2023-10-16 18:44:14,992 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:44:14,993 Model: "SequenceTagger( (embeddings): TransformerWordEmbeddings( (model): BertModel( (embeddings): BertEmbeddings( (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): BertEncoder( (layer): ModuleList( (0-11): 12 x BertLayer( (attention): BertAttention( (self): BertSelfAttention( (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): BertSelfOutput( (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): BertIntermediate( (dense): Linear(in_features=768, out_features=3072, bias=True) (intermediate_act_fn): GELUActivation() ) (output): BertOutput( (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) ) ) ) ) (pooler): BertPooler( (dense): Linear(in_features=768, out_features=768, bias=True) (activation): Tanh() ) ) ) (locked_dropout): LockedDropout(p=0.5) (linear): Linear(in_features=768, out_features=17, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-16 18:44:14,993 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:44:14,993 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 2023-10-16 18:44:14,993 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:44:14,993 Train: 1166 sentences 2023-10-16 18:44:14,993 (train_with_dev=False, train_with_test=False) 2023-10-16 18:44:14,993 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:44:14,993 Training Params: 2023-10-16 18:44:14,993 - learning_rate: "5e-05" 2023-10-16 18:44:14,993 - mini_batch_size: "8" 2023-10-16 18:44:14,993 - max_epochs: "10" 2023-10-16 18:44:14,993 - shuffle: "True" 2023-10-16 18:44:14,994 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:44:14,994 Plugins: 2023-10-16 18:44:14,994 - LinearScheduler | warmup_fraction: '0.1' 2023-10-16 18:44:14,994 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:44:14,994 Final evaluation on model from best epoch (best-model.pt) 2023-10-16 18:44:14,994 - metric: "('micro avg', 'f1-score')" 2023-10-16 18:44:14,994 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:44:14,994 Computation: 2023-10-16 18:44:14,994 - compute on device: cuda:0 2023-10-16 18:44:14,994 - embedding storage: none 2023-10-16 18:44:14,994 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:44:14,994 Model training base path: "hmbench-newseye/fi-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4" 2023-10-16 18:44:14,994 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:44:14,994 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:44:16,450 epoch 1 - iter 14/146 - loss 2.93554078 - time (sec): 1.46 - samples/sec: 3031.39 - lr: 0.000004 - momentum: 0.000000 2023-10-16 18:44:17,598 epoch 1 - iter 28/146 - loss 2.59641910 - time (sec): 2.60 - samples/sec: 3093.50 - lr: 0.000009 - momentum: 0.000000 2023-10-16 18:44:19,197 epoch 1 - iter 42/146 - loss 1.98647565 - time (sec): 4.20 - samples/sec: 3046.18 - lr: 0.000014 - momentum: 0.000000 2023-10-16 18:44:20,958 epoch 1 - iter 56/146 - loss 1.65493251 - time (sec): 5.96 - samples/sec: 2908.07 - lr: 0.000019 - momentum: 0.000000 2023-10-16 18:44:22,265 epoch 1 - iter 70/146 - loss 1.48549199 - time (sec): 7.27 - samples/sec: 2906.30 - lr: 0.000024 - momentum: 0.000000 2023-10-16 18:44:23,957 epoch 1 - iter 84/146 - loss 1.33028113 - time (sec): 8.96 - samples/sec: 2878.44 - lr: 0.000028 - momentum: 0.000000 2023-10-16 18:44:25,337 epoch 1 - iter 98/146 - loss 1.19387586 - time (sec): 10.34 - samples/sec: 2907.41 - lr: 0.000033 - momentum: 0.000000 2023-10-16 18:44:26,724 epoch 1 - iter 112/146 - loss 1.07702304 - time (sec): 11.73 - samples/sec: 2939.16 - lr: 0.000038 - momentum: 0.000000 2023-10-16 18:44:28,150 epoch 1 - iter 126/146 - loss 0.98680341 - time (sec): 13.16 - samples/sec: 2960.74 - lr: 0.000043 - momentum: 0.000000 2023-10-16 18:44:29,393 epoch 1 - iter 140/146 - loss 0.91930004 - time (sec): 14.40 - samples/sec: 2986.90 - lr: 0.000048 - momentum: 0.000000 2023-10-16 18:44:29,886 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:44:29,887 EPOCH 1 done: loss 0.8989 - lr: 0.000048 2023-10-16 18:44:30,915 DEV : loss 0.19385209679603577 - f1-score (micro avg) 0.4388 2023-10-16 18:44:30,919 saving best model 2023-10-16 18:44:31,329 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:44:32,656 epoch 2 - iter 14/146 - loss 0.26646816 - time (sec): 1.33 - samples/sec: 3202.19 - lr: 0.000050 - momentum: 0.000000 2023-10-16 18:44:34,281 epoch 2 - iter 28/146 - loss 0.27245130 - time (sec): 2.95 - samples/sec: 3188.14 - lr: 0.000049 - momentum: 0.000000 2023-10-16 18:44:35,932 epoch 2 - iter 42/146 - loss 0.28481191 - time (sec): 4.60 - samples/sec: 2962.86 - lr: 0.000048 - momentum: 0.000000 2023-10-16 18:44:37,632 epoch 2 - iter 56/146 - loss 0.25721835 - time (sec): 6.30 - samples/sec: 2913.56 - lr: 0.000048 - momentum: 0.000000 2023-10-16 18:44:38,924 epoch 2 - iter 70/146 - loss 0.24376985 - time (sec): 7.59 - samples/sec: 2929.54 - lr: 0.000047 - momentum: 0.000000 2023-10-16 18:44:40,297 epoch 2 - iter 84/146 - loss 0.23798723 - time (sec): 8.97 - samples/sec: 2952.15 - lr: 0.000047 - momentum: 0.000000 2023-10-16 18:44:41,482 epoch 2 - iter 98/146 - loss 0.23172878 - time (sec): 10.15 - samples/sec: 2977.39 - lr: 0.000046 - momentum: 0.000000 2023-10-16 18:44:42,768 epoch 2 - iter 112/146 - loss 0.21818927 - time (sec): 11.44 - samples/sec: 3000.61 - lr: 0.000046 - momentum: 0.000000 2023-10-16 18:44:44,349 epoch 2 - iter 126/146 - loss 0.21149100 - time (sec): 13.02 - samples/sec: 2974.30 - lr: 0.000045 - momentum: 0.000000 2023-10-16 18:44:45,592 epoch 2 - iter 140/146 - loss 0.20682752 - time (sec): 14.26 - samples/sec: 2983.62 - lr: 0.000045 - momentum: 0.000000 2023-10-16 18:44:46,258 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:44:46,259 EPOCH 2 done: loss 0.2017 - lr: 0.000045 2023-10-16 18:44:47,492 DEV : loss 0.11957067251205444 - f1-score (micro avg) 0.6291 2023-10-16 18:44:47,503 saving best model 2023-10-16 18:44:47,979 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:44:49,571 epoch 3 - iter 14/146 - loss 0.11191478 - time (sec): 1.58 - samples/sec: 3023.30 - lr: 0.000044 - momentum: 0.000000 2023-10-16 18:44:51,442 epoch 3 - iter 28/146 - loss 0.11158050 - time (sec): 3.45 - samples/sec: 2823.35 - lr: 0.000043 - momentum: 0.000000 2023-10-16 18:44:52,817 epoch 3 - iter 42/146 - loss 0.11844144 - time (sec): 4.83 - samples/sec: 2769.61 - lr: 0.000043 - momentum: 0.000000 2023-10-16 18:44:53,989 epoch 3 - iter 56/146 - loss 0.11296688 - time (sec): 6.00 - samples/sec: 2807.39 - lr: 0.000042 - momentum: 0.000000 2023-10-16 18:44:55,247 epoch 3 - iter 70/146 - loss 0.11117130 - time (sec): 7.26 - samples/sec: 2862.75 - lr: 0.000042 - momentum: 0.000000 2023-10-16 18:44:56,780 epoch 3 - iter 84/146 - loss 0.10769328 - time (sec): 8.79 - samples/sec: 2918.14 - lr: 0.000041 - momentum: 0.000000 2023-10-16 18:44:58,413 epoch 3 - iter 98/146 - loss 0.10822287 - time (sec): 10.42 - samples/sec: 2956.84 - lr: 0.000041 - momentum: 0.000000 2023-10-16 18:44:59,893 epoch 3 - iter 112/146 - loss 0.10974325 - time (sec): 11.90 - samples/sec: 2946.24 - lr: 0.000040 - momentum: 0.000000 2023-10-16 18:45:01,233 epoch 3 - iter 126/146 - loss 0.10961191 - time (sec): 13.24 - samples/sec: 2948.19 - lr: 0.000040 - momentum: 0.000000 2023-10-16 18:45:02,762 epoch 3 - iter 140/146 - loss 0.11159692 - time (sec): 14.77 - samples/sec: 2908.76 - lr: 0.000039 - momentum: 0.000000 2023-10-16 18:45:03,277 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:45:03,277 EPOCH 3 done: loss 0.1116 - lr: 0.000039 2023-10-16 18:45:04,525 DEV : loss 0.10484768450260162 - f1-score (micro avg) 0.7265 2023-10-16 18:45:04,529 saving best model 2023-10-16 18:45:05,074 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:45:06,751 epoch 4 - iter 14/146 - loss 0.06773359 - time (sec): 1.67 - samples/sec: 3019.84 - lr: 0.000038 - momentum: 0.000000 2023-10-16 18:45:08,400 epoch 4 - iter 28/146 - loss 0.08808391 - time (sec): 3.32 - samples/sec: 2840.89 - lr: 0.000038 - momentum: 0.000000 2023-10-16 18:45:09,895 epoch 4 - iter 42/146 - loss 0.07321162 - time (sec): 4.82 - samples/sec: 2896.89 - lr: 0.000037 - momentum: 0.000000 2023-10-16 18:45:11,044 epoch 4 - iter 56/146 - loss 0.07713816 - time (sec): 5.97 - samples/sec: 2905.45 - lr: 0.000037 - momentum: 0.000000 2023-10-16 18:45:12,599 epoch 4 - iter 70/146 - loss 0.07754016 - time (sec): 7.52 - samples/sec: 2913.73 - lr: 0.000036 - momentum: 0.000000 2023-10-16 18:45:14,247 epoch 4 - iter 84/146 - loss 0.07736573 - time (sec): 9.17 - samples/sec: 2861.76 - lr: 0.000036 - momentum: 0.000000 2023-10-16 18:45:15,521 epoch 4 - iter 98/146 - loss 0.07596628 - time (sec): 10.44 - samples/sec: 2860.25 - lr: 0.000035 - momentum: 0.000000 2023-10-16 18:45:17,303 epoch 4 - iter 112/146 - loss 0.07335282 - time (sec): 12.23 - samples/sec: 2866.41 - lr: 0.000035 - momentum: 0.000000 2023-10-16 18:45:18,600 epoch 4 - iter 126/146 - loss 0.07440373 - time (sec): 13.52 - samples/sec: 2899.44 - lr: 0.000034 - momentum: 0.000000 2023-10-16 18:45:19,917 epoch 4 - iter 140/146 - loss 0.07551967 - time (sec): 14.84 - samples/sec: 2900.64 - lr: 0.000034 - momentum: 0.000000 2023-10-16 18:45:20,359 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:45:20,359 EPOCH 4 done: loss 0.0754 - lr: 0.000034 2023-10-16 18:45:21,637 DEV : loss 0.106049545109272 - f1-score (micro avg) 0.7489 2023-10-16 18:45:21,642 saving best model 2023-10-16 18:45:22,190 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:45:23,719 epoch 5 - iter 14/146 - loss 0.04848651 - time (sec): 1.53 - samples/sec: 2833.76 - lr: 0.000033 - momentum: 0.000000 2023-10-16 18:45:25,226 epoch 5 - iter 28/146 - loss 0.04053376 - time (sec): 3.03 - samples/sec: 2740.38 - lr: 0.000032 - momentum: 0.000000 2023-10-16 18:45:26,737 epoch 5 - iter 42/146 - loss 0.04042929 - time (sec): 4.54 - samples/sec: 2797.98 - lr: 0.000032 - momentum: 0.000000 2023-10-16 18:45:28,427 epoch 5 - iter 56/146 - loss 0.04755111 - time (sec): 6.23 - samples/sec: 2885.92 - lr: 0.000031 - momentum: 0.000000 2023-10-16 18:45:30,011 epoch 5 - iter 70/146 - loss 0.04653289 - time (sec): 7.82 - samples/sec: 2907.59 - lr: 0.000031 - momentum: 0.000000 2023-10-16 18:45:31,331 epoch 5 - iter 84/146 - loss 0.05035831 - time (sec): 9.14 - samples/sec: 2918.02 - lr: 0.000030 - momentum: 0.000000 2023-10-16 18:45:32,949 epoch 5 - iter 98/146 - loss 0.04699916 - time (sec): 10.76 - samples/sec: 2946.14 - lr: 0.000030 - momentum: 0.000000 2023-10-16 18:45:34,106 epoch 5 - iter 112/146 - loss 0.04895681 - time (sec): 11.91 - samples/sec: 2962.47 - lr: 0.000029 - momentum: 0.000000 2023-10-16 18:45:35,479 epoch 5 - iter 126/146 - loss 0.05041908 - time (sec): 13.29 - samples/sec: 2959.37 - lr: 0.000029 - momentum: 0.000000 2023-10-16 18:45:36,721 epoch 5 - iter 140/146 - loss 0.05285443 - time (sec): 14.53 - samples/sec: 2974.98 - lr: 0.000028 - momentum: 0.000000 2023-10-16 18:45:37,198 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:45:37,198 EPOCH 5 done: loss 0.0529 - lr: 0.000028 2023-10-16 18:45:38,518 DEV : loss 0.11109450459480286 - f1-score (micro avg) 0.7431 2023-10-16 18:45:38,525 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:45:39,890 epoch 6 - iter 14/146 - loss 0.05005263 - time (sec): 1.36 - samples/sec: 3099.52 - lr: 0.000027 - momentum: 0.000000 2023-10-16 18:45:41,470 epoch 6 - iter 28/146 - loss 0.04617285 - time (sec): 2.94 - samples/sec: 3109.63 - lr: 0.000027 - momentum: 0.000000 2023-10-16 18:45:42,781 epoch 6 - iter 42/146 - loss 0.03762395 - time (sec): 4.25 - samples/sec: 3087.76 - lr: 0.000026 - momentum: 0.000000 2023-10-16 18:45:44,214 epoch 6 - iter 56/146 - loss 0.03821664 - time (sec): 5.69 - samples/sec: 3009.65 - lr: 0.000026 - momentum: 0.000000 2023-10-16 18:45:45,976 epoch 6 - iter 70/146 - loss 0.03792029 - time (sec): 7.45 - samples/sec: 2884.01 - lr: 0.000025 - momentum: 0.000000 2023-10-16 18:45:47,431 epoch 6 - iter 84/146 - loss 0.03947725 - time (sec): 8.90 - samples/sec: 2851.63 - lr: 0.000025 - momentum: 0.000000 2023-10-16 18:45:48,916 epoch 6 - iter 98/146 - loss 0.03763630 - time (sec): 10.39 - samples/sec: 2885.48 - lr: 0.000024 - momentum: 0.000000 2023-10-16 18:45:50,323 epoch 6 - iter 112/146 - loss 0.03716618 - time (sec): 11.80 - samples/sec: 2911.43 - lr: 0.000024 - momentum: 0.000000 2023-10-16 18:45:51,783 epoch 6 - iter 126/146 - loss 0.03715143 - time (sec): 13.26 - samples/sec: 2934.88 - lr: 0.000023 - momentum: 0.000000 2023-10-16 18:45:53,097 epoch 6 - iter 140/146 - loss 0.03776802 - time (sec): 14.57 - samples/sec: 2928.64 - lr: 0.000023 - momentum: 0.000000 2023-10-16 18:45:53,735 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:45:53,735 EPOCH 6 done: loss 0.0377 - lr: 0.000023 2023-10-16 18:45:55,064 DEV : loss 0.10371122509241104 - f1-score (micro avg) 0.7702 2023-10-16 18:45:55,070 saving best model 2023-10-16 18:45:55,594 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:45:56,996 epoch 7 - iter 14/146 - loss 0.02717694 - time (sec): 1.40 - samples/sec: 3059.75 - lr: 0.000022 - momentum: 0.000000 2023-10-16 18:45:58,279 epoch 7 - iter 28/146 - loss 0.02170902 - time (sec): 2.68 - samples/sec: 3094.36 - lr: 0.000021 - momentum: 0.000000 2023-10-16 18:45:59,710 epoch 7 - iter 42/146 - loss 0.01935375 - time (sec): 4.11 - samples/sec: 3129.38 - lr: 0.000021 - momentum: 0.000000 2023-10-16 18:46:01,234 epoch 7 - iter 56/146 - loss 0.02181448 - time (sec): 5.64 - samples/sec: 3090.41 - lr: 0.000020 - momentum: 0.000000 2023-10-16 18:46:02,563 epoch 7 - iter 70/146 - loss 0.02092834 - time (sec): 6.97 - samples/sec: 3056.05 - lr: 0.000020 - momentum: 0.000000 2023-10-16 18:46:04,330 epoch 7 - iter 84/146 - loss 0.02507712 - time (sec): 8.73 - samples/sec: 2999.06 - lr: 0.000019 - momentum: 0.000000 2023-10-16 18:46:05,695 epoch 7 - iter 98/146 - loss 0.02423133 - time (sec): 10.10 - samples/sec: 3005.29 - lr: 0.000019 - momentum: 0.000000 2023-10-16 18:46:07,233 epoch 7 - iter 112/146 - loss 0.02532614 - time (sec): 11.64 - samples/sec: 2964.77 - lr: 0.000018 - momentum: 0.000000 2023-10-16 18:46:08,741 epoch 7 - iter 126/146 - loss 0.02581130 - time (sec): 13.15 - samples/sec: 2945.18 - lr: 0.000018 - momentum: 0.000000 2023-10-16 18:46:10,281 epoch 7 - iter 140/146 - loss 0.02646009 - time (sec): 14.69 - samples/sec: 2926.02 - lr: 0.000017 - momentum: 0.000000 2023-10-16 18:46:10,770 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:46:10,770 EPOCH 7 done: loss 0.0263 - lr: 0.000017 2023-10-16 18:46:12,029 DEV : loss 0.12524546682834625 - f1-score (micro avg) 0.7613 2023-10-16 18:46:12,034 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:46:13,421 epoch 8 - iter 14/146 - loss 0.01540096 - time (sec): 1.39 - samples/sec: 3120.75 - lr: 0.000016 - momentum: 0.000000 2023-10-16 18:46:14,944 epoch 8 - iter 28/146 - loss 0.01738761 - time (sec): 2.91 - samples/sec: 3063.83 - lr: 0.000016 - momentum: 0.000000 2023-10-16 18:46:16,589 epoch 8 - iter 42/146 - loss 0.02301455 - time (sec): 4.55 - samples/sec: 3009.18 - lr: 0.000015 - momentum: 0.000000 2023-10-16 18:46:17,768 epoch 8 - iter 56/146 - loss 0.02368094 - time (sec): 5.73 - samples/sec: 2956.04 - lr: 0.000015 - momentum: 0.000000 2023-10-16 18:46:19,225 epoch 8 - iter 70/146 - loss 0.02328872 - time (sec): 7.19 - samples/sec: 2990.68 - lr: 0.000014 - momentum: 0.000000 2023-10-16 18:46:20,897 epoch 8 - iter 84/146 - loss 0.02214130 - time (sec): 8.86 - samples/sec: 2969.74 - lr: 0.000014 - momentum: 0.000000 2023-10-16 18:46:22,215 epoch 8 - iter 98/146 - loss 0.02037658 - time (sec): 10.18 - samples/sec: 3002.75 - lr: 0.000013 - momentum: 0.000000 2023-10-16 18:46:23,724 epoch 8 - iter 112/146 - loss 0.02107436 - time (sec): 11.69 - samples/sec: 2975.20 - lr: 0.000013 - momentum: 0.000000 2023-10-16 18:46:25,210 epoch 8 - iter 126/146 - loss 0.02013553 - time (sec): 13.18 - samples/sec: 2976.87 - lr: 0.000012 - momentum: 0.000000 2023-10-16 18:46:26,439 epoch 8 - iter 140/146 - loss 0.01954804 - time (sec): 14.40 - samples/sec: 2984.73 - lr: 0.000012 - momentum: 0.000000 2023-10-16 18:46:26,930 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:46:26,930 EPOCH 8 done: loss 0.0194 - lr: 0.000012 2023-10-16 18:46:28,209 DEV : loss 0.1471555233001709 - f1-score (micro avg) 0.7468 2023-10-16 18:46:28,214 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:46:29,632 epoch 9 - iter 14/146 - loss 0.02032134 - time (sec): 1.42 - samples/sec: 3324.63 - lr: 0.000011 - momentum: 0.000000 2023-10-16 18:46:31,349 epoch 9 - iter 28/146 - loss 0.01946398 - time (sec): 3.13 - samples/sec: 2988.38 - lr: 0.000010 - momentum: 0.000000 2023-10-16 18:46:32,688 epoch 9 - iter 42/146 - loss 0.01654113 - time (sec): 4.47 - samples/sec: 2913.92 - lr: 0.000010 - momentum: 0.000000 2023-10-16 18:46:34,080 epoch 9 - iter 56/146 - loss 0.01628118 - time (sec): 5.86 - samples/sec: 2922.13 - lr: 0.000009 - momentum: 0.000000 2023-10-16 18:46:35,284 epoch 9 - iter 70/146 - loss 0.01650357 - time (sec): 7.07 - samples/sec: 2956.01 - lr: 0.000009 - momentum: 0.000000 2023-10-16 18:46:36,626 epoch 9 - iter 84/146 - loss 0.01631769 - time (sec): 8.41 - samples/sec: 2971.67 - lr: 0.000008 - momentum: 0.000000 2023-10-16 18:46:38,373 epoch 9 - iter 98/146 - loss 0.01409268 - time (sec): 10.16 - samples/sec: 2910.90 - lr: 0.000008 - momentum: 0.000000 2023-10-16 18:46:39,652 epoch 9 - iter 112/146 - loss 0.01357809 - time (sec): 11.44 - samples/sec: 2907.10 - lr: 0.000007 - momentum: 0.000000 2023-10-16 18:46:41,271 epoch 9 - iter 126/146 - loss 0.01361320 - time (sec): 13.06 - samples/sec: 2887.92 - lr: 0.000007 - momentum: 0.000000 2023-10-16 18:46:42,819 epoch 9 - iter 140/146 - loss 0.01401163 - time (sec): 14.60 - samples/sec: 2903.52 - lr: 0.000006 - momentum: 0.000000 2023-10-16 18:46:43,393 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:46:43,393 EPOCH 9 done: loss 0.0137 - lr: 0.000006 2023-10-16 18:46:44,724 DEV : loss 0.14237351715564728 - f1-score (micro avg) 0.7425 2023-10-16 18:46:44,729 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:46:46,002 epoch 10 - iter 14/146 - loss 0.01164471 - time (sec): 1.27 - samples/sec: 2898.78 - lr: 0.000005 - momentum: 0.000000 2023-10-16 18:46:47,276 epoch 10 - iter 28/146 - loss 0.00769121 - time (sec): 2.55 - samples/sec: 3100.67 - lr: 0.000005 - momentum: 0.000000 2023-10-16 18:46:48,584 epoch 10 - iter 42/146 - loss 0.01041592 - time (sec): 3.85 - samples/sec: 3128.38 - lr: 0.000004 - momentum: 0.000000 2023-10-16 18:46:50,364 epoch 10 - iter 56/146 - loss 0.01063731 - time (sec): 5.63 - samples/sec: 3008.74 - lr: 0.000004 - momentum: 0.000000 2023-10-16 18:46:51,888 epoch 10 - iter 70/146 - loss 0.01116153 - time (sec): 7.16 - samples/sec: 3038.46 - lr: 0.000003 - momentum: 0.000000 2023-10-16 18:46:53,303 epoch 10 - iter 84/146 - loss 0.01395303 - time (sec): 8.57 - samples/sec: 3047.73 - lr: 0.000003 - momentum: 0.000000 2023-10-16 18:46:54,633 epoch 10 - iter 98/146 - loss 0.01344263 - time (sec): 9.90 - samples/sec: 3046.71 - lr: 0.000002 - momentum: 0.000000 2023-10-16 18:46:56,080 epoch 10 - iter 112/146 - loss 0.01227917 - time (sec): 11.35 - samples/sec: 3019.23 - lr: 0.000002 - momentum: 0.000000 2023-10-16 18:46:57,603 epoch 10 - iter 126/146 - loss 0.01125604 - time (sec): 12.87 - samples/sec: 3017.04 - lr: 0.000001 - momentum: 0.000000 2023-10-16 18:46:59,081 epoch 10 - iter 140/146 - loss 0.01038127 - time (sec): 14.35 - samples/sec: 3020.01 - lr: 0.000000 - momentum: 0.000000 2023-10-16 18:46:59,533 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:46:59,533 EPOCH 10 done: loss 0.0104 - lr: 0.000000 2023-10-16 18:47:00,784 DEV : loss 0.14098528027534485 - f1-score (micro avg) 0.7532 2023-10-16 18:47:01,250 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:47:01,252 Loading model from best epoch ... 2023-10-16 18:47:02,920 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 2023-10-16 18:47:05,389 Results: - F-score (micro) 0.7535 - F-score (macro) 0.6922 - Accuracy 0.6267 By class: precision recall f1-score support PER 0.7902 0.8333 0.8112 348 LOC 0.6719 0.8161 0.7370 261 ORG 0.4340 0.4423 0.4381 52 HumanProd 0.7500 0.8182 0.7826 22 micro avg 0.7148 0.7965 0.7535 683 macro avg 0.6615 0.7275 0.6922 683 weighted avg 0.7166 0.7965 0.7535 683 2023-10-16 18:47:05,390 ----------------------------------------------------------------------------------------------------