2023-09-03 18:52:51,771 ---------------------------------------------------------------------------------------------------- 2023-09-03 18:52:51,772 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=21, bias=True) (loss_function): CrossEntropyLoss() )" 2023-09-03 18:52:51,772 ---------------------------------------------------------------------------------------------------- 2023-09-03 18:52:51,772 MultiCorpus: 3575 train + 1235 dev + 1266 test sentences - NER_HIPE_2022 Corpus: 3575 train + 1235 dev + 1266 test sentences - /app/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/de/with_doc_seperator 2023-09-03 18:52:51,772 ---------------------------------------------------------------------------------------------------- 2023-09-03 18:52:51,772 Train: 3575 sentences 2023-09-03 18:52:51,772 (train_with_dev=False, train_with_test=False) 2023-09-03 18:52:51,772 ---------------------------------------------------------------------------------------------------- 2023-09-03 18:52:51,773 Training Params: 2023-09-03 18:52:51,773 - learning_rate: "3e-05" 2023-09-03 18:52:51,773 - mini_batch_size: "4" 2023-09-03 18:52:51,773 - max_epochs: "10" 2023-09-03 18:52:51,773 - shuffle: "True" 2023-09-03 18:52:51,773 ---------------------------------------------------------------------------------------------------- 2023-09-03 18:52:51,773 Plugins: 2023-09-03 18:52:51,773 - LinearScheduler | warmup_fraction: '0.1' 2023-09-03 18:52:51,773 ---------------------------------------------------------------------------------------------------- 2023-09-03 18:52:51,773 Final evaluation on model from best epoch (best-model.pt) 2023-09-03 18:52:51,773 - metric: "('micro avg', 'f1-score')" 2023-09-03 18:52:51,773 ---------------------------------------------------------------------------------------------------- 2023-09-03 18:52:51,773 Computation: 2023-09-03 18:52:51,773 - compute on device: cuda:0 2023-09-03 18:52:51,773 - embedding storage: none 2023-09-03 18:52:51,773 ---------------------------------------------------------------------------------------------------- 2023-09-03 18:52:51,773 Model training base path: "hmbench-hipe2020/de-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-1" 2023-09-03 18:52:51,773 ---------------------------------------------------------------------------------------------------- 2023-09-03 18:52:51,773 ---------------------------------------------------------------------------------------------------- 2023-09-03 18:53:01,513 epoch 1 - iter 89/894 - loss 2.99699689 - time (sec): 9.74 - samples/sec: 982.05 - lr: 0.000003 - momentum: 0.000000 2023-09-03 18:53:10,325 epoch 1 - iter 178/894 - loss 2.06986610 - time (sec): 18.55 - samples/sec: 946.08 - lr: 0.000006 - momentum: 0.000000 2023-09-03 18:53:18,954 epoch 1 - iter 267/894 - loss 1.58199999 - time (sec): 27.18 - samples/sec: 930.69 - lr: 0.000009 - momentum: 0.000000 2023-09-03 18:53:27,950 epoch 1 - iter 356/894 - loss 1.28323944 - time (sec): 36.18 - samples/sec: 935.84 - lr: 0.000012 - momentum: 0.000000 2023-09-03 18:53:37,044 epoch 1 - iter 445/894 - loss 1.10202507 - time (sec): 45.27 - samples/sec: 928.12 - lr: 0.000015 - momentum: 0.000000 2023-09-03 18:53:46,353 epoch 1 - iter 534/894 - loss 0.96300193 - time (sec): 54.58 - samples/sec: 933.42 - lr: 0.000018 - momentum: 0.000000 2023-09-03 18:53:55,548 epoch 1 - iter 623/894 - loss 0.86405084 - time (sec): 63.77 - samples/sec: 933.57 - lr: 0.000021 - momentum: 0.000000 2023-09-03 18:54:05,482 epoch 1 - iter 712/894 - loss 0.77769083 - time (sec): 73.71 - samples/sec: 937.92 - lr: 0.000024 - momentum: 0.000000 2023-09-03 18:54:14,345 epoch 1 - iter 801/894 - loss 0.72155531 - time (sec): 82.57 - samples/sec: 933.95 - lr: 0.000027 - momentum: 0.000000 2023-09-03 18:54:23,977 epoch 1 - iter 890/894 - loss 0.67418228 - time (sec): 92.20 - samples/sec: 933.80 - lr: 0.000030 - momentum: 0.000000 2023-09-03 18:54:24,479 ---------------------------------------------------------------------------------------------------- 2023-09-03 18:54:24,479 EPOCH 1 done: loss 0.6716 - lr: 0.000030 2023-09-03 18:54:35,493 DEV : loss 0.19621583819389343 - f1-score (micro avg) 0.5235 2023-09-03 18:54:35,520 saving best model 2023-09-03 18:54:35,970 ---------------------------------------------------------------------------------------------------- 2023-09-03 18:54:44,922 epoch 2 - iter 89/894 - loss 0.17209427 - time (sec): 8.95 - samples/sec: 965.42 - lr: 0.000030 - momentum: 0.000000 2023-09-03 18:54:53,969 epoch 2 - iter 178/894 - loss 0.18691490 - time (sec): 18.00 - samples/sec: 959.15 - lr: 0.000029 - momentum: 0.000000 2023-09-03 18:55:02,872 epoch 2 - iter 267/894 - loss 0.18352945 - time (sec): 26.90 - samples/sec: 964.77 - lr: 0.000029 - momentum: 0.000000 2023-09-03 18:55:12,051 epoch 2 - iter 356/894 - loss 0.17845196 - time (sec): 36.08 - samples/sec: 952.37 - lr: 0.000029 - momentum: 0.000000 2023-09-03 18:55:20,789 epoch 2 - iter 445/894 - loss 0.17332474 - time (sec): 44.82 - samples/sec: 945.81 - lr: 0.000028 - momentum: 0.000000 2023-09-03 18:55:29,919 epoch 2 - iter 534/894 - loss 0.16393436 - time (sec): 53.95 - samples/sec: 948.31 - lr: 0.000028 - momentum: 0.000000 2023-09-03 18:55:39,031 epoch 2 - iter 623/894 - loss 0.16288261 - time (sec): 63.06 - samples/sec: 957.44 - lr: 0.000028 - momentum: 0.000000 2023-09-03 18:55:47,840 epoch 2 - iter 712/894 - loss 0.16084894 - time (sec): 71.87 - samples/sec: 955.50 - lr: 0.000027 - momentum: 0.000000 2023-09-03 18:55:56,568 epoch 2 - iter 801/894 - loss 0.16255124 - time (sec): 80.60 - samples/sec: 952.88 - lr: 0.000027 - momentum: 0.000000 2023-09-03 18:56:06,100 epoch 2 - iter 890/894 - loss 0.15871408 - time (sec): 90.13 - samples/sec: 956.94 - lr: 0.000027 - momentum: 0.000000 2023-09-03 18:56:06,466 ---------------------------------------------------------------------------------------------------- 2023-09-03 18:56:06,467 EPOCH 2 done: loss 0.1585 - lr: 0.000027 2023-09-03 18:56:19,172 DEV : loss 0.13305141031742096 - f1-score (micro avg) 0.7028 2023-09-03 18:56:19,199 saving best model 2023-09-03 18:56:20,516 ---------------------------------------------------------------------------------------------------- 2023-09-03 18:56:29,013 epoch 3 - iter 89/894 - loss 0.08962806 - time (sec): 8.50 - samples/sec: 919.40 - lr: 0.000026 - momentum: 0.000000 2023-09-03 18:56:37,719 epoch 3 - iter 178/894 - loss 0.08780954 - time (sec): 17.20 - samples/sec: 940.66 - lr: 0.000026 - momentum: 0.000000 2023-09-03 18:56:46,548 epoch 3 - iter 267/894 - loss 0.09281134 - time (sec): 26.03 - samples/sec: 935.89 - lr: 0.000026 - momentum: 0.000000 2023-09-03 18:56:55,781 epoch 3 - iter 356/894 - loss 0.08453025 - time (sec): 35.26 - samples/sec: 942.54 - lr: 0.000025 - momentum: 0.000000 2023-09-03 18:57:05,448 epoch 3 - iter 445/894 - loss 0.09105320 - time (sec): 44.93 - samples/sec: 944.93 - lr: 0.000025 - momentum: 0.000000 2023-09-03 18:57:14,265 epoch 3 - iter 534/894 - loss 0.08878389 - time (sec): 53.75 - samples/sec: 955.67 - lr: 0.000025 - momentum: 0.000000 2023-09-03 18:57:23,378 epoch 3 - iter 623/894 - loss 0.09032482 - time (sec): 62.86 - samples/sec: 954.16 - lr: 0.000024 - momentum: 0.000000 2023-09-03 18:57:32,300 epoch 3 - iter 712/894 - loss 0.09053936 - time (sec): 71.78 - samples/sec: 954.45 - lr: 0.000024 - momentum: 0.000000 2023-09-03 18:57:40,998 epoch 3 - iter 801/894 - loss 0.09166475 - time (sec): 80.48 - samples/sec: 956.22 - lr: 0.000024 - momentum: 0.000000 2023-09-03 18:57:50,674 epoch 3 - iter 890/894 - loss 0.09112858 - time (sec): 90.16 - samples/sec: 956.64 - lr: 0.000023 - momentum: 0.000000 2023-09-03 18:57:51,028 ---------------------------------------------------------------------------------------------------- 2023-09-03 18:57:51,028 EPOCH 3 done: loss 0.0910 - lr: 0.000023 2023-09-03 18:58:04,037 DEV : loss 0.14156648516654968 - f1-score (micro avg) 0.7312 2023-09-03 18:58:04,064 saving best model 2023-09-03 18:58:05,396 ---------------------------------------------------------------------------------------------------- 2023-09-03 18:58:14,302 epoch 4 - iter 89/894 - loss 0.06843269 - time (sec): 8.91 - samples/sec: 1012.66 - lr: 0.000023 - momentum: 0.000000 2023-09-03 18:58:23,123 epoch 4 - iter 178/894 - loss 0.06184409 - time (sec): 17.73 - samples/sec: 968.09 - lr: 0.000023 - momentum: 0.000000 2023-09-03 18:58:32,577 epoch 4 - iter 267/894 - loss 0.05872966 - time (sec): 27.18 - samples/sec: 966.73 - lr: 0.000022 - momentum: 0.000000 2023-09-03 18:58:42,449 epoch 4 - iter 356/894 - loss 0.05557241 - time (sec): 37.05 - samples/sec: 975.48 - lr: 0.000022 - momentum: 0.000000 2023-09-03 18:58:51,812 epoch 4 - iter 445/894 - loss 0.05467367 - time (sec): 46.41 - samples/sec: 965.79 - lr: 0.000022 - momentum: 0.000000 2023-09-03 18:59:01,031 epoch 4 - iter 534/894 - loss 0.05481570 - time (sec): 55.63 - samples/sec: 962.60 - lr: 0.000021 - momentum: 0.000000 2023-09-03 18:59:09,843 epoch 4 - iter 623/894 - loss 0.05465991 - time (sec): 64.45 - samples/sec: 959.38 - lr: 0.000021 - momentum: 0.000000 2023-09-03 18:59:18,839 epoch 4 - iter 712/894 - loss 0.05541652 - time (sec): 73.44 - samples/sec: 956.47 - lr: 0.000021 - momentum: 0.000000 2023-09-03 18:59:27,327 epoch 4 - iter 801/894 - loss 0.05463429 - time (sec): 81.93 - samples/sec: 947.52 - lr: 0.000020 - momentum: 0.000000 2023-09-03 18:59:36,709 epoch 4 - iter 890/894 - loss 0.05453935 - time (sec): 91.31 - samples/sec: 944.18 - lr: 0.000020 - momentum: 0.000000 2023-09-03 18:59:37,086 ---------------------------------------------------------------------------------------------------- 2023-09-03 18:59:37,086 EPOCH 4 done: loss 0.0549 - lr: 0.000020 2023-09-03 18:59:50,540 DEV : loss 0.21218053996562958 - f1-score (micro avg) 0.7671 2023-09-03 18:59:50,566 saving best model 2023-09-03 18:59:52,162 ---------------------------------------------------------------------------------------------------- 2023-09-03 19:00:02,477 epoch 5 - iter 89/894 - loss 0.04240015 - time (sec): 10.31 - samples/sec: 942.29 - lr: 0.000020 - momentum: 0.000000 2023-09-03 19:00:11,515 epoch 5 - iter 178/894 - loss 0.04116963 - time (sec): 19.35 - samples/sec: 917.53 - lr: 0.000019 - momentum: 0.000000 2023-09-03 19:00:20,920 epoch 5 - iter 267/894 - loss 0.04532837 - time (sec): 28.76 - samples/sec: 924.58 - lr: 0.000019 - momentum: 0.000000 2023-09-03 19:00:29,740 epoch 5 - iter 356/894 - loss 0.04477082 - time (sec): 37.58 - samples/sec: 923.25 - lr: 0.000019 - momentum: 0.000000 2023-09-03 19:00:39,349 epoch 5 - iter 445/894 - loss 0.04118324 - time (sec): 47.19 - samples/sec: 926.62 - lr: 0.000018 - momentum: 0.000000 2023-09-03 19:00:48,518 epoch 5 - iter 534/894 - loss 0.04162335 - time (sec): 56.35 - samples/sec: 931.35 - lr: 0.000018 - momentum: 0.000000 2023-09-03 19:00:57,557 epoch 5 - iter 623/894 - loss 0.03943254 - time (sec): 65.39 - samples/sec: 928.17 - lr: 0.000018 - momentum: 0.000000 2023-09-03 19:01:06,797 epoch 5 - iter 712/894 - loss 0.03955068 - time (sec): 74.63 - samples/sec: 930.83 - lr: 0.000017 - momentum: 0.000000 2023-09-03 19:01:15,902 epoch 5 - iter 801/894 - loss 0.03921819 - time (sec): 83.74 - samples/sec: 928.46 - lr: 0.000017 - momentum: 0.000000 2023-09-03 19:01:24,861 epoch 5 - iter 890/894 - loss 0.04074719 - time (sec): 92.70 - samples/sec: 929.92 - lr: 0.000017 - momentum: 0.000000 2023-09-03 19:01:25,294 ---------------------------------------------------------------------------------------------------- 2023-09-03 19:01:25,294 EPOCH 5 done: loss 0.0407 - lr: 0.000017 2023-09-03 19:01:38,788 DEV : loss 0.22197993099689484 - f1-score (micro avg) 0.7586 2023-09-03 19:01:38,815 ---------------------------------------------------------------------------------------------------- 2023-09-03 19:01:48,171 epoch 6 - iter 89/894 - loss 0.01977262 - time (sec): 9.36 - samples/sec: 926.42 - lr: 0.000016 - momentum: 0.000000 2023-09-03 19:01:57,047 epoch 6 - iter 178/894 - loss 0.02907246 - time (sec): 18.23 - samples/sec: 898.61 - lr: 0.000016 - momentum: 0.000000 2023-09-03 19:02:06,793 epoch 6 - iter 267/894 - loss 0.02606421 - time (sec): 27.98 - samples/sec: 913.25 - lr: 0.000016 - momentum: 0.000000 2023-09-03 19:02:15,974 epoch 6 - iter 356/894 - loss 0.02546854 - time (sec): 37.16 - samples/sec: 926.86 - lr: 0.000015 - momentum: 0.000000 2023-09-03 19:02:24,624 epoch 6 - iter 445/894 - loss 0.02436981 - time (sec): 45.81 - samples/sec: 917.13 - lr: 0.000015 - momentum: 0.000000 2023-09-03 19:02:33,659 epoch 6 - iter 534/894 - loss 0.02428737 - time (sec): 54.84 - samples/sec: 916.86 - lr: 0.000015 - momentum: 0.000000 2023-09-03 19:02:42,539 epoch 6 - iter 623/894 - loss 0.02613954 - time (sec): 63.72 - samples/sec: 912.52 - lr: 0.000014 - momentum: 0.000000 2023-09-03 19:02:52,855 epoch 6 - iter 712/894 - loss 0.02543888 - time (sec): 74.04 - samples/sec: 923.24 - lr: 0.000014 - momentum: 0.000000 2023-09-03 19:03:01,984 epoch 6 - iter 801/894 - loss 0.02680363 - time (sec): 83.17 - samples/sec: 924.51 - lr: 0.000014 - momentum: 0.000000 2023-09-03 19:03:11,677 epoch 6 - iter 890/894 - loss 0.02684866 - time (sec): 92.86 - samples/sec: 927.32 - lr: 0.000013 - momentum: 0.000000 2023-09-03 19:03:12,085 ---------------------------------------------------------------------------------------------------- 2023-09-03 19:03:12,085 EPOCH 6 done: loss 0.0269 - lr: 0.000013 2023-09-03 19:03:25,620 DEV : loss 0.23417820036411285 - f1-score (micro avg) 0.7583 2023-09-03 19:03:25,647 ---------------------------------------------------------------------------------------------------- 2023-09-03 19:03:34,698 epoch 7 - iter 89/894 - loss 0.02350094 - time (sec): 9.05 - samples/sec: 971.10 - lr: 0.000013 - momentum: 0.000000 2023-09-03 19:03:43,659 epoch 7 - iter 178/894 - loss 0.01866029 - time (sec): 18.01 - samples/sec: 949.80 - lr: 0.000013 - momentum: 0.000000 2023-09-03 19:03:54,334 epoch 7 - iter 267/894 - loss 0.01757408 - time (sec): 28.69 - samples/sec: 950.15 - lr: 0.000012 - momentum: 0.000000 2023-09-03 19:04:03,555 epoch 7 - iter 356/894 - loss 0.01632661 - time (sec): 37.91 - samples/sec: 939.89 - lr: 0.000012 - momentum: 0.000000 2023-09-03 19:04:12,843 epoch 7 - iter 445/894 - loss 0.01696761 - time (sec): 47.20 - samples/sec: 940.77 - lr: 0.000012 - momentum: 0.000000 2023-09-03 19:04:21,813 epoch 7 - iter 534/894 - loss 0.01791246 - time (sec): 56.17 - samples/sec: 936.56 - lr: 0.000011 - momentum: 0.000000 2023-09-03 19:04:31,035 epoch 7 - iter 623/894 - loss 0.01771411 - time (sec): 65.39 - samples/sec: 932.24 - lr: 0.000011 - momentum: 0.000000 2023-09-03 19:04:40,185 epoch 7 - iter 712/894 - loss 0.01755280 - time (sec): 74.54 - samples/sec: 928.22 - lr: 0.000011 - momentum: 0.000000 2023-09-03 19:04:49,163 epoch 7 - iter 801/894 - loss 0.01881491 - time (sec): 83.52 - samples/sec: 922.93 - lr: 0.000010 - momentum: 0.000000 2023-09-03 19:04:58,887 epoch 7 - iter 890/894 - loss 0.01856801 - time (sec): 93.24 - samples/sec: 925.29 - lr: 0.000010 - momentum: 0.000000 2023-09-03 19:04:59,269 ---------------------------------------------------------------------------------------------------- 2023-09-03 19:04:59,269 EPOCH 7 done: loss 0.0185 - lr: 0.000010 2023-09-03 19:05:12,824 DEV : loss 0.2219560593366623 - f1-score (micro avg) 0.7553 2023-09-03 19:05:12,853 ---------------------------------------------------------------------------------------------------- 2023-09-03 19:05:22,261 epoch 8 - iter 89/894 - loss 0.00890074 - time (sec): 9.41 - samples/sec: 921.36 - lr: 0.000010 - momentum: 0.000000 2023-09-03 19:05:31,949 epoch 8 - iter 178/894 - loss 0.01187531 - time (sec): 19.09 - samples/sec: 931.09 - lr: 0.000009 - momentum: 0.000000 2023-09-03 19:05:41,200 epoch 8 - iter 267/894 - loss 0.01028799 - time (sec): 28.35 - samples/sec: 948.27 - lr: 0.000009 - momentum: 0.000000 2023-09-03 19:05:51,068 epoch 8 - iter 356/894 - loss 0.00904645 - time (sec): 38.21 - samples/sec: 953.90 - lr: 0.000009 - momentum: 0.000000 2023-09-03 19:06:00,185 epoch 8 - iter 445/894 - loss 0.01215640 - time (sec): 47.33 - samples/sec: 932.74 - lr: 0.000008 - momentum: 0.000000 2023-09-03 19:06:09,410 epoch 8 - iter 534/894 - loss 0.01237356 - time (sec): 56.56 - samples/sec: 930.61 - lr: 0.000008 - momentum: 0.000000 2023-09-03 19:06:18,481 epoch 8 - iter 623/894 - loss 0.01215753 - time (sec): 65.63 - samples/sec: 939.67 - lr: 0.000008 - momentum: 0.000000 2023-09-03 19:06:27,206 epoch 8 - iter 712/894 - loss 0.01202186 - time (sec): 74.35 - samples/sec: 940.57 - lr: 0.000007 - momentum: 0.000000 2023-09-03 19:06:35,843 epoch 8 - iter 801/894 - loss 0.01148155 - time (sec): 82.99 - samples/sec: 941.04 - lr: 0.000007 - momentum: 0.000000 2023-09-03 19:06:44,618 epoch 8 - iter 890/894 - loss 0.01161529 - time (sec): 91.76 - samples/sec: 939.13 - lr: 0.000007 - momentum: 0.000000 2023-09-03 19:06:45,013 ---------------------------------------------------------------------------------------------------- 2023-09-03 19:06:45,013 EPOCH 8 done: loss 0.0116 - lr: 0.000007 2023-09-03 19:06:57,836 DEV : loss 0.2321111261844635 - f1-score (micro avg) 0.7811 2023-09-03 19:06:57,864 saving best model 2023-09-03 19:06:59,193 ---------------------------------------------------------------------------------------------------- 2023-09-03 19:07:07,874 epoch 9 - iter 89/894 - loss 0.00777350 - time (sec): 8.68 - samples/sec: 952.56 - lr: 0.000006 - momentum: 0.000000 2023-09-03 19:07:17,061 epoch 9 - iter 178/894 - loss 0.00609744 - time (sec): 17.87 - samples/sec: 983.74 - lr: 0.000006 - momentum: 0.000000 2023-09-03 19:07:26,154 epoch 9 - iter 267/894 - loss 0.00658637 - time (sec): 26.96 - samples/sec: 960.39 - lr: 0.000006 - momentum: 0.000000 2023-09-03 19:07:35,421 epoch 9 - iter 356/894 - loss 0.00606818 - time (sec): 36.23 - samples/sec: 975.85 - lr: 0.000005 - momentum: 0.000000 2023-09-03 19:07:44,752 epoch 9 - iter 445/894 - loss 0.00770679 - time (sec): 45.56 - samples/sec: 970.09 - lr: 0.000005 - momentum: 0.000000 2023-09-03 19:07:54,114 epoch 9 - iter 534/894 - loss 0.00758096 - time (sec): 54.92 - samples/sec: 973.48 - lr: 0.000005 - momentum: 0.000000 2023-09-03 19:08:02,757 epoch 9 - iter 623/894 - loss 0.00665079 - time (sec): 63.56 - samples/sec: 975.67 - lr: 0.000004 - momentum: 0.000000 2023-09-03 19:08:11,354 epoch 9 - iter 712/894 - loss 0.00691389 - time (sec): 72.16 - samples/sec: 970.27 - lr: 0.000004 - momentum: 0.000000 2023-09-03 19:08:19,860 epoch 9 - iter 801/894 - loss 0.00689763 - time (sec): 80.67 - samples/sec: 969.33 - lr: 0.000004 - momentum: 0.000000 2023-09-03 19:08:28,559 epoch 9 - iter 890/894 - loss 0.00703245 - time (sec): 89.36 - samples/sec: 963.76 - lr: 0.000003 - momentum: 0.000000 2023-09-03 19:08:28,937 ---------------------------------------------------------------------------------------------------- 2023-09-03 19:08:28,937 EPOCH 9 done: loss 0.0070 - lr: 0.000003 2023-09-03 19:08:41,698 DEV : loss 0.24807557463645935 - f1-score (micro avg) 0.7864 2023-09-03 19:08:41,727 saving best model 2023-09-03 19:08:43,045 ---------------------------------------------------------------------------------------------------- 2023-09-03 19:08:52,067 epoch 10 - iter 89/894 - loss 0.00626017 - time (sec): 9.02 - samples/sec: 973.79 - lr: 0.000003 - momentum: 0.000000 2023-09-03 19:09:00,618 epoch 10 - iter 178/894 - loss 0.00580896 - time (sec): 17.57 - samples/sec: 951.60 - lr: 0.000003 - momentum: 0.000000 2023-09-03 19:09:09,338 epoch 10 - iter 267/894 - loss 0.00557829 - time (sec): 26.29 - samples/sec: 961.57 - lr: 0.000002 - momentum: 0.000000 2023-09-03 19:09:18,325 epoch 10 - iter 356/894 - loss 0.00543496 - time (sec): 35.28 - samples/sec: 966.14 - lr: 0.000002 - momentum: 0.000000 2023-09-03 19:09:27,844 epoch 10 - iter 445/894 - loss 0.00500160 - time (sec): 44.80 - samples/sec: 965.53 - lr: 0.000002 - momentum: 0.000000 2023-09-03 19:09:37,143 epoch 10 - iter 534/894 - loss 0.00502452 - time (sec): 54.10 - samples/sec: 965.26 - lr: 0.000001 - momentum: 0.000000 2023-09-03 19:09:46,513 epoch 10 - iter 623/894 - loss 0.00467325 - time (sec): 63.47 - samples/sec: 960.27 - lr: 0.000001 - momentum: 0.000000 2023-09-03 19:09:55,177 epoch 10 - iter 712/894 - loss 0.00489280 - time (sec): 72.13 - samples/sec: 959.59 - lr: 0.000001 - momentum: 0.000000 2023-09-03 19:10:03,944 epoch 10 - iter 801/894 - loss 0.00473651 - time (sec): 80.90 - samples/sec: 953.32 - lr: 0.000000 - momentum: 0.000000 2023-09-03 19:10:13,430 epoch 10 - iter 890/894 - loss 0.00488641 - time (sec): 90.38 - samples/sec: 953.57 - lr: 0.000000 - momentum: 0.000000 2023-09-03 19:10:13,811 ---------------------------------------------------------------------------------------------------- 2023-09-03 19:10:13,811 EPOCH 10 done: loss 0.0049 - lr: 0.000000 2023-09-03 19:10:26,993 DEV : loss 0.24311408400535583 - f1-score (micro avg) 0.7859 2023-09-03 19:10:27,473 ---------------------------------------------------------------------------------------------------- 2023-09-03 19:10:27,475 Loading model from best epoch ... 2023-09-03 19:10:29,409 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 2023-09-03 19:10:39,948 Results: - F-score (micro) 0.7513 - F-score (macro) 0.6694 - Accuracy 0.6173 By class: precision recall f1-score support loc 0.8276 0.8540 0.8406 596 pers 0.6974 0.7267 0.7118 333 org 0.5769 0.4545 0.5085 132 prod 0.6739 0.4697 0.5536 66 time 0.7115 0.7551 0.7327 49 micro avg 0.7552 0.7474 0.7513 1176 macro avg 0.6975 0.6520 0.6694 1176 weighted avg 0.7492 0.7474 0.7462 1176 2023-09-03 19:10:39,948 ----------------------------------------------------------------------------------------------------