2023-10-17 17:26:59,563 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:26:59,564 Model: "SequenceTagger( (embeddings): TransformerWordEmbeddings( (model): ElectraModel( (embeddings): ElectraEmbeddings( (word_embeddings): Embedding(32001, 768) (position_embeddings): Embedding(512, 768) (token_type_embeddings): Embedding(2, 768) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) (encoder): ElectraEncoder( (layer): ModuleList( (0-11): 12 x ElectraLayer( (attention): ElectraAttention( (self): ElectraSelfAttention( (query): Linear(in_features=768, out_features=768, bias=True) (key): Linear(in_features=768, out_features=768, bias=True) (value): Linear(in_features=768, out_features=768, bias=True) (dropout): Dropout(p=0.1, inplace=False) ) (output): ElectraSelfOutput( (dense): Linear(in_features=768, out_features=768, bias=True) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) (intermediate): ElectraIntermediate( (dense): Linear(in_features=768, out_features=3072, bias=True) (intermediate_act_fn): GELUActivation() ) (output): ElectraOutput( (dense): Linear(in_features=3072, out_features=768, bias=True) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) ) ) ) (locked_dropout): LockedDropout(p=0.5) (linear): Linear(in_features=768, out_features=17, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-17 17:26:59,564 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:26:59,564 MultiCorpus: 1166 train + 165 dev + 415 test sentences - NER_HIPE_2022 Corpus: 1166 train + 165 dev + 415 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fi/with_doc_seperator 2023-10-17 17:26:59,564 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:26:59,564 Train: 1166 sentences 2023-10-17 17:26:59,564 (train_with_dev=False, train_with_test=False) 2023-10-17 17:26:59,565 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:26:59,565 Training Params: 2023-10-17 17:26:59,565 - learning_rate: "3e-05" 2023-10-17 17:26:59,565 - mini_batch_size: "4" 2023-10-17 17:26:59,565 - max_epochs: "10" 2023-10-17 17:26:59,565 - shuffle: "True" 2023-10-17 17:26:59,565 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:26:59,565 Plugins: 2023-10-17 17:26:59,565 - TensorboardLogger 2023-10-17 17:26:59,565 - LinearScheduler | warmup_fraction: '0.1' 2023-10-17 17:26:59,565 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:26:59,565 Final evaluation on model from best epoch (best-model.pt) 2023-10-17 17:26:59,565 - metric: "('micro avg', 'f1-score')" 2023-10-17 17:26:59,565 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:26:59,565 Computation: 2023-10-17 17:26:59,565 - compute on device: cuda:0 2023-10-17 17:26:59,565 - embedding storage: none 2023-10-17 17:26:59,565 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:26:59,565 Model training base path: "hmbench-newseye/fi-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-1" 2023-10-17 17:26:59,566 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:26:59,566 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:26:59,566 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-17 17:27:02,559 epoch 1 - iter 29/292 - loss 3.38067707 - time (sec): 2.99 - samples/sec: 1274.79 - lr: 0.000003 - momentum: 0.000000 2023-10-17 17:27:04,259 epoch 1 - iter 58/292 - loss 2.84271477 - time (sec): 4.69 - samples/sec: 1733.47 - lr: 0.000006 - momentum: 0.000000 2023-10-17 17:27:06,435 epoch 1 - iter 87/292 - loss 2.15025489 - time (sec): 6.87 - samples/sec: 2025.87 - lr: 0.000009 - momentum: 0.000000 2023-10-17 17:27:08,024 epoch 1 - iter 116/292 - loss 1.73892010 - time (sec): 8.46 - samples/sec: 2196.36 - lr: 0.000012 - momentum: 0.000000 2023-10-17 17:27:09,750 epoch 1 - iter 145/292 - loss 1.52932502 - time (sec): 10.18 - samples/sec: 2213.85 - lr: 0.000015 - momentum: 0.000000 2023-10-17 17:27:11,391 epoch 1 - iter 174/292 - loss 1.35737795 - time (sec): 11.82 - samples/sec: 2255.55 - lr: 0.000018 - momentum: 0.000000 2023-10-17 17:27:13,018 epoch 1 - iter 203/292 - loss 1.21956588 - time (sec): 13.45 - samples/sec: 2279.46 - lr: 0.000021 - momentum: 0.000000 2023-10-17 17:27:14,949 epoch 1 - iter 232/292 - loss 1.09424639 - time (sec): 15.38 - samples/sec: 2310.72 - lr: 0.000024 - momentum: 0.000000 2023-10-17 17:27:16,712 epoch 1 - iter 261/292 - loss 1.01634785 - time (sec): 17.14 - samples/sec: 2301.10 - lr: 0.000027 - momentum: 0.000000 2023-10-17 17:27:18,606 epoch 1 - iter 290/292 - loss 0.93625762 - time (sec): 19.04 - samples/sec: 2320.43 - lr: 0.000030 - momentum: 0.000000 2023-10-17 17:27:18,714 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:27:18,715 EPOCH 1 done: loss 0.9322 - lr: 0.000030 2023-10-17 17:27:19,545 DEV : loss 0.19820411503314972 - f1-score (micro avg) 0.4404 2023-10-17 17:27:19,551 saving best model 2023-10-17 17:27:19,982 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:27:21,872 epoch 2 - iter 29/292 - loss 0.22906131 - time (sec): 1.89 - samples/sec: 2440.74 - lr: 0.000030 - momentum: 0.000000 2023-10-17 17:27:23,585 epoch 2 - iter 58/292 - loss 0.20380503 - time (sec): 3.60 - samples/sec: 2427.27 - lr: 0.000029 - momentum: 0.000000 2023-10-17 17:27:25,351 epoch 2 - iter 87/292 - loss 0.19601726 - time (sec): 5.37 - samples/sec: 2453.85 - lr: 0.000029 - momentum: 0.000000 2023-10-17 17:27:27,325 epoch 2 - iter 116/292 - loss 0.18101298 - time (sec): 7.34 - samples/sec: 2492.83 - lr: 0.000029 - momentum: 0.000000 2023-10-17 17:27:29,067 epoch 2 - iter 145/292 - loss 0.17978471 - time (sec): 9.08 - samples/sec: 2536.20 - lr: 0.000028 - momentum: 0.000000 2023-10-17 17:27:30,696 epoch 2 - iter 174/292 - loss 0.17845205 - time (sec): 10.71 - samples/sec: 2574.15 - lr: 0.000028 - momentum: 0.000000 2023-10-17 17:27:32,232 epoch 2 - iter 203/292 - loss 0.18417685 - time (sec): 12.25 - samples/sec: 2562.26 - lr: 0.000028 - momentum: 0.000000 2023-10-17 17:27:33,820 epoch 2 - iter 232/292 - loss 0.19378925 - time (sec): 13.84 - samples/sec: 2538.03 - lr: 0.000027 - momentum: 0.000000 2023-10-17 17:27:35,628 epoch 2 - iter 261/292 - loss 0.19190160 - time (sec): 15.64 - samples/sec: 2570.21 - lr: 0.000027 - momentum: 0.000000 2023-10-17 17:27:37,313 epoch 2 - iter 290/292 - loss 0.18637235 - time (sec): 17.33 - samples/sec: 2554.65 - lr: 0.000027 - momentum: 0.000000 2023-10-17 17:27:37,418 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:27:37,419 EPOCH 2 done: loss 0.1866 - lr: 0.000027 2023-10-17 17:27:38,858 DEV : loss 0.13132032752037048 - f1-score (micro avg) 0.6523 2023-10-17 17:27:38,881 saving best model 2023-10-17 17:27:39,381 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:27:41,149 epoch 3 - iter 29/292 - loss 0.11668019 - time (sec): 1.76 - samples/sec: 2582.00 - lr: 0.000026 - momentum: 0.000000 2023-10-17 17:27:42,789 epoch 3 - iter 58/292 - loss 0.10524911 - time (sec): 3.40 - samples/sec: 2641.65 - lr: 0.000026 - momentum: 0.000000 2023-10-17 17:27:44,415 epoch 3 - iter 87/292 - loss 0.12104987 - time (sec): 5.03 - samples/sec: 2720.98 - lr: 0.000026 - momentum: 0.000000 2023-10-17 17:27:45,968 epoch 3 - iter 116/292 - loss 0.11755234 - time (sec): 6.58 - samples/sec: 2686.04 - lr: 0.000025 - momentum: 0.000000 2023-10-17 17:27:47,599 epoch 3 - iter 145/292 - loss 0.11284499 - time (sec): 8.21 - samples/sec: 2658.66 - lr: 0.000025 - momentum: 0.000000 2023-10-17 17:27:49,217 epoch 3 - iter 174/292 - loss 0.10811762 - time (sec): 9.83 - samples/sec: 2628.27 - lr: 0.000025 - momentum: 0.000000 2023-10-17 17:27:51,020 epoch 3 - iter 203/292 - loss 0.10863850 - time (sec): 11.63 - samples/sec: 2659.81 - lr: 0.000024 - momentum: 0.000000 2023-10-17 17:27:52,752 epoch 3 - iter 232/292 - loss 0.10627802 - time (sec): 13.37 - samples/sec: 2663.02 - lr: 0.000024 - momentum: 0.000000 2023-10-17 17:27:54,311 epoch 3 - iter 261/292 - loss 0.10575975 - time (sec): 14.93 - samples/sec: 2655.88 - lr: 0.000024 - momentum: 0.000000 2023-10-17 17:27:56,137 epoch 3 - iter 290/292 - loss 0.10736363 - time (sec): 16.75 - samples/sec: 2634.68 - lr: 0.000023 - momentum: 0.000000 2023-10-17 17:27:56,257 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:27:56,257 EPOCH 3 done: loss 0.1078 - lr: 0.000023 2023-10-17 17:27:57,516 DEV : loss 0.1270725578069687 - f1-score (micro avg) 0.7313 2023-10-17 17:27:57,521 saving best model 2023-10-17 17:27:57,991 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:27:59,500 epoch 4 - iter 29/292 - loss 0.11337783 - time (sec): 1.50 - samples/sec: 2364.25 - lr: 0.000023 - momentum: 0.000000 2023-10-17 17:28:01,109 epoch 4 - iter 58/292 - loss 0.09246785 - time (sec): 3.11 - samples/sec: 2522.43 - lr: 0.000023 - momentum: 0.000000 2023-10-17 17:28:03,000 epoch 4 - iter 87/292 - loss 0.07732073 - time (sec): 5.01 - samples/sec: 2562.90 - lr: 0.000022 - momentum: 0.000000 2023-10-17 17:28:04,764 epoch 4 - iter 116/292 - loss 0.07279139 - time (sec): 6.77 - samples/sec: 2578.42 - lr: 0.000022 - momentum: 0.000000 2023-10-17 17:28:06,391 epoch 4 - iter 145/292 - loss 0.06862627 - time (sec): 8.40 - samples/sec: 2594.33 - lr: 0.000022 - momentum: 0.000000 2023-10-17 17:28:08,168 epoch 4 - iter 174/292 - loss 0.06941418 - time (sec): 10.17 - samples/sec: 2623.37 - lr: 0.000021 - momentum: 0.000000 2023-10-17 17:28:09,936 epoch 4 - iter 203/292 - loss 0.07292749 - time (sec): 11.94 - samples/sec: 2633.95 - lr: 0.000021 - momentum: 0.000000 2023-10-17 17:28:11,626 epoch 4 - iter 232/292 - loss 0.07213824 - time (sec): 13.63 - samples/sec: 2620.05 - lr: 0.000021 - momentum: 0.000000 2023-10-17 17:28:13,204 epoch 4 - iter 261/292 - loss 0.07275720 - time (sec): 15.21 - samples/sec: 2641.76 - lr: 0.000020 - momentum: 0.000000 2023-10-17 17:28:14,908 epoch 4 - iter 290/292 - loss 0.07210992 - time (sec): 16.91 - samples/sec: 2611.71 - lr: 0.000020 - momentum: 0.000000 2023-10-17 17:28:15,014 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:28:15,015 EPOCH 4 done: loss 0.0719 - lr: 0.000020 2023-10-17 17:28:16,266 DEV : loss 0.12534381449222565 - f1-score (micro avg) 0.7749 2023-10-17 17:28:16,271 saving best model 2023-10-17 17:28:16,794 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:28:18,744 epoch 5 - iter 29/292 - loss 0.05953152 - time (sec): 1.95 - samples/sec: 2665.79 - lr: 0.000020 - momentum: 0.000000 2023-10-17 17:28:20,508 epoch 5 - iter 58/292 - loss 0.04305700 - time (sec): 3.71 - samples/sec: 2662.09 - lr: 0.000019 - momentum: 0.000000 2023-10-17 17:28:22,098 epoch 5 - iter 87/292 - loss 0.05274874 - time (sec): 5.30 - samples/sec: 2689.83 - lr: 0.000019 - momentum: 0.000000 2023-10-17 17:28:23,608 epoch 5 - iter 116/292 - loss 0.05816026 - time (sec): 6.81 - samples/sec: 2652.68 - lr: 0.000019 - momentum: 0.000000 2023-10-17 17:28:25,334 epoch 5 - iter 145/292 - loss 0.05818204 - time (sec): 8.54 - samples/sec: 2630.17 - lr: 0.000018 - momentum: 0.000000 2023-10-17 17:28:27,100 epoch 5 - iter 174/292 - loss 0.05810976 - time (sec): 10.30 - samples/sec: 2649.02 - lr: 0.000018 - momentum: 0.000000 2023-10-17 17:28:28,766 epoch 5 - iter 203/292 - loss 0.05835023 - time (sec): 11.97 - samples/sec: 2652.40 - lr: 0.000018 - momentum: 0.000000 2023-10-17 17:28:30,397 epoch 5 - iter 232/292 - loss 0.05556762 - time (sec): 13.60 - samples/sec: 2658.25 - lr: 0.000017 - momentum: 0.000000 2023-10-17 17:28:32,060 epoch 5 - iter 261/292 - loss 0.05489206 - time (sec): 15.26 - samples/sec: 2619.23 - lr: 0.000017 - momentum: 0.000000 2023-10-17 17:28:33,627 epoch 5 - iter 290/292 - loss 0.05276153 - time (sec): 16.83 - samples/sec: 2620.96 - lr: 0.000017 - momentum: 0.000000 2023-10-17 17:28:33,744 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:28:33,744 EPOCH 5 done: loss 0.0524 - lr: 0.000017 2023-10-17 17:28:35,003 DEV : loss 0.1388275921344757 - f1-score (micro avg) 0.7404 2023-10-17 17:28:35,007 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:28:36,580 epoch 6 - iter 29/292 - loss 0.04318866 - time (sec): 1.57 - samples/sec: 2847.50 - lr: 0.000016 - momentum: 0.000000 2023-10-17 17:28:38,085 epoch 6 - iter 58/292 - loss 0.04282065 - time (sec): 3.08 - samples/sec: 2629.42 - lr: 0.000016 - momentum: 0.000000 2023-10-17 17:28:39,949 epoch 6 - iter 87/292 - loss 0.04016784 - time (sec): 4.94 - samples/sec: 2609.49 - lr: 0.000016 - momentum: 0.000000 2023-10-17 17:28:41,597 epoch 6 - iter 116/292 - loss 0.04140804 - time (sec): 6.59 - samples/sec: 2608.75 - lr: 0.000015 - momentum: 0.000000 2023-10-17 17:28:43,095 epoch 6 - iter 145/292 - loss 0.04001813 - time (sec): 8.09 - samples/sec: 2562.13 - lr: 0.000015 - momentum: 0.000000 2023-10-17 17:28:44,814 epoch 6 - iter 174/292 - loss 0.03934836 - time (sec): 9.80 - samples/sec: 2549.72 - lr: 0.000015 - momentum: 0.000000 2023-10-17 17:28:46,650 epoch 6 - iter 203/292 - loss 0.03998330 - time (sec): 11.64 - samples/sec: 2553.75 - lr: 0.000014 - momentum: 0.000000 2023-10-17 17:28:48,456 epoch 6 - iter 232/292 - loss 0.03957101 - time (sec): 13.45 - samples/sec: 2559.54 - lr: 0.000014 - momentum: 0.000000 2023-10-17 17:28:50,115 epoch 6 - iter 261/292 - loss 0.03892236 - time (sec): 15.11 - samples/sec: 2565.17 - lr: 0.000014 - momentum: 0.000000 2023-10-17 17:28:51,917 epoch 6 - iter 290/292 - loss 0.03724276 - time (sec): 16.91 - samples/sec: 2618.81 - lr: 0.000013 - momentum: 0.000000 2023-10-17 17:28:52,008 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:28:52,008 EPOCH 6 done: loss 0.0373 - lr: 0.000013 2023-10-17 17:28:53,260 DEV : loss 0.15327706933021545 - f1-score (micro avg) 0.7414 2023-10-17 17:28:53,265 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:28:55,107 epoch 7 - iter 29/292 - loss 0.01499409 - time (sec): 1.84 - samples/sec: 2673.73 - lr: 0.000013 - momentum: 0.000000 2023-10-17 17:28:56,892 epoch 7 - iter 58/292 - loss 0.02746269 - time (sec): 3.63 - samples/sec: 2535.72 - lr: 0.000013 - momentum: 0.000000 2023-10-17 17:28:58,571 epoch 7 - iter 87/292 - loss 0.03058176 - time (sec): 5.31 - samples/sec: 2609.09 - lr: 0.000012 - momentum: 0.000000 2023-10-17 17:29:00,399 epoch 7 - iter 116/292 - loss 0.02783010 - time (sec): 7.13 - samples/sec: 2613.54 - lr: 0.000012 - momentum: 0.000000 2023-10-17 17:29:02,208 epoch 7 - iter 145/292 - loss 0.02763787 - time (sec): 8.94 - samples/sec: 2636.93 - lr: 0.000012 - momentum: 0.000000 2023-10-17 17:29:03,848 epoch 7 - iter 174/292 - loss 0.02864891 - time (sec): 10.58 - samples/sec: 2631.66 - lr: 0.000011 - momentum: 0.000000 2023-10-17 17:29:05,535 epoch 7 - iter 203/292 - loss 0.02705060 - time (sec): 12.27 - samples/sec: 2655.11 - lr: 0.000011 - momentum: 0.000000 2023-10-17 17:29:07,098 epoch 7 - iter 232/292 - loss 0.02619522 - time (sec): 13.83 - samples/sec: 2656.91 - lr: 0.000011 - momentum: 0.000000 2023-10-17 17:29:08,606 epoch 7 - iter 261/292 - loss 0.02763773 - time (sec): 15.34 - samples/sec: 2636.76 - lr: 0.000010 - momentum: 0.000000 2023-10-17 17:29:10,173 epoch 7 - iter 290/292 - loss 0.02674452 - time (sec): 16.91 - samples/sec: 2618.73 - lr: 0.000010 - momentum: 0.000000 2023-10-17 17:29:10,271 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:29:10,271 EPOCH 7 done: loss 0.0267 - lr: 0.000010 2023-10-17 17:29:11,518 DEV : loss 0.16855847835540771 - f1-score (micro avg) 0.7511 2023-10-17 17:29:11,523 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:29:13,452 epoch 8 - iter 29/292 - loss 0.02027046 - time (sec): 1.93 - samples/sec: 2456.17 - lr: 0.000010 - momentum: 0.000000 2023-10-17 17:29:15,350 epoch 8 - iter 58/292 - loss 0.02601394 - time (sec): 3.83 - samples/sec: 2461.62 - lr: 0.000009 - momentum: 0.000000 2023-10-17 17:29:17,197 epoch 8 - iter 87/292 - loss 0.02915936 - time (sec): 5.67 - samples/sec: 2511.81 - lr: 0.000009 - momentum: 0.000000 2023-10-17 17:29:18,764 epoch 8 - iter 116/292 - loss 0.02597150 - time (sec): 7.24 - samples/sec: 2514.51 - lr: 0.000009 - momentum: 0.000000 2023-10-17 17:29:20,355 epoch 8 - iter 145/292 - loss 0.02298947 - time (sec): 8.83 - samples/sec: 2498.33 - lr: 0.000008 - momentum: 0.000000 2023-10-17 17:29:21,943 epoch 8 - iter 174/292 - loss 0.02214849 - time (sec): 10.42 - samples/sec: 2477.29 - lr: 0.000008 - momentum: 0.000000 2023-10-17 17:29:23,569 epoch 8 - iter 203/292 - loss 0.02160627 - time (sec): 12.04 - samples/sec: 2515.45 - lr: 0.000008 - momentum: 0.000000 2023-10-17 17:29:25,193 epoch 8 - iter 232/292 - loss 0.01994902 - time (sec): 13.67 - samples/sec: 2512.24 - lr: 0.000007 - momentum: 0.000000 2023-10-17 17:29:27,208 epoch 8 - iter 261/292 - loss 0.02019007 - time (sec): 15.68 - samples/sec: 2572.58 - lr: 0.000007 - momentum: 0.000000 2023-10-17 17:29:28,717 epoch 8 - iter 290/292 - loss 0.02000299 - time (sec): 17.19 - samples/sec: 2580.42 - lr: 0.000007 - momentum: 0.000000 2023-10-17 17:29:28,810 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:29:28,811 EPOCH 8 done: loss 0.0200 - lr: 0.000007 2023-10-17 17:29:30,046 DEV : loss 0.16279011964797974 - f1-score (micro avg) 0.7297 2023-10-17 17:29:30,050 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:29:31,831 epoch 9 - iter 29/292 - loss 0.01144995 - time (sec): 1.78 - samples/sec: 2396.24 - lr: 0.000006 - momentum: 0.000000 2023-10-17 17:29:33,608 epoch 9 - iter 58/292 - loss 0.01623989 - time (sec): 3.56 - samples/sec: 2291.67 - lr: 0.000006 - momentum: 0.000000 2023-10-17 17:29:35,363 epoch 9 - iter 87/292 - loss 0.01401842 - time (sec): 5.31 - samples/sec: 2234.76 - lr: 0.000006 - momentum: 0.000000 2023-10-17 17:29:37,138 epoch 9 - iter 116/292 - loss 0.01205765 - time (sec): 7.09 - samples/sec: 2251.76 - lr: 0.000005 - momentum: 0.000000 2023-10-17 17:29:38,916 epoch 9 - iter 145/292 - loss 0.01173432 - time (sec): 8.86 - samples/sec: 2348.03 - lr: 0.000005 - momentum: 0.000000 2023-10-17 17:29:40,652 epoch 9 - iter 174/292 - loss 0.01383228 - time (sec): 10.60 - samples/sec: 2400.23 - lr: 0.000005 - momentum: 0.000000 2023-10-17 17:29:42,275 epoch 9 - iter 203/292 - loss 0.01267356 - time (sec): 12.22 - samples/sec: 2423.12 - lr: 0.000004 - momentum: 0.000000 2023-10-17 17:29:43,999 epoch 9 - iter 232/292 - loss 0.01196640 - time (sec): 13.95 - samples/sec: 2479.86 - lr: 0.000004 - momentum: 0.000000 2023-10-17 17:29:45,663 epoch 9 - iter 261/292 - loss 0.01366318 - time (sec): 15.61 - samples/sec: 2524.91 - lr: 0.000004 - momentum: 0.000000 2023-10-17 17:29:47,441 epoch 9 - iter 290/292 - loss 0.01473423 - time (sec): 17.39 - samples/sec: 2527.37 - lr: 0.000003 - momentum: 0.000000 2023-10-17 17:29:47,644 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:29:47,644 EPOCH 9 done: loss 0.0147 - lr: 0.000003 2023-10-17 17:29:48,905 DEV : loss 0.175223708152771 - f1-score (micro avg) 0.7467 2023-10-17 17:29:48,910 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:29:50,670 epoch 10 - iter 29/292 - loss 0.01438218 - time (sec): 1.76 - samples/sec: 2780.63 - lr: 0.000003 - momentum: 0.000000 2023-10-17 17:29:52,357 epoch 10 - iter 58/292 - loss 0.01415184 - time (sec): 3.45 - samples/sec: 2802.79 - lr: 0.000003 - momentum: 0.000000 2023-10-17 17:29:54,032 epoch 10 - iter 87/292 - loss 0.01816413 - time (sec): 5.12 - samples/sec: 2710.02 - lr: 0.000002 - momentum: 0.000000 2023-10-17 17:29:55,669 epoch 10 - iter 116/292 - loss 0.01481988 - time (sec): 6.76 - samples/sec: 2678.17 - lr: 0.000002 - momentum: 0.000000 2023-10-17 17:29:57,326 epoch 10 - iter 145/292 - loss 0.01203336 - time (sec): 8.41 - samples/sec: 2674.61 - lr: 0.000002 - momentum: 0.000000 2023-10-17 17:29:59,003 epoch 10 - iter 174/292 - loss 0.01214143 - time (sec): 10.09 - samples/sec: 2658.32 - lr: 0.000001 - momentum: 0.000000 2023-10-17 17:30:00,852 epoch 10 - iter 203/292 - loss 0.01276323 - time (sec): 11.94 - samples/sec: 2663.21 - lr: 0.000001 - momentum: 0.000000 2023-10-17 17:30:02,439 epoch 10 - iter 232/292 - loss 0.01297237 - time (sec): 13.53 - samples/sec: 2638.23 - lr: 0.000001 - momentum: 0.000000 2023-10-17 17:30:04,230 epoch 10 - iter 261/292 - loss 0.01231085 - time (sec): 15.32 - samples/sec: 2651.34 - lr: 0.000000 - momentum: 0.000000 2023-10-17 17:30:05,785 epoch 10 - iter 290/292 - loss 0.01144895 - time (sec): 16.87 - samples/sec: 2623.09 - lr: 0.000000 - momentum: 0.000000 2023-10-17 17:30:05,893 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:30:05,893 EPOCH 10 done: loss 0.0114 - lr: 0.000000 2023-10-17 17:30:07,136 DEV : loss 0.17449024319648743 - f1-score (micro avg) 0.7387 2023-10-17 17:30:07,536 ---------------------------------------------------------------------------------------------------- 2023-10-17 17:30:07,537 Loading model from best epoch ... 2023-10-17 17:30:09,536 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 2023-10-17 17:30:12,032 Results: - F-score (micro) 0.7261 - F-score (macro) 0.6066 - Accuracy 0.5897 By class: precision recall f1-score support PER 0.8437 0.8218 0.8326 348 LOC 0.6073 0.7701 0.6791 261 ORG 0.3077 0.2308 0.2637 52 HumanProd 0.6667 0.6364 0.6512 22 micro avg 0.7027 0.7511 0.7261 683 macro avg 0.6063 0.6148 0.6066 683 weighted avg 0.7068 0.7511 0.7248 683 2023-10-17 17:30:12,032 ----------------------------------------------------------------------------------------------------