2023-10-19 21:04:04,067 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:04:04,067 Model: "SequenceTagger( (embeddings): TransformerWordEmbeddings( (model): BertModel( (embeddings): BertEmbeddings( (word_embeddings): Embedding(32001, 128) (position_embeddings): Embedding(512, 128) (token_type_embeddings): Embedding(2, 128) (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) (encoder): BertEncoder( (layer): ModuleList( (0-1): 2 x BertLayer( (attention): BertAttention( (self): BertSelfAttention( (query): Linear(in_features=128, out_features=128, bias=True) (key): Linear(in_features=128, out_features=128, bias=True) (value): Linear(in_features=128, out_features=128, bias=True) (dropout): Dropout(p=0.1, inplace=False) ) (output): BertSelfOutput( (dense): Linear(in_features=128, out_features=128, bias=True) (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) (intermediate): BertIntermediate( (dense): Linear(in_features=128, out_features=512, bias=True) (intermediate_act_fn): GELUActivation() ) (output): BertOutput( (dense): Linear(in_features=512, out_features=128, bias=True) (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) ) (pooler): BertPooler( (dense): Linear(in_features=128, out_features=128, bias=True) (activation): Tanh() ) ) ) (locked_dropout): LockedDropout(p=0.5) (linear): Linear(in_features=128, out_features=17, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-19 21:04:04,067 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:04:04,067 MultiCorpus: 7142 train + 698 dev + 2570 test sentences - NER_HIPE_2022 Corpus: 7142 train + 698 dev + 2570 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fr/with_doc_seperator 2023-10-19 21:04:04,067 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:04:04,067 Train: 7142 sentences 2023-10-19 21:04:04,068 (train_with_dev=False, train_with_test=False) 2023-10-19 21:04:04,068 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:04:04,068 Training Params: 2023-10-19 21:04:04,068 - learning_rate: "3e-05" 2023-10-19 21:04:04,068 - mini_batch_size: "8" 2023-10-19 21:04:04,068 - max_epochs: "10" 2023-10-19 21:04:04,068 - shuffle: "True" 2023-10-19 21:04:04,068 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:04:04,068 Plugins: 2023-10-19 21:04:04,068 - TensorboardLogger 2023-10-19 21:04:04,068 - LinearScheduler | warmup_fraction: '0.1' 2023-10-19 21:04:04,068 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:04:04,068 Final evaluation on model from best epoch (best-model.pt) 2023-10-19 21:04:04,068 - metric: "('micro avg', 'f1-score')" 2023-10-19 21:04:04,068 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:04:04,068 Computation: 2023-10-19 21:04:04,068 - compute on device: cuda:0 2023-10-19 21:04:04,068 - embedding storage: none 2023-10-19 21:04:04,068 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:04:04,068 Model training base path: "hmbench-newseye/fr-dbmdz/bert-tiny-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5" 2023-10-19 21:04:04,068 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:04:04,068 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:04:04,068 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-19 21:04:06,450 epoch 1 - iter 89/893 - loss 3.46516027 - time (sec): 2.38 - samples/sec: 10355.47 - lr: 0.000003 - momentum: 0.000000 2023-10-19 21:04:08,890 epoch 1 - iter 178/893 - loss 3.31633199 - time (sec): 4.82 - samples/sec: 10596.78 - lr: 0.000006 - momentum: 0.000000 2023-10-19 21:04:11,278 epoch 1 - iter 267/893 - loss 3.01321278 - time (sec): 7.21 - samples/sec: 10667.08 - lr: 0.000009 - momentum: 0.000000 2023-10-19 21:04:13,675 epoch 1 - iter 356/893 - loss 2.63035363 - time (sec): 9.61 - samples/sec: 10662.06 - lr: 0.000012 - momentum: 0.000000 2023-10-19 21:04:16,000 epoch 1 - iter 445/893 - loss 2.31455106 - time (sec): 11.93 - samples/sec: 10567.39 - lr: 0.000015 - momentum: 0.000000 2023-10-19 21:04:18,356 epoch 1 - iter 534/893 - loss 2.07618924 - time (sec): 14.29 - samples/sec: 10509.93 - lr: 0.000018 - momentum: 0.000000 2023-10-19 21:04:21,075 epoch 1 - iter 623/893 - loss 1.88468649 - time (sec): 17.01 - samples/sec: 10330.37 - lr: 0.000021 - momentum: 0.000000 2023-10-19 21:04:23,229 epoch 1 - iter 712/893 - loss 1.73722430 - time (sec): 19.16 - samples/sec: 10479.92 - lr: 0.000024 - momentum: 0.000000 2023-10-19 21:04:25,515 epoch 1 - iter 801/893 - loss 1.61864786 - time (sec): 21.45 - samples/sec: 10471.92 - lr: 0.000027 - momentum: 0.000000 2023-10-19 21:04:27,744 epoch 1 - iter 890/893 - loss 1.52299965 - time (sec): 23.67 - samples/sec: 10476.93 - lr: 0.000030 - momentum: 0.000000 2023-10-19 21:04:27,807 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:04:27,808 EPOCH 1 done: loss 1.5206 - lr: 0.000030 2023-10-19 21:04:28,776 DEV : loss 0.36043813824653625 - f1-score (micro avg) 0.0026 2023-10-19 21:04:28,791 saving best model 2023-10-19 21:04:28,825 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:04:30,944 epoch 2 - iter 89/893 - loss 0.54958316 - time (sec): 2.12 - samples/sec: 11145.85 - lr: 0.000030 - momentum: 0.000000 2023-10-19 21:04:33,239 epoch 2 - iter 178/893 - loss 0.53358732 - time (sec): 4.41 - samples/sec: 11059.90 - lr: 0.000029 - momentum: 0.000000 2023-10-19 21:04:35,553 epoch 2 - iter 267/893 - loss 0.52508342 - time (sec): 6.73 - samples/sec: 10890.14 - lr: 0.000029 - momentum: 0.000000 2023-10-19 21:04:37,785 epoch 2 - iter 356/893 - loss 0.51537618 - time (sec): 8.96 - samples/sec: 10867.78 - lr: 0.000029 - momentum: 0.000000 2023-10-19 21:04:40,048 epoch 2 - iter 445/893 - loss 0.50151060 - time (sec): 11.22 - samples/sec: 10908.00 - lr: 0.000028 - momentum: 0.000000 2023-10-19 21:04:42,322 epoch 2 - iter 534/893 - loss 0.49081103 - time (sec): 13.50 - samples/sec: 10977.85 - lr: 0.000028 - momentum: 0.000000 2023-10-19 21:04:44,594 epoch 2 - iter 623/893 - loss 0.48334925 - time (sec): 15.77 - samples/sec: 11038.71 - lr: 0.000028 - momentum: 0.000000 2023-10-19 21:04:46,841 epoch 2 - iter 712/893 - loss 0.48240244 - time (sec): 18.01 - samples/sec: 10944.63 - lr: 0.000027 - momentum: 0.000000 2023-10-19 21:04:49,143 epoch 2 - iter 801/893 - loss 0.47466459 - time (sec): 20.32 - samples/sec: 10920.37 - lr: 0.000027 - momentum: 0.000000 2023-10-19 21:04:51,445 epoch 2 - iter 890/893 - loss 0.47084872 - time (sec): 22.62 - samples/sec: 10964.45 - lr: 0.000027 - momentum: 0.000000 2023-10-19 21:04:51,519 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:04:51,519 EPOCH 2 done: loss 0.4711 - lr: 0.000027 2023-10-19 21:04:54,349 DEV : loss 0.2766191065311432 - f1-score (micro avg) 0.2593 2023-10-19 21:04:54,364 saving best model 2023-10-19 21:04:54,398 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:04:56,666 epoch 3 - iter 89/893 - loss 0.37256391 - time (sec): 2.27 - samples/sec: 11380.92 - lr: 0.000026 - momentum: 0.000000 2023-10-19 21:04:58,952 epoch 3 - iter 178/893 - loss 0.39452300 - time (sec): 4.55 - samples/sec: 10884.87 - lr: 0.000026 - momentum: 0.000000 2023-10-19 21:05:01,194 epoch 3 - iter 267/893 - loss 0.39982878 - time (sec): 6.80 - samples/sec: 10778.53 - lr: 0.000026 - momentum: 0.000000 2023-10-19 21:05:03,441 epoch 3 - iter 356/893 - loss 0.39305945 - time (sec): 9.04 - samples/sec: 10883.19 - lr: 0.000025 - momentum: 0.000000 2023-10-19 21:05:05,683 epoch 3 - iter 445/893 - loss 0.39651155 - time (sec): 11.28 - samples/sec: 10974.19 - lr: 0.000025 - momentum: 0.000000 2023-10-19 21:05:07,927 epoch 3 - iter 534/893 - loss 0.39642494 - time (sec): 13.53 - samples/sec: 11049.80 - lr: 0.000025 - momentum: 0.000000 2023-10-19 21:05:10,208 epoch 3 - iter 623/893 - loss 0.39408089 - time (sec): 15.81 - samples/sec: 11044.80 - lr: 0.000024 - momentum: 0.000000 2023-10-19 21:05:12,406 epoch 3 - iter 712/893 - loss 0.39397319 - time (sec): 18.01 - samples/sec: 11043.21 - lr: 0.000024 - momentum: 0.000000 2023-10-19 21:05:14,604 epoch 3 - iter 801/893 - loss 0.39353180 - time (sec): 20.21 - samples/sec: 11015.76 - lr: 0.000024 - momentum: 0.000000 2023-10-19 21:05:16,849 epoch 3 - iter 890/893 - loss 0.39134018 - time (sec): 22.45 - samples/sec: 11048.56 - lr: 0.000023 - momentum: 0.000000 2023-10-19 21:05:16,926 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:05:16,926 EPOCH 3 done: loss 0.3922 - lr: 0.000023 2023-10-19 21:05:19,751 DEV : loss 0.2515345811843872 - f1-score (micro avg) 0.3323 2023-10-19 21:05:19,765 saving best model 2023-10-19 21:05:19,799 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:05:22,171 epoch 4 - iter 89/893 - loss 0.36067185 - time (sec): 2.37 - samples/sec: 10806.53 - lr: 0.000023 - momentum: 0.000000 2023-10-19 21:05:24,441 epoch 4 - iter 178/893 - loss 0.35369799 - time (sec): 4.64 - samples/sec: 10634.18 - lr: 0.000023 - momentum: 0.000000 2023-10-19 21:05:26,669 epoch 4 - iter 267/893 - loss 0.35728690 - time (sec): 6.87 - samples/sec: 10774.66 - lr: 0.000022 - momentum: 0.000000 2023-10-19 21:05:28,712 epoch 4 - iter 356/893 - loss 0.36526060 - time (sec): 8.91 - samples/sec: 10963.32 - lr: 0.000022 - momentum: 0.000000 2023-10-19 21:05:30,959 epoch 4 - iter 445/893 - loss 0.36088528 - time (sec): 11.16 - samples/sec: 10981.58 - lr: 0.000022 - momentum: 0.000000 2023-10-19 21:05:33,190 epoch 4 - iter 534/893 - loss 0.36222174 - time (sec): 13.39 - samples/sec: 10925.92 - lr: 0.000021 - momentum: 0.000000 2023-10-19 21:05:35,432 epoch 4 - iter 623/893 - loss 0.35964595 - time (sec): 15.63 - samples/sec: 10947.56 - lr: 0.000021 - momentum: 0.000000 2023-10-19 21:05:37,503 epoch 4 - iter 712/893 - loss 0.35640730 - time (sec): 17.70 - samples/sec: 11213.08 - lr: 0.000021 - momentum: 0.000000 2023-10-19 21:05:39,612 epoch 4 - iter 801/893 - loss 0.35414262 - time (sec): 19.81 - samples/sec: 11249.80 - lr: 0.000020 - momentum: 0.000000 2023-10-19 21:05:41,879 epoch 4 - iter 890/893 - loss 0.35252586 - time (sec): 22.08 - samples/sec: 11214.76 - lr: 0.000020 - momentum: 0.000000 2023-10-19 21:05:41,953 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:05:41,953 EPOCH 4 done: loss 0.3521 - lr: 0.000020 2023-10-19 21:05:44,308 DEV : loss 0.24030107259750366 - f1-score (micro avg) 0.3985 2023-10-19 21:05:44,322 saving best model 2023-10-19 21:05:44,355 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:05:46,615 epoch 5 - iter 89/893 - loss 0.33201364 - time (sec): 2.26 - samples/sec: 10449.33 - lr: 0.000020 - momentum: 0.000000 2023-10-19 21:05:48,898 epoch 5 - iter 178/893 - loss 0.32799586 - time (sec): 4.54 - samples/sec: 10848.84 - lr: 0.000019 - momentum: 0.000000 2023-10-19 21:05:51,137 epoch 5 - iter 267/893 - loss 0.33434089 - time (sec): 6.78 - samples/sec: 10828.88 - lr: 0.000019 - momentum: 0.000000 2023-10-19 21:05:53,351 epoch 5 - iter 356/893 - loss 0.33073391 - time (sec): 9.00 - samples/sec: 10953.80 - lr: 0.000019 - momentum: 0.000000 2023-10-19 21:05:55,663 epoch 5 - iter 445/893 - loss 0.33199483 - time (sec): 11.31 - samples/sec: 10906.21 - lr: 0.000018 - momentum: 0.000000 2023-10-19 21:05:57,878 epoch 5 - iter 534/893 - loss 0.32808703 - time (sec): 13.52 - samples/sec: 10862.79 - lr: 0.000018 - momentum: 0.000000 2023-10-19 21:06:00,156 epoch 5 - iter 623/893 - loss 0.32729446 - time (sec): 15.80 - samples/sec: 11023.39 - lr: 0.000018 - momentum: 0.000000 2023-10-19 21:06:02,408 epoch 5 - iter 712/893 - loss 0.32383033 - time (sec): 18.05 - samples/sec: 10996.11 - lr: 0.000017 - momentum: 0.000000 2023-10-19 21:06:04,750 epoch 5 - iter 801/893 - loss 0.32640604 - time (sec): 20.39 - samples/sec: 10960.58 - lr: 0.000017 - momentum: 0.000000 2023-10-19 21:06:07,022 epoch 5 - iter 890/893 - loss 0.32481877 - time (sec): 22.67 - samples/sec: 10929.86 - lr: 0.000017 - momentum: 0.000000 2023-10-19 21:06:07,109 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:06:07,109 EPOCH 5 done: loss 0.3243 - lr: 0.000017 2023-10-19 21:06:09,945 DEV : loss 0.22777576744556427 - f1-score (micro avg) 0.4374 2023-10-19 21:06:09,958 saving best model 2023-10-19 21:06:09,993 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:06:12,246 epoch 6 - iter 89/893 - loss 0.30901463 - time (sec): 2.25 - samples/sec: 10491.78 - lr: 0.000016 - momentum: 0.000000 2023-10-19 21:06:14,534 epoch 6 - iter 178/893 - loss 0.30776053 - time (sec): 4.54 - samples/sec: 10896.61 - lr: 0.000016 - momentum: 0.000000 2023-10-19 21:06:16,805 epoch 6 - iter 267/893 - loss 0.31279529 - time (sec): 6.81 - samples/sec: 10972.40 - lr: 0.000016 - momentum: 0.000000 2023-10-19 21:06:19,052 epoch 6 - iter 356/893 - loss 0.31110898 - time (sec): 9.06 - samples/sec: 11069.80 - lr: 0.000015 - momentum: 0.000000 2023-10-19 21:06:21,259 epoch 6 - iter 445/893 - loss 0.31454604 - time (sec): 11.27 - samples/sec: 10919.77 - lr: 0.000015 - momentum: 0.000000 2023-10-19 21:06:23,421 epoch 6 - iter 534/893 - loss 0.31388124 - time (sec): 13.43 - samples/sec: 10919.47 - lr: 0.000015 - momentum: 0.000000 2023-10-19 21:06:25,685 epoch 6 - iter 623/893 - loss 0.31175351 - time (sec): 15.69 - samples/sec: 10969.54 - lr: 0.000014 - momentum: 0.000000 2023-10-19 21:06:28,029 epoch 6 - iter 712/893 - loss 0.30862470 - time (sec): 18.04 - samples/sec: 10966.02 - lr: 0.000014 - momentum: 0.000000 2023-10-19 21:06:30,350 epoch 6 - iter 801/893 - loss 0.30704706 - time (sec): 20.36 - samples/sec: 10969.15 - lr: 0.000014 - momentum: 0.000000 2023-10-19 21:06:32,616 epoch 6 - iter 890/893 - loss 0.30734708 - time (sec): 22.62 - samples/sec: 10971.21 - lr: 0.000013 - momentum: 0.000000 2023-10-19 21:06:32,682 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:06:32,682 EPOCH 6 done: loss 0.3076 - lr: 0.000013 2023-10-19 21:06:35,530 DEV : loss 0.21941223740577698 - f1-score (micro avg) 0.4564 2023-10-19 21:06:35,544 saving best model 2023-10-19 21:06:35,582 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:06:37,807 epoch 7 - iter 89/893 - loss 0.30509027 - time (sec): 2.22 - samples/sec: 10655.57 - lr: 0.000013 - momentum: 0.000000 2023-10-19 21:06:40,129 epoch 7 - iter 178/893 - loss 0.28654601 - time (sec): 4.55 - samples/sec: 10694.39 - lr: 0.000013 - momentum: 0.000000 2023-10-19 21:06:42,497 epoch 7 - iter 267/893 - loss 0.28403086 - time (sec): 6.91 - samples/sec: 10509.95 - lr: 0.000012 - momentum: 0.000000 2023-10-19 21:06:44,843 epoch 7 - iter 356/893 - loss 0.28827472 - time (sec): 9.26 - samples/sec: 10509.54 - lr: 0.000012 - momentum: 0.000000 2023-10-19 21:06:47,165 epoch 7 - iter 445/893 - loss 0.29045580 - time (sec): 11.58 - samples/sec: 10583.77 - lr: 0.000012 - momentum: 0.000000 2023-10-19 21:06:49,400 epoch 7 - iter 534/893 - loss 0.30162964 - time (sec): 13.82 - samples/sec: 10697.48 - lr: 0.000011 - momentum: 0.000000 2023-10-19 21:06:51,632 epoch 7 - iter 623/893 - loss 0.30089497 - time (sec): 16.05 - samples/sec: 10683.90 - lr: 0.000011 - momentum: 0.000000 2023-10-19 21:06:53,942 epoch 7 - iter 712/893 - loss 0.29776092 - time (sec): 18.36 - samples/sec: 10762.55 - lr: 0.000011 - momentum: 0.000000 2023-10-19 21:06:56,209 epoch 7 - iter 801/893 - loss 0.29561950 - time (sec): 20.63 - samples/sec: 10809.02 - lr: 0.000010 - momentum: 0.000000 2023-10-19 21:06:58,427 epoch 7 - iter 890/893 - loss 0.29349579 - time (sec): 22.84 - samples/sec: 10841.48 - lr: 0.000010 - momentum: 0.000000 2023-10-19 21:06:58,506 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:06:58,506 EPOCH 7 done: loss 0.2932 - lr: 0.000010 2023-10-19 21:07:00,835 DEV : loss 0.2180010825395584 - f1-score (micro avg) 0.4616 2023-10-19 21:07:00,849 saving best model 2023-10-19 21:07:00,884 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:07:03,134 epoch 8 - iter 89/893 - loss 0.28270141 - time (sec): 2.25 - samples/sec: 11490.73 - lr: 0.000010 - momentum: 0.000000 2023-10-19 21:07:05,412 epoch 8 - iter 178/893 - loss 0.28238330 - time (sec): 4.53 - samples/sec: 11628.05 - lr: 0.000009 - momentum: 0.000000 2023-10-19 21:07:07,718 epoch 8 - iter 267/893 - loss 0.27604606 - time (sec): 6.83 - samples/sec: 11561.75 - lr: 0.000009 - momentum: 0.000000 2023-10-19 21:07:10,009 epoch 8 - iter 356/893 - loss 0.27685070 - time (sec): 9.12 - samples/sec: 11354.16 - lr: 0.000009 - momentum: 0.000000 2023-10-19 21:07:12,333 epoch 8 - iter 445/893 - loss 0.28502945 - time (sec): 11.45 - samples/sec: 11152.36 - lr: 0.000008 - momentum: 0.000000 2023-10-19 21:07:14,695 epoch 8 - iter 534/893 - loss 0.28106511 - time (sec): 13.81 - samples/sec: 11140.31 - lr: 0.000008 - momentum: 0.000000 2023-10-19 21:07:16,982 epoch 8 - iter 623/893 - loss 0.28187735 - time (sec): 16.10 - samples/sec: 10994.00 - lr: 0.000008 - momentum: 0.000000 2023-10-19 21:07:19,237 epoch 8 - iter 712/893 - loss 0.27973353 - time (sec): 18.35 - samples/sec: 10944.00 - lr: 0.000007 - momentum: 0.000000 2023-10-19 21:07:21,454 epoch 8 - iter 801/893 - loss 0.28007865 - time (sec): 20.57 - samples/sec: 10940.20 - lr: 0.000007 - momentum: 0.000000 2023-10-19 21:07:23,659 epoch 8 - iter 890/893 - loss 0.28450041 - time (sec): 22.77 - samples/sec: 10890.11 - lr: 0.000007 - momentum: 0.000000 2023-10-19 21:07:23,729 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:07:23,729 EPOCH 8 done: loss 0.2843 - lr: 0.000007 2023-10-19 21:07:26,548 DEV : loss 0.21302838623523712 - f1-score (micro avg) 0.4614 2023-10-19 21:07:26,562 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:07:28,823 epoch 9 - iter 89/893 - loss 0.27434130 - time (sec): 2.26 - samples/sec: 11242.35 - lr: 0.000006 - momentum: 0.000000 2023-10-19 21:07:31,078 epoch 9 - iter 178/893 - loss 0.27958386 - time (sec): 4.52 - samples/sec: 11145.59 - lr: 0.000006 - momentum: 0.000000 2023-10-19 21:07:33,324 epoch 9 - iter 267/893 - loss 0.28201334 - time (sec): 6.76 - samples/sec: 10968.95 - lr: 0.000006 - momentum: 0.000000 2023-10-19 21:07:35,560 epoch 9 - iter 356/893 - loss 0.28336896 - time (sec): 9.00 - samples/sec: 10917.79 - lr: 0.000005 - momentum: 0.000000 2023-10-19 21:07:37,762 epoch 9 - iter 445/893 - loss 0.28043243 - time (sec): 11.20 - samples/sec: 10900.38 - lr: 0.000005 - momentum: 0.000000 2023-10-19 21:07:40,039 epoch 9 - iter 534/893 - loss 0.27712063 - time (sec): 13.48 - samples/sec: 11038.22 - lr: 0.000005 - momentum: 0.000000 2023-10-19 21:07:42,253 epoch 9 - iter 623/893 - loss 0.27542232 - time (sec): 15.69 - samples/sec: 11006.18 - lr: 0.000004 - momentum: 0.000000 2023-10-19 21:07:44,549 epoch 9 - iter 712/893 - loss 0.27658911 - time (sec): 17.99 - samples/sec: 11007.29 - lr: 0.000004 - momentum: 0.000000 2023-10-19 21:07:46,811 epoch 9 - iter 801/893 - loss 0.27538194 - time (sec): 20.25 - samples/sec: 10997.20 - lr: 0.000004 - momentum: 0.000000 2023-10-19 21:07:49,039 epoch 9 - iter 890/893 - loss 0.27416186 - time (sec): 22.48 - samples/sec: 11034.55 - lr: 0.000003 - momentum: 0.000000 2023-10-19 21:07:49,117 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:07:49,117 EPOCH 9 done: loss 0.2741 - lr: 0.000003 2023-10-19 21:07:51,965 DEV : loss 0.21206232905387878 - f1-score (micro avg) 0.4562 2023-10-19 21:07:51,980 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:07:54,185 epoch 10 - iter 89/893 - loss 0.25869512 - time (sec): 2.20 - samples/sec: 10712.67 - lr: 0.000003 - momentum: 0.000000 2023-10-19 21:07:56,391 epoch 10 - iter 178/893 - loss 0.26163812 - time (sec): 4.41 - samples/sec: 10800.80 - lr: 0.000003 - momentum: 0.000000 2023-10-19 21:07:58,723 epoch 10 - iter 267/893 - loss 0.26555406 - time (sec): 6.74 - samples/sec: 11025.94 - lr: 0.000002 - momentum: 0.000000 2023-10-19 21:08:00,990 epoch 10 - iter 356/893 - loss 0.26148161 - time (sec): 9.01 - samples/sec: 10868.20 - lr: 0.000002 - momentum: 0.000000 2023-10-19 21:08:03,290 epoch 10 - iter 445/893 - loss 0.26190440 - time (sec): 11.31 - samples/sec: 10794.51 - lr: 0.000002 - momentum: 0.000000 2023-10-19 21:08:05,593 epoch 10 - iter 534/893 - loss 0.26271915 - time (sec): 13.61 - samples/sec: 10856.21 - lr: 0.000001 - momentum: 0.000000 2023-10-19 21:08:07,843 epoch 10 - iter 623/893 - loss 0.26537037 - time (sec): 15.86 - samples/sec: 10830.62 - lr: 0.000001 - momentum: 0.000000 2023-10-19 21:08:10,148 epoch 10 - iter 712/893 - loss 0.27004143 - time (sec): 18.17 - samples/sec: 10879.72 - lr: 0.000001 - momentum: 0.000000 2023-10-19 21:08:12,367 epoch 10 - iter 801/893 - loss 0.27091574 - time (sec): 20.39 - samples/sec: 10981.73 - lr: 0.000000 - momentum: 0.000000 2023-10-19 21:08:14,609 epoch 10 - iter 890/893 - loss 0.27304663 - time (sec): 22.63 - samples/sec: 10963.71 - lr: 0.000000 - momentum: 0.000000 2023-10-19 21:08:14,679 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:08:14,679 EPOCH 10 done: loss 0.2730 - lr: 0.000000 2023-10-19 21:08:17,043 DEV : loss 0.21014845371246338 - f1-score (micro avg) 0.4628 2023-10-19 21:08:17,057 saving best model 2023-10-19 21:08:17,117 ---------------------------------------------------------------------------------------------------- 2023-10-19 21:08:17,118 Loading model from best epoch ... 2023-10-19 21:08:17,199 SequenceTagger predicts: Dictionary with 17 tags: O, S-PER, B-PER, E-PER, I-PER, S-LOC, B-LOC, E-LOC, I-LOC, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd 2023-10-19 21:08:21,777 Results: - F-score (micro) 0.3645 - F-score (macro) 0.2036 - Accuracy 0.2332 By class: precision recall f1-score support LOC 0.3682 0.4849 0.4186 1095 PER 0.3427 0.4219 0.3782 1012 ORG 0.0426 0.0112 0.0177 357 HumanProd 0.0000 0.0000 0.0000 33 micro avg 0.3458 0.3853 0.3645 2497 macro avg 0.1884 0.2295 0.2036 2497 weighted avg 0.3065 0.3853 0.3394 2497 2023-10-19 21:08:21,778 ----------------------------------------------------------------------------------------------------