521.14 - lr: 0.000064 - momentum: 0.000000 2023-10-15 02:33:12,285 epoch 1 - iter 445/894 - loss 2.35433260 - time (sec): 85.53 - samples/sec: 520.19 - lr: 0.000079 - momentum: 0.000000 2023-10-15 02:33:29,031 epoch 1 - iter 534/894 - loss 2.10818662 - time (sec): 102.28 - samples/sec: 520.02 - lr: 0.000095 - momentum: 0.000000 2023-10-15 02:33:45,298 epoch 1 - iter 623/894 - loss 1.90435758 - time (sec): 118.54 - samples/sec: 518.72 - lr: 0.000111 - momentum: 0.000000 2023-10-15 02:34:01,309 epoch 1 - iter 712/894 - loss 1.75522635 - time (sec): 134.55 - samples/sec: 514.33 - lr: 0.000127 - momentum: 0.000000 2023-10-15 02:34:17,945 epoch 1 - iter 801/894 - loss 1.61964852 - time (sec): 151.19 - samples/sec: 514.93 - lr: 0.000143 - momentum: 0.000000 2023-10-15 02:34:34,731 epoch 1 - iter 890/894 - loss 1.50331592 - time (sec): 167.98 - samples/sec: 513.77 - lr: 0.000159 - momentum: 0.000000 2023-10-15 02:34:35,368 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:34:35,368 EPOCH 1 done: loss 1.5004 - lr: 0.000159 2023-10-15 02:34:58,797 DEV : loss 0.3780643343925476 - f1-score (micro avg) 0.0 2023-10-15 02:34:58,823 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:35:15,413 epoch 2 - iter 89/894 - loss 0.35647784 - time (sec): 16.59 - samples/sec: 506.48 - lr: 0.000158 - momentum: 0.000000 2023-10-15 02:35:32,068 epoch 2 - iter 178/894 - loss 0.34495992 - time (sec): 33.24 - samples/sec: 488.93 - lr: 0.000156 - momentum: 0.000000 2023-10-15 02:35:48,763 epoch 2 - iter 267/894 - loss 0.32633621 - time (sec): 49.94 - samples/sec: 498.77 - lr: 0.000155 - momentum: 0.000000 2023-10-15 02:36:05,361 epoch 2 - iter 356/894 - loss 0.30999671 - time (sec): 66.54 - samples/sec: 501.31 - lr: 0.000153 - momentum: 0.000000 2023-10-15 02:36:22,230 epoch 2 - iter 445/894 - loss 0.29402029 - time (sec): 83.41 - samples/sec: 504.29 - lr: 0.000151 - momentum: 0.000000 2023-10-15 02:36:38,633 epoch 2 - iter 534/894 - loss 0.29260383 - time (sec): 99.81 - samples/sec: 504.68 - lr: 0.000149 - momentum: 0.000000 2023-10-15 02:36:55,012 epoch 2 - iter 623/894 - loss 0.28062458 - time (sec): 116.19 - samples/sec: 505.86 - lr: 0.000148 - momentum: 0.000000 2023-10-15 02:37:11,215 epoch 2 - iter 712/894 - loss 0.27449208 - time (sec): 132.39 - samples/sec: 505.71 - lr: 0.000146 - momentum: 0.000000 2023-10-15 02:37:28,432 epoch 2 - iter 801/894 - loss 0.26359435 - time (sec): 149.61 - samples/sec: 509.57 - lr: 0.000144 - momentum: 0.000000 2023-10-15 02:37:47,059 epoch 2 - iter 890/894 - loss 0.25229683 - time (sec): 168.23 - samples/sec: 511.90 - lr: 0.000142 - momentum: 0.000000 2023-10-15 02:37:47,825 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:37:47,825 EPOCH 2 done: loss 0.2512 - lr: 0.000142 2023-10-15 02:38:12,900 DEV : loss 0.19516690075397491 - f1-score (micro avg) 0.6578 2023-10-15 02:38:12,926 saving best model 2023-10-15 02:38:13,600 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:38:30,620 epoch 3 - iter 89/894 - loss 0.19012779 - time (sec): 17.02 - samples/sec: 538.74 - lr: 0.000140 - momentum: 0.000000 2023-10-15 02:38:47,328 epoch 3 - iter 178/894 - loss 0.15803457 - time (sec): 33.73 - samples/sec: 540.12 - lr: 0.000139 - momentum: 0.000000 2023-10-15 02:39:04,061 epoch 3 - iter 267/894 - loss 0.14197851 - time (sec): 50.46 - samples/sec: 531.87 - lr: 0.000137 - momentum: 0.000000 2023-10-15 02:39:20,354 epoch 3 - iter 356/894 - loss 0.14302842 - time (sec): 66.75 - samples/sec: 525.00 - lr: 0.000135 - momentum: 0.000000 2023-10-15 02:39:36,895 epoch 3 - iter 445/894 - loss 0.13867980 - time (sec): 83.29 - samples/sec: 524.43 - lr: 0.000133 - momentum: 0.000000 2023-10-15 02:39:52,835 epoch 3 - iter 534/894 - loss 0.13292117 - time (sec): 99.23 - samples/sec: 516.98 - lr: 0.000132 - momentum: 0.000000 2023-10-15 02:40:09,133 epoch 3 - iter 623/894 - loss 0.13020535 - time (sec): 115.53 - samples/sec: 513.60 - lr: 0.000130 - momentum: 0.000000 2023-10-15 02:40:25,404 epoch 3 - iter 712/894 - loss 0.12587355 - time (sec): 131.80 - samples/sec: 512.31 - lr: 0.000128 - momentum: 0.000000 2023-10-15 02:40:43,932 epoch 3 - iter 801/894 - loss 0.12594823 - time (sec): 150.33 - samples/sec: 513.37 - lr: 0.000126 - momentum: 0.000000 2023-10-15 02:41:01,043 epoch 3 - iter 890/894 - loss 0.12151089 - time (sec): 167.44 - samples/sec: 514.04 - lr: 0.000125 - momentum: 0.000000 2023-10-15 02:41:01,800 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:41:01,800 EPOCH 3 done: loss 0.1218 - lr: 0.000125 2023-10-15 02:41:27,066 DEV : loss 0.15591633319854736 - f1-score (micro avg) 0.7298 2023-10-15 02:41:27,092 saving best model 2023-10-15 02:41:27,983 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:41:44,324 epoch 4 - iter 89/894 - loss 0.07103724 - time (sec): 16.34 - samples/sec: 506.27 - lr: 0.000123 - momentum: 0.000000 2023-10-15 02:42:01,963 epoch 4 - iter 178/894 - loss 0.06257623 - time (sec): 33.98 - samples/sec: 517.81 - lr: 0.000121 - momentum: 0.000000 2023-10-15 02:42:18,072 epoch 4 - iter 267/894 - loss 0.06910373 - time (sec): 50.09 - samples/sec: 511.23 - lr: 0.000119 - momentum: 0.000000 2023-10-15 02:42:34,820 epoch 4 - iter 356/894 - loss 0.07107525 - time (sec): 66.83 - samples/sec: 520.12 - lr: 0.000117 - momentum: 0.000000 2023-10-15 02:42:51,325 epoch 4 - iter 445/894 - loss 0.06802427 - time (sec): 83.34 - samples/sec: 522.37 - lr: 0.000116 - momentum: 0.000000 2023-10-15 02:43:07,291 epoch 4 - iter 534/894 - loss 0.06713645 - time (sec): 99.30 - samples/sec: 519.31 - lr: 0.000114 - momentum: 0.000000 2023-10-15 02:43:23,747 epoch 4 - iter 623/894 - loss 0.06876201 - time (sec): 115.76 - samples/sec: 521.62 - lr: 0.000112 - momentum: 0.000000 2023-10-15 02:43:41,754 epoch 4 - iter 712/894 - loss 0.07019890 - time (sec): 133.77 - samples/sec: 519.61 - lr: 0.000110 - momentum: 0.000000 2023-10-15 02:43:57,924 epoch 4 - iter 801/894 - loss 0.06984123 - time (sec): 149.94 - samples/sec: 518.42 - lr: 0.000109 - momentum: 0.000000 2023-10-15 02:44:14,341 epoch 4 - iter 890/894 - loss 0.06910896 - time (sec): 166.35 - samples/sec: 518.79 - lr: 0.000107 - momentum: 0.000000 2023-10-15 02:44:14,960 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:44:14,961 EPOCH 4 done: loss 0.0690 - lr: 0.000107 2023-10-15 02:44:39,949 DEV : loss 0.1609293520450592 - f1-score (micro avg) 0.7516 2023-10-15 02:44:39,975 saving best model 2023-10-15 02:44:40,880 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:44:57,848 epoch 5 - iter 89/894 - loss 0.03775608 - time (sec): 16.97 - samples/sec: 531.50 - lr: 0.000105 - momentum: 0.000000 2023-10-15 02:45:14,380 epoch 5 - iter 178/894 - loss 0.03532971 - time (sec): 33.50 - samples/sec: 524.28 - lr: 0.000103 - momentum: 0.000000 2023-10-15 02:45:31,264 epoch 5 - iter 267/894 - loss 0.03513730 - time (sec): 50.38 - samples/sec: 530.22 - lr: 0.000101 - momentum: 0.000000 2023-10-15 02:45:47,982 epoch 5 - iter 356/894 - loss 0.03993442 - time (sec): 67.10 - samples/sec: 528.81 - lr: 0.000100 - momentum: 0.000000 2023-10-15 02:46:06,528 epoch 5 - iter 445/894 - loss 0.04620953 - time (sec): 85.65 - samples/sec: 529.50 - lr: 0.000098 - momentum: 0.000000 2023-10-15 02:46:22,864 epoch 5 - iter 534/894 - loss 0.04713336 - time (sec): 101.98 - samples/sec: 527.63 - lr: 0.000096 - momentum: 0.000000 2023-10-15 02:46:38,857 epoch 5 - iter 623/894 - loss 0.04525939 - time (sec): 117.97 - samples/sec: 521.37 - lr: 0.000094 - momentum: 0.000000 2023-10-15 02:46:55,181 epoch 5 - iter 712/894 - loss 0.04267360 - time (sec): 134.30 - samples/sec: 519.69 - lr: 0.000093 - momentum: 0.000000 2023-10-15 02:47:11,200 epoch 5 - iter 801/894 - loss 0.04096140 - time (sec): 150.32 - samples/sec: 516.27 - lr: 0.000091 - momentum: 0.000000 2023-10-15 02:47:27,819 epoch 5 - iter 890/894 - loss 0.04206314 - time (sec): 166.94 - samples/sec: 516.61 - lr: 0.000089 - momentum: 0.000000 2023-10-15 02:47:28,490 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:47:28,490 EPOCH 5 done: loss 0.0419 - lr: 0.000089 2023-10-15 02:47:53,602 DEV : loss 0.1873023360967636 - f1-score (micro avg) 0.7704 2023-10-15 02:47:53,628 saving best model 2023-10-15 02:47:54,607 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:48:12,579 epoch 6 - iter 89/894 - loss 0.04090361 - time (sec): 17.97 - samples/sec: 505.20 - lr: 0.000087 - momentum: 0.000000 2023-10-15 02:48:30,268 epoch 6 - iter 178/894 - loss 0.03139198 - time (sec): 35.66 - samples/sec: 515.84 - lr: 0.000085 - momentum: 0.000000 2023-10-15 02:48:46,856 epoch 6 - iter 267/894 - loss 0.03064485 - time (sec): 52.25 - samples/sec: 522.23 - lr: 0.000084 - momentum: 0.000000 2023-10-15 02:49:03,120 epoch 6 - iter 356/894 - loss 0.02852187 - time (sec): 68.51 - samples/sec: 516.68 - lr: 0.000082 - momentum: 0.000000 2023-10-15 02:49:20,293 epoch 6 - iter 445/894 - loss 0.02788748 - time (sec): 85.68 - samples/sec: 508.91 - lr: 0.000080 - momentum: 0.000000 2023-10-15 02:49:37,772 epoch 6 - iter 534/894 - loss 0.02794351 - time (sec): 103.16 - samples/sec: 506.10 - lr: 0.000078 - momentum: 0.000000 2023-10-15 02:49:55,339 epoch 6 - iter 623/894 - loss 0.02802367 - time (sec): 120.73 - samples/sec: 502.99 - lr: 0.000077 - momentum: 0.000000 2023-10-15 02:50:12,915 epoch 6 - iter 712/894 - loss 0.02792162 - time (sec): 138.31 - samples/sec: 500.00 - lr: 0.000075 - momentum: 0.000000 2023-10-15 02:50:29,545 epoch 6 - iter 801/894 - loss 0.02919230 - time (sec): 154.94 - samples/sec: 499.81 - lr: 0.000073 - momentum: 0.000000 2023-10-15 02:50:46,019 epoch 6 - iter 890/894 - loss 0.02781489 - time (sec): 171.41 - samples/sec: 503.33 - lr: 0.000071 - momentum: 0.000000 2023-10-15 02:50:46,685 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:50:46,685 EPOCH 6 done: loss 0.0278 - lr: 0.000071 2023-10-15 02:51:13,116 DEV : loss 0.21434210240840912 - f1-score (micro avg) 0.7671 2023-10-15 02:51:13,144 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:51:32,358 epoch 7 - iter 89/894 - loss 0.03055746 - time (sec): 19.21 - samples/sec: 506.79 - lr: 0.000069 - momentum: 0.000000 2023-10-15 02:51:49,459 epoch 7 - iter 178/894 - loss 0.03178652 - time (sec): 36.31 - samples/sec: 508.13 - lr: 0.000068 - momentum: 0.000000 2023-10-15 02:52:05,517 epoch 7 - iter 267/894 - loss 0.02947788 - time (sec): 52.37 - samples/sec: 496.56 - lr: 0.000066 - momentum: 0.000000 2023-10-15 02:52:21,830 epoch 7 - iter 356/894 - loss 0.02621194 - time (sec): 68.68 - samples/sec: 497.65 - lr: 0.000064 - momentum: 0.000000 2023-10-15 02:52:38,189 epoch 7 - iter 445/894 - loss 0.02435064 - time (sec): 85.04 - samples/sec: 498.38 - lr: 0.000062 - momentum: 0.000000 2023-10-15 02:52:54,896 epoch 7 - iter 534/894 - loss 0.02215860 - time (sec): 101.75 - samples/sec: 503.67 - lr: 0.000061 - momentum: 0.000000 2023-10-15 02:53:11,073 epoch 7 - iter 623/894 - loss 0.02099170 - time (sec): 117.93 - samples/sec: 501.57 - lr: 0.000059 - momentum: 0.000000 2023-10-15 02:53:27,755 epoch 7 - iter 712/894 - loss 0.01995778 - time (sec): 134.61 - samples/sec: 504.04 - lr: 0.000057 - momentum: 0.000000 2023-10-15 02:53:44,630 epoch 7 - iter 801/894 - loss 0.01902734 - time (sec): 151.48 - samples/sec: 507.27 - lr: 0.000055 - momentum: 0.000000 2023-10-15 02:54:02,012 epoch 7 - iter 890/894 - loss 0.01833316 - time (sec): 168.87 - samples/sec: 510.97 - lr: 0.000053 - momentum: 0.000000 2023-10-15 02:54:02,650 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:54:02,651 EPOCH 7 done: loss 0.0185 - lr: 0.000053 2023-10-15 02:54:28,844 DEV : loss 0.21627573668956757 - f1-score (micro avg) 0.7695 2023-10-15 02:54:28,870 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:54:45,753 epoch 8 - iter 89/894 - loss 0.01084712 - time (sec): 16.88 - samples/sec: 505.28 - lr: 0.000052 - momentum: 0.000000 2023-10-15 02:55:03,142 epoch 8 - iter 178/894 - loss 0.01079641 - time (sec): 34.27 - samples/sec: 513.12 - lr: 0.000050 - momentum: 0.000000 2023-10-15 02:55:20,027 epoch 8 - iter 267/894 - loss 0.00894694 - time (sec): 51.16 - samples/sec: 521.76 - lr: 0.000048 - momentum: 0.000000 2023-10-15 02:55:36,682 epoch 8 - iter 356/894 - loss 0.00974459 - time (sec): 67.81 - samples/sec: 522.44 - lr: 0.000046 - momentum: 0.000000 2023-10-15 02:55:53,263 epoch 8 - iter 445/894 - loss 0.01110363 - time (sec): 84.39 - samples/sec: 522.29 - lr: 0.000045 - momentum: 0.000000 2023-10-15 02:56:09,389 epoch 8 - iter 534/894 - loss 0.01181829 - time (sec): 100.52 - samples/sec: 517.26 - lr: 0.000043 - momentum: 0.000000 2023-10-15 02:56:26,017 epoch 8 - iter 623/894 - loss 0.01176535 - time (sec): 117.15 - samples/sec: 516.61 - lr: 0.000041 - momentum: 0.000000 2023-10-15 02:56:42,522 epoch 8 - iter 712/894 - loss 0.01085512 - time (sec): 133.65 - samples/sec: 516.06 - lr: 0.000039 - momentum: 0.000000 2023-10-15 02:57:01,040 epoch 8 - iter 801/894 - loss 0.01135469 - time (sec): 152.17 - samples/sec: 515.39 - lr: 0.000038 - momentum: 0.000000 2023-10-15 02:57:17,154 epoch 8 - iter 890/894 - loss 0.01148836 - time (sec): 168.28 - samples/sec: 512.32 - lr: 0.000036 - momentum: 0.000000 2023-10-15 02:57:17,838 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:57:17,839 EPOCH 8 done: loss 0.0118 - lr: 0.000036 2023-10-15 02:57:44,212 DEV : loss 0.22242531180381775 - f1-score (micro avg) 0.7719 2023-10-15 02:57:44,240 saving best model 2023-10-15 02:57:46,752 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:58:02,946 epoch 9 - iter 89/894 - loss 0.00595299 - time (sec): 16.19 - samples/sec: 494.63 - lr: 0.000034 - momentum: 0.000000 2023-10-15 02:58:19,361 epoch 9 - iter 178/894 - loss 0.00644078 - time (sec): 32.61 - samples/sec: 498.46 - lr: 0.000032 - momentum: 0.000000 2023-10-15 02:58:35,843 epoch 9 - iter 267/894 - loss 0.01132599 - time (sec): 49.09 - samples/sec: 505.33 - lr: 0.000030 - momentum: 0.000000 2023-10-15 02:58:52,419 epoch 9 - iter 356/894 - loss 0.00904748 - time (sec): 65.67 - samples/sec: 510.07 - lr: 0.000029 - momentum: 0.000000 2023-10-15 02:59:10,359 epoch 9 - iter 445/894 - loss 0.01098782 - time (sec): 83.60 - samples/sec: 506.84 - lr: 0.000027 - momentum: 0.000000 2023-10-15 02:59:27,525 epoch 9 - iter 534/894 - loss 0.01062907 - time (sec): 100.77 - samples/sec: 512.80 - lr: 0.000025 - momentum: 0.000000 2023-10-15 02:59:44,394 epoch 9 - iter 623/894 - loss 0.01018680 - time (sec): 117.64 - samples/sec: 512.48 - lr: 0.000023 - momentum: 0.000000 2023-10-15 03:00:01,280 epoch 9 - iter 712/894 - loss 0.00968917 - time (sec): 134.53 - samples/sec: 513.69 - lr: 0.000022 - momentum: 0.000000 2023-10-15 03:00:17,808 epoch 9 - iter 801/894 - loss 0.00910352 - time (sec): 151.05 - samples/sec: 515.92 - lr: 0.000020 - momentum: 0.000000 2023-10-15 03:00:34,257 epoch 9 - iter 890/894 - loss 0.00910813 - time (sec): 167.50 - samples/sec: 514.73 - lr: 0.000018 - momentum: 0.000000 2023-10-15 03:00:34,964 ---------------------------------------------------------------------------------------------------- 2023-10-15 03:00:34,964 EPOCH 9 done: loss 0.0091 - lr: 0.000018 2023-10-15 03:01:01,017 DEV : loss 0.23208962380886078 - f1-score (micro avg) 0.7894 2023-10-15 03:01:01,043 saving best model 2023-10-15 03:01:03,945 ---------------------------------------------------------------------------------------------------- 2023-10-15 03:01:22,920 epoch 10 - iter 89/894 - loss 0.00741155 - time (sec): 18.97 - samples/sec: 530.99 - lr: 0.000016 - momentum: 0.000000 2023-10-15 03:01:39,312 epoch 10 - iter 178/894 - loss 0.00663561 - time (sec): 35.36 - samples/sec: 518.24 - lr: 0.000014 - momentum: 0.000000 2023-10-15 03:01:55,571 epoch 10 - iter 267/894 - loss 0.00512243 - time (sec): 51.62 - samples/sec: 515.38 - lr: 0.000013 - momentum: 0.000000 2023-10-15 03:02:11,926 epoch 10 - iter 356/894 - loss 0.00494199 - time (sec): 67.98 - samples/sec: 513.59 - lr: 0.000011 - momentum: 0.000000 2023-10-15 03:02:29,205 epoch 10 - iter 445/894 - loss 0.00519207 - time (sec): 85.26 - samples/sec: 519.56 - lr: 0.000009 - momentum: 0.000000 2023-10-15 03:02:45,598 epoch 10 - iter 534/894 - loss 0.00528129 - time (sec): 101.65 - samples/sec: 517.28 - lr: 0.000007 - momentum: 0.000000 2023-10-15 03:03:02,594 epoch 10 - iter 623/894 - loss 0.00652465 - time (sec): 118.65 - samples/sec: 515.94 - lr: 0.000006 - momentum: 0.000000 2023-10-15 03:03:18,989 epoch 10 - iter 712/894 - loss 0.00616289 - time (sec): 135.04 - samples/sec: 517.96 - lr: 0.000004 - momentum: 0.000000 2023-10-15 03:03:35,161 epoch 10 - iter 801/894 - loss 0.00585123 - time (sec): 151.21 - samples/sec: 516.29 - lr: 0.000002 - momentum: 0.000000 2023-10-15 03:03:51,457 epoch 10 - iter 890/894 - loss 0.00587903 - time (sec): 167.51 - samples/sec: 515.29 - lr: 0.000000 - momentum: 0.000000 2023-10-15 03:03:52,093 ---------------------------------------------------------------------------------------------------- 2023-10-15 03:03:52,093 EPOCH 10 done: loss 0.0059 - lr: 0.000000 2023-10-15 03:04:17,600 DEV : loss 0.2378893792629242 - f1-score (micro avg) 0.7848 2023-10-15 03:04:18,259 ---------------------------------------------------------------------------------------------------- 2023-10-15 03:04:18,261 Loading model from best epoch ... 2023-10-15 03:04:26,176 SequenceTagger predicts: Dictionary with 21 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, S-prod, B-prod, E-prod, I-prod, S-time, B-time, E-time, I-time 2023-10-15 03:04:49,929 Results: - F-score (micro) 0.7741 - F-score (macro) 0.6991 - Accuracy 0.6448 By class: precision recall f1-score support loc 0.8588 0.8674 0.8631 596 pers 0.6898 0.7748 0.7298 333 org 0.5923 0.5833 0.5878 132 prod 0.6731 0.5303 0.5932 66 time 0.7292 0.7143 0.7216 49 micro avg 0.7645 0.7840 0.7741 1176 macro avg 0.7086 0.6940 0.6991 1176 weighted avg 0.7652 0.7840 0.7734 1176 2023-10-15 03:04:49,929 ----------------------------------------------------------------------------------------------------