2023-10-17 18:15:25,593 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:15:25,594 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 18:15:25,594 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:15:25,594 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 18:15:25,594 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:15:25,594 Train: 1166 sentences 2023-10-17 18:15:25,594 (train_with_dev=False, train_with_test=False) 2023-10-17 18:15:25,594 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:15:25,594 Training Params: 2023-10-17 18:15:25,594 - learning_rate: "3e-05" 2023-10-17 18:15:25,594 - mini_batch_size: "8" 2023-10-17 18:15:25,594 - max_epochs: "10" 2023-10-17 18:15:25,594 - shuffle: "True" 2023-10-17 18:15:25,594 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:15:25,595 Plugins: 2023-10-17 18:15:25,595 - TensorboardLogger 2023-10-17 18:15:25,595 - LinearScheduler | warmup_fraction: '0.1' 2023-10-17 18:15:25,595 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:15:25,595 Final evaluation on model from best epoch (best-model.pt) 2023-10-17 18:15:25,595 - metric: "('micro avg', 'f1-score')" 2023-10-17 18:15:25,595 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:15:25,595 Computation: 2023-10-17 18:15:25,595 - compute on device: cuda:0 2023-10-17 18:15:25,595 - embedding storage: none 2023-10-17 18:15:25,595 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:15:25,595 Model training base path: "hmbench-newseye/fi-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4" 2023-10-17 18:15:25,595 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:15:25,595 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:15:25,595 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-17 18:15:27,065 epoch 1 - iter 14/146 - loss 3.61689160 - time (sec): 1.47 - samples/sec: 3134.32 - lr: 0.000003 - momentum: 0.000000 2023-10-17 18:15:28,697 epoch 1 - iter 28/146 - loss 3.45583081 - time (sec): 3.10 - samples/sec: 2937.46 - lr: 0.000006 - momentum: 0.000000 2023-10-17 18:15:30,233 epoch 1 - iter 42/146 - loss 3.03740792 - time (sec): 4.64 - samples/sec: 2855.93 - lr: 0.000008 - momentum: 0.000000 2023-10-17 18:15:31,892 epoch 1 - iter 56/146 - loss 2.50761450 - time (sec): 6.30 - samples/sec: 2892.85 - lr: 0.000011 - momentum: 0.000000 2023-10-17 18:15:33,437 epoch 1 - iter 70/146 - loss 2.08084143 - time (sec): 7.84 - samples/sec: 2944.69 - lr: 0.000014 - momentum: 0.000000 2023-10-17 18:15:34,911 epoch 1 - iter 84/146 - loss 1.83799606 - time (sec): 9.32 - samples/sec: 2926.52 - lr: 0.000017 - momentum: 0.000000 2023-10-17 18:15:36,193 epoch 1 - iter 98/146 - loss 1.68258983 - time (sec): 10.60 - samples/sec: 2958.89 - lr: 0.000020 - momentum: 0.000000 2023-10-17 18:15:37,405 epoch 1 - iter 112/146 - loss 1.54466778 - time (sec): 11.81 - samples/sec: 2975.35 - lr: 0.000023 - momentum: 0.000000 2023-10-17 18:15:38,570 epoch 1 - iter 126/146 - loss 1.43607685 - time (sec): 12.97 - samples/sec: 2973.29 - lr: 0.000026 - momentum: 0.000000 2023-10-17 18:15:40,140 epoch 1 - iter 140/146 - loss 1.32061406 - time (sec): 14.54 - samples/sec: 2952.12 - lr: 0.000029 - momentum: 0.000000 2023-10-17 18:15:40,645 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:15:40,646 EPOCH 1 done: loss 1.2846 - lr: 0.000029 2023-10-17 18:15:41,637 DEV : loss 0.19552753865718842 - f1-score (micro avg) 0.4402 2023-10-17 18:15:41,642 saving best model 2023-10-17 18:15:41,975 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:15:43,448 epoch 2 - iter 14/146 - loss 0.25210580 - time (sec): 1.47 - samples/sec: 3201.40 - lr: 0.000030 - momentum: 0.000000 2023-10-17 18:15:44,863 epoch 2 - iter 28/146 - loss 0.23091153 - time (sec): 2.89 - samples/sec: 3036.34 - lr: 0.000029 - momentum: 0.000000 2023-10-17 18:15:46,028 epoch 2 - iter 42/146 - loss 0.25363766 - time (sec): 4.05 - samples/sec: 3121.72 - lr: 0.000029 - momentum: 0.000000 2023-10-17 18:15:47,571 epoch 2 - iter 56/146 - loss 0.25203999 - time (sec): 5.60 - samples/sec: 3007.28 - lr: 0.000029 - momentum: 0.000000 2023-10-17 18:15:49,098 epoch 2 - iter 70/146 - loss 0.23974587 - time (sec): 7.12 - samples/sec: 3000.34 - lr: 0.000028 - momentum: 0.000000 2023-10-17 18:15:50,767 epoch 2 - iter 84/146 - loss 0.23286681 - time (sec): 8.79 - samples/sec: 2998.47 - lr: 0.000028 - momentum: 0.000000 2023-10-17 18:15:52,419 epoch 2 - iter 98/146 - loss 0.21984282 - time (sec): 10.44 - samples/sec: 3007.83 - lr: 0.000028 - momentum: 0.000000 2023-10-17 18:15:53,712 epoch 2 - iter 112/146 - loss 0.22405276 - time (sec): 11.74 - samples/sec: 3031.68 - lr: 0.000027 - momentum: 0.000000 2023-10-17 18:15:55,192 epoch 2 - iter 126/146 - loss 0.21982674 - time (sec): 13.22 - samples/sec: 2965.80 - lr: 0.000027 - momentum: 0.000000 2023-10-17 18:15:56,565 epoch 2 - iter 140/146 - loss 0.21373885 - time (sec): 14.59 - samples/sec: 2968.47 - lr: 0.000027 - momentum: 0.000000 2023-10-17 18:15:56,974 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:15:56,975 EPOCH 2 done: loss 0.2122 - lr: 0.000027 2023-10-17 18:15:58,208 DEV : loss 0.13967962563037872 - f1-score (micro avg) 0.6198 2023-10-17 18:15:58,215 saving best model 2023-10-17 18:15:58,640 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:16:00,122 epoch 3 - iter 14/146 - loss 0.11131063 - time (sec): 1.47 - samples/sec: 2930.17 - lr: 0.000026 - momentum: 0.000000 2023-10-17 18:16:01,582 epoch 3 - iter 28/146 - loss 0.13050366 - time (sec): 2.93 - samples/sec: 3052.87 - lr: 0.000026 - momentum: 0.000000 2023-10-17 18:16:02,892 epoch 3 - iter 42/146 - loss 0.12162901 - time (sec): 4.24 - samples/sec: 2912.72 - lr: 0.000026 - momentum: 0.000000 2023-10-17 18:16:04,557 epoch 3 - iter 56/146 - loss 0.11964735 - time (sec): 5.91 - samples/sec: 2946.82 - lr: 0.000025 - momentum: 0.000000 2023-10-17 18:16:06,006 epoch 3 - iter 70/146 - loss 0.11783808 - time (sec): 7.36 - samples/sec: 2983.56 - lr: 0.000025 - momentum: 0.000000 2023-10-17 18:16:07,147 epoch 3 - iter 84/146 - loss 0.11516116 - time (sec): 8.50 - samples/sec: 2982.07 - lr: 0.000025 - momentum: 0.000000 2023-10-17 18:16:08,617 epoch 3 - iter 98/146 - loss 0.11251443 - time (sec): 9.97 - samples/sec: 2996.69 - lr: 0.000024 - momentum: 0.000000 2023-10-17 18:16:10,335 epoch 3 - iter 112/146 - loss 0.12177305 - time (sec): 11.69 - samples/sec: 2921.42 - lr: 0.000024 - momentum: 0.000000 2023-10-17 18:16:11,812 epoch 3 - iter 126/146 - loss 0.12094707 - time (sec): 13.17 - samples/sec: 2920.99 - lr: 0.000024 - momentum: 0.000000 2023-10-17 18:16:13,234 epoch 3 - iter 140/146 - loss 0.12211446 - time (sec): 14.59 - samples/sec: 2919.44 - lr: 0.000024 - momentum: 0.000000 2023-10-17 18:16:13,860 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:16:13,860 EPOCH 3 done: loss 0.1230 - lr: 0.000024 2023-10-17 18:16:15,099 DEV : loss 0.12319271266460419 - f1-score (micro avg) 0.7021 2023-10-17 18:16:15,103 saving best model 2023-10-17 18:16:15,516 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:16:16,910 epoch 4 - iter 14/146 - loss 0.07446918 - time (sec): 1.39 - samples/sec: 3126.83 - lr: 0.000023 - momentum: 0.000000 2023-10-17 18:16:18,406 epoch 4 - iter 28/146 - loss 0.07544335 - time (sec): 2.88 - samples/sec: 3098.50 - lr: 0.000023 - momentum: 0.000000 2023-10-17 18:16:19,573 epoch 4 - iter 42/146 - loss 0.07942405 - time (sec): 4.05 - samples/sec: 3153.97 - lr: 0.000022 - momentum: 0.000000 2023-10-17 18:16:21,016 epoch 4 - iter 56/146 - loss 0.08085290 - time (sec): 5.49 - samples/sec: 3103.44 - lr: 0.000022 - momentum: 0.000000 2023-10-17 18:16:22,472 epoch 4 - iter 70/146 - loss 0.07764363 - time (sec): 6.95 - samples/sec: 3028.09 - lr: 0.000022 - momentum: 0.000000 2023-10-17 18:16:23,964 epoch 4 - iter 84/146 - loss 0.07552423 - time (sec): 8.44 - samples/sec: 2977.51 - lr: 0.000021 - momentum: 0.000000 2023-10-17 18:16:25,675 epoch 4 - iter 98/146 - loss 0.08142206 - time (sec): 10.15 - samples/sec: 2955.13 - lr: 0.000021 - momentum: 0.000000 2023-10-17 18:16:27,189 epoch 4 - iter 112/146 - loss 0.08418781 - time (sec): 11.67 - samples/sec: 2965.89 - lr: 0.000021 - momentum: 0.000000 2023-10-17 18:16:28,598 epoch 4 - iter 126/146 - loss 0.08204856 - time (sec): 13.08 - samples/sec: 2939.69 - lr: 0.000021 - momentum: 0.000000 2023-10-17 18:16:30,075 epoch 4 - iter 140/146 - loss 0.08171260 - time (sec): 14.55 - samples/sec: 2942.50 - lr: 0.000020 - momentum: 0.000000 2023-10-17 18:16:30,568 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:16:30,568 EPOCH 4 done: loss 0.0810 - lr: 0.000020 2023-10-17 18:16:31,869 DEV : loss 0.11123495548963547 - f1-score (micro avg) 0.7265 2023-10-17 18:16:31,874 saving best model 2023-10-17 18:16:32,310 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:16:33,707 epoch 5 - iter 14/146 - loss 0.07595815 - time (sec): 1.40 - samples/sec: 2719.13 - lr: 0.000020 - momentum: 0.000000 2023-10-17 18:16:35,138 epoch 5 - iter 28/146 - loss 0.06655941 - time (sec): 2.83 - samples/sec: 2843.09 - lr: 0.000019 - momentum: 0.000000 2023-10-17 18:16:36,720 epoch 5 - iter 42/146 - loss 0.05683304 - time (sec): 4.41 - samples/sec: 2898.57 - lr: 0.000019 - momentum: 0.000000 2023-10-17 18:16:37,969 epoch 5 - iter 56/146 - loss 0.04997737 - time (sec): 5.66 - samples/sec: 2912.50 - lr: 0.000019 - momentum: 0.000000 2023-10-17 18:16:39,377 epoch 5 - iter 70/146 - loss 0.05412388 - time (sec): 7.07 - samples/sec: 2958.66 - lr: 0.000018 - momentum: 0.000000 2023-10-17 18:16:40,880 epoch 5 - iter 84/146 - loss 0.05953456 - time (sec): 8.57 - samples/sec: 2958.21 - lr: 0.000018 - momentum: 0.000000 2023-10-17 18:16:42,379 epoch 5 - iter 98/146 - loss 0.05702498 - time (sec): 10.07 - samples/sec: 2960.03 - lr: 0.000018 - momentum: 0.000000 2023-10-17 18:16:43,443 epoch 5 - iter 112/146 - loss 0.05850990 - time (sec): 11.13 - samples/sec: 2970.86 - lr: 0.000018 - momentum: 0.000000 2023-10-17 18:16:44,906 epoch 5 - iter 126/146 - loss 0.05714029 - time (sec): 12.59 - samples/sec: 2983.70 - lr: 0.000017 - momentum: 0.000000 2023-10-17 18:16:46,694 epoch 5 - iter 140/146 - loss 0.05542780 - time (sec): 14.38 - samples/sec: 2963.74 - lr: 0.000017 - momentum: 0.000000 2023-10-17 18:16:47,334 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:16:47,334 EPOCH 5 done: loss 0.0550 - lr: 0.000017 2023-10-17 18:16:48,612 DEV : loss 0.11169999092817307 - f1-score (micro avg) 0.754 2023-10-17 18:16:48,618 saving best model 2023-10-17 18:16:49,041 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:16:50,331 epoch 6 - iter 14/146 - loss 0.03957065 - time (sec): 1.28 - samples/sec: 2984.68 - lr: 0.000016 - momentum: 0.000000 2023-10-17 18:16:51,800 epoch 6 - iter 28/146 - loss 0.04961863 - time (sec): 2.75 - samples/sec: 2888.36 - lr: 0.000016 - momentum: 0.000000 2023-10-17 18:16:53,397 epoch 6 - iter 42/146 - loss 0.04313213 - time (sec): 4.35 - samples/sec: 2896.92 - lr: 0.000016 - momentum: 0.000000 2023-10-17 18:16:54,981 epoch 6 - iter 56/146 - loss 0.04127585 - time (sec): 5.93 - samples/sec: 2880.26 - lr: 0.000015 - momentum: 0.000000 2023-10-17 18:16:56,440 epoch 6 - iter 70/146 - loss 0.04424832 - time (sec): 7.39 - samples/sec: 2849.49 - lr: 0.000015 - momentum: 0.000000 2023-10-17 18:16:57,963 epoch 6 - iter 84/146 - loss 0.04266501 - time (sec): 8.92 - samples/sec: 2812.03 - lr: 0.000015 - momentum: 0.000000 2023-10-17 18:16:59,396 epoch 6 - iter 98/146 - loss 0.04351976 - time (sec): 10.35 - samples/sec: 2840.40 - lr: 0.000015 - momentum: 0.000000 2023-10-17 18:17:01,045 epoch 6 - iter 112/146 - loss 0.04419179 - time (sec): 12.00 - samples/sec: 2870.34 - lr: 0.000014 - momentum: 0.000000 2023-10-17 18:17:02,529 epoch 6 - iter 126/146 - loss 0.04318850 - time (sec): 13.48 - samples/sec: 2885.05 - lr: 0.000014 - momentum: 0.000000 2023-10-17 18:17:03,791 epoch 6 - iter 140/146 - loss 0.04286066 - time (sec): 14.74 - samples/sec: 2902.70 - lr: 0.000014 - momentum: 0.000000 2023-10-17 18:17:04,362 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:17:04,362 EPOCH 6 done: loss 0.0422 - lr: 0.000014 2023-10-17 18:17:05,596 DEV : loss 0.11784238368272781 - f1-score (micro avg) 0.7661 2023-10-17 18:17:05,601 saving best model 2023-10-17 18:17:06,046 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:17:07,424 epoch 7 - iter 14/146 - loss 0.03385022 - time (sec): 1.37 - samples/sec: 2927.23 - lr: 0.000013 - momentum: 0.000000 2023-10-17 18:17:08,837 epoch 7 - iter 28/146 - loss 0.03372986 - time (sec): 2.79 - samples/sec: 3052.80 - lr: 0.000013 - momentum: 0.000000 2023-10-17 18:17:10,292 epoch 7 - iter 42/146 - loss 0.03560445 - time (sec): 4.24 - samples/sec: 2957.91 - lr: 0.000012 - momentum: 0.000000 2023-10-17 18:17:11,474 epoch 7 - iter 56/146 - loss 0.03370650 - time (sec): 5.42 - samples/sec: 2950.45 - lr: 0.000012 - momentum: 0.000000 2023-10-17 18:17:13,086 epoch 7 - iter 70/146 - loss 0.03515693 - time (sec): 7.03 - samples/sec: 2910.61 - lr: 0.000012 - momentum: 0.000000 2023-10-17 18:17:14,681 epoch 7 - iter 84/146 - loss 0.03267296 - time (sec): 8.63 - samples/sec: 2930.16 - lr: 0.000012 - momentum: 0.000000 2023-10-17 18:17:16,030 epoch 7 - iter 98/146 - loss 0.03209009 - time (sec): 9.98 - samples/sec: 2945.41 - lr: 0.000011 - momentum: 0.000000 2023-10-17 18:17:17,833 epoch 7 - iter 112/146 - loss 0.03060223 - time (sec): 11.78 - samples/sec: 2915.36 - lr: 0.000011 - momentum: 0.000000 2023-10-17 18:17:19,181 epoch 7 - iter 126/146 - loss 0.03048273 - time (sec): 13.13 - samples/sec: 2939.36 - lr: 0.000011 - momentum: 0.000000 2023-10-17 18:17:20,728 epoch 7 - iter 140/146 - loss 0.03009423 - time (sec): 14.68 - samples/sec: 2922.40 - lr: 0.000010 - momentum: 0.000000 2023-10-17 18:17:21,244 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:17:21,244 EPOCH 7 done: loss 0.0305 - lr: 0.000010 2023-10-17 18:17:22,487 DEV : loss 0.12970943748950958 - f1-score (micro avg) 0.7652 2023-10-17 18:17:22,491 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:17:24,101 epoch 8 - iter 14/146 - loss 0.03967823 - time (sec): 1.61 - samples/sec: 2808.92 - lr: 0.000010 - momentum: 0.000000 2023-10-17 18:17:25,640 epoch 8 - iter 28/146 - loss 0.04068898 - time (sec): 3.15 - samples/sec: 2938.80 - lr: 0.000009 - momentum: 0.000000 2023-10-17 18:17:27,175 epoch 8 - iter 42/146 - loss 0.03075275 - time (sec): 4.68 - samples/sec: 2926.50 - lr: 0.000009 - momentum: 0.000000 2023-10-17 18:17:28,734 epoch 8 - iter 56/146 - loss 0.02745463 - time (sec): 6.24 - samples/sec: 2938.76 - lr: 0.000009 - momentum: 0.000000 2023-10-17 18:17:30,104 epoch 8 - iter 70/146 - loss 0.02452288 - time (sec): 7.61 - samples/sec: 2947.94 - lr: 0.000009 - momentum: 0.000000 2023-10-17 18:17:31,279 epoch 8 - iter 84/146 - loss 0.02648615 - time (sec): 8.79 - samples/sec: 2979.13 - lr: 0.000008 - momentum: 0.000000 2023-10-17 18:17:32,802 epoch 8 - iter 98/146 - loss 0.02578929 - time (sec): 10.31 - samples/sec: 2995.43 - lr: 0.000008 - momentum: 0.000000 2023-10-17 18:17:34,106 epoch 8 - iter 112/146 - loss 0.02578146 - time (sec): 11.61 - samples/sec: 2997.10 - lr: 0.000008 - momentum: 0.000000 2023-10-17 18:17:35,181 epoch 8 - iter 126/146 - loss 0.02544646 - time (sec): 12.69 - samples/sec: 2999.56 - lr: 0.000007 - momentum: 0.000000 2023-10-17 18:17:36,877 epoch 8 - iter 140/146 - loss 0.02347594 - time (sec): 14.38 - samples/sec: 2964.33 - lr: 0.000007 - momentum: 0.000000 2023-10-17 18:17:37,393 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:17:37,393 EPOCH 8 done: loss 0.0231 - lr: 0.000007 2023-10-17 18:17:38,624 DEV : loss 0.12909089028835297 - f1-score (micro avg) 0.7606 2023-10-17 18:17:38,629 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:17:40,113 epoch 9 - iter 14/146 - loss 0.01610542 - time (sec): 1.48 - samples/sec: 2717.22 - lr: 0.000006 - momentum: 0.000000 2023-10-17 18:17:41,485 epoch 9 - iter 28/146 - loss 0.02338741 - time (sec): 2.86 - samples/sec: 3014.68 - lr: 0.000006 - momentum: 0.000000 2023-10-17 18:17:43,099 epoch 9 - iter 42/146 - loss 0.01898118 - time (sec): 4.47 - samples/sec: 2995.68 - lr: 0.000006 - momentum: 0.000000 2023-10-17 18:17:44,530 epoch 9 - iter 56/146 - loss 0.01556164 - time (sec): 5.90 - samples/sec: 2986.01 - lr: 0.000006 - momentum: 0.000000 2023-10-17 18:17:46,025 epoch 9 - iter 70/146 - loss 0.01504665 - time (sec): 7.39 - samples/sec: 2968.23 - lr: 0.000005 - momentum: 0.000000 2023-10-17 18:17:47,261 epoch 9 - iter 84/146 - loss 0.01595822 - time (sec): 8.63 - samples/sec: 2956.70 - lr: 0.000005 - momentum: 0.000000 2023-10-17 18:17:48,678 epoch 9 - iter 98/146 - loss 0.01550102 - time (sec): 10.05 - samples/sec: 2941.07 - lr: 0.000005 - momentum: 0.000000 2023-10-17 18:17:50,319 epoch 9 - iter 112/146 - loss 0.01652017 - time (sec): 11.69 - samples/sec: 2918.00 - lr: 0.000004 - momentum: 0.000000 2023-10-17 18:17:51,698 epoch 9 - iter 126/146 - loss 0.01700961 - time (sec): 13.07 - samples/sec: 2927.73 - lr: 0.000004 - momentum: 0.000000 2023-10-17 18:17:53,503 epoch 9 - iter 140/146 - loss 0.01918912 - time (sec): 14.87 - samples/sec: 2894.91 - lr: 0.000004 - momentum: 0.000000 2023-10-17 18:17:54,135 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:17:54,136 EPOCH 9 done: loss 0.0195 - lr: 0.000004 2023-10-17 18:17:55,386 DEV : loss 0.129518061876297 - f1-score (micro avg) 0.7489 2023-10-17 18:17:55,392 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:17:56,839 epoch 10 - iter 14/146 - loss 0.01275460 - time (sec): 1.45 - samples/sec: 2691.79 - lr: 0.000003 - momentum: 0.000000 2023-10-17 18:17:58,169 epoch 10 - iter 28/146 - loss 0.00987069 - time (sec): 2.78 - samples/sec: 2816.79 - lr: 0.000003 - momentum: 0.000000 2023-10-17 18:17:59,720 epoch 10 - iter 42/146 - loss 0.00991440 - time (sec): 4.33 - samples/sec: 2815.40 - lr: 0.000003 - momentum: 0.000000 2023-10-17 18:18:01,134 epoch 10 - iter 56/146 - loss 0.00988055 - time (sec): 5.74 - samples/sec: 2865.60 - lr: 0.000002 - momentum: 0.000000 2023-10-17 18:18:02,734 epoch 10 - iter 70/146 - loss 0.01038453 - time (sec): 7.34 - samples/sec: 2965.40 - lr: 0.000002 - momentum: 0.000000 2023-10-17 18:18:04,048 epoch 10 - iter 84/146 - loss 0.01235361 - time (sec): 8.66 - samples/sec: 2968.13 - lr: 0.000002 - momentum: 0.000000 2023-10-17 18:18:05,765 epoch 10 - iter 98/146 - loss 0.01293458 - time (sec): 10.37 - samples/sec: 2910.71 - lr: 0.000001 - momentum: 0.000000 2023-10-17 18:18:07,131 epoch 10 - iter 112/146 - loss 0.01413777 - time (sec): 11.74 - samples/sec: 2923.45 - lr: 0.000001 - momentum: 0.000000 2023-10-17 18:18:08,559 epoch 10 - iter 126/146 - loss 0.01605264 - time (sec): 13.17 - samples/sec: 2882.86 - lr: 0.000001 - momentum: 0.000000 2023-10-17 18:18:09,996 epoch 10 - iter 140/146 - loss 0.01671795 - time (sec): 14.60 - samples/sec: 2927.32 - lr: 0.000000 - momentum: 0.000000 2023-10-17 18:18:10,578 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:18:10,578 EPOCH 10 done: loss 0.0165 - lr: 0.000000 2023-10-17 18:18:12,075 DEV : loss 0.13104750216007233 - f1-score (micro avg) 0.7522 2023-10-17 18:18:12,423 ---------------------------------------------------------------------------------------------------- 2023-10-17 18:18:12,425 Loading model from best epoch ... 2023-10-17 18:18:13,828 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 18:18:16,268 Results: - F-score (micro) 0.7599 - F-score (macro) 0.6534 - Accuracy 0.6298 By class: precision recall f1-score support PER 0.8110 0.8506 0.8303 348 LOC 0.6879 0.8276 0.7513 261 ORG 0.3654 0.3654 0.3654 52 HumanProd 0.6522 0.6818 0.6667 22 micro avg 0.7241 0.7994 0.7599 683 macro avg 0.6291 0.6813 0.6534 683 weighted avg 0.7249 0.7994 0.7594 683 2023-10-17 18:18:16,268 ----------------------------------------------------------------------------------------------------