2023-10-15 02:14:00,073 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:14:00,074 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=13, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-15 02:14:00,074 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:14:00,074 MultiCorpus: 14465 train + 1392 dev + 2432 test sentences - NER_HIPE_2022 Corpus: 14465 train + 1392 dev + 2432 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/letemps/fr/with_doc_seperator 2023-10-15 02:14:00,074 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:14:00,074 Train: 14465 sentences 2023-10-15 02:14:00,074 (train_with_dev=False, train_with_test=False) 2023-10-15 02:14:00,074 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:14:00,075 Training Params: 2023-10-15 02:14:00,075 - learning_rate: "3e-05" 2023-10-15 02:14:00,075 - mini_batch_size: "8" 2023-10-15 02:14:00,075 - max_epochs: "10" 2023-10-15 02:14:00,075 - shuffle: "True" 2023-10-15 02:14:00,075 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:14:00,075 Plugins: 2023-10-15 02:14:00,075 - LinearScheduler | warmup_fraction: '0.1' 2023-10-15 02:14:00,075 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:14:00,075 Final evaluation on model from best epoch (best-model.pt) 2023-10-15 02:14:00,075 - metric: "('micro avg', 'f1-score')" 2023-10-15 02:14:00,075 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:14:00,075 Computation: 2023-10-15 02:14:00,075 - compute on device: cuda:0 2023-10-15 02:14:00,075 - embedding storage: none 2023-10-15 02:14:00,075 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:14:00,075 Model training base path: "hmbench-letemps/fr-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5" 2023-10-15 02:14:00,075 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:14:00,075 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:14:10,927 epoch 1 - iter 180/1809 - loss 1.71620325 - time (sec): 10.85 - samples/sec: 3306.55 - lr: 0.000003 - momentum: 0.000000 2023-10-15 02:14:22,349 epoch 1 - iter 360/1809 - loss 0.94066267 - time (sec): 22.27 - samples/sec: 3372.78 - lr: 0.000006 - momentum: 0.000000 2023-10-15 02:14:33,394 epoch 1 - iter 540/1809 - loss 0.69247038 - time (sec): 33.32 - samples/sec: 3362.94 - lr: 0.000009 - momentum: 0.000000 2023-10-15 02:14:44,314 epoch 1 - iter 720/1809 - loss 0.55675267 - time (sec): 44.24 - samples/sec: 3376.46 - lr: 0.000012 - momentum: 0.000000 2023-10-15 02:14:55,634 epoch 1 - iter 900/1809 - loss 0.46969736 - time (sec): 55.56 - samples/sec: 3377.19 - lr: 0.000015 - momentum: 0.000000 2023-10-15 02:15:06,901 epoch 1 - iter 1080/1809 - loss 0.41189108 - time (sec): 66.83 - samples/sec: 3355.68 - lr: 0.000018 - momentum: 0.000000 2023-10-15 02:15:18,514 epoch 1 - iter 1260/1809 - loss 0.36613576 - time (sec): 78.44 - samples/sec: 3364.39 - lr: 0.000021 - momentum: 0.000000 2023-10-15 02:15:30,173 epoch 1 - iter 1440/1809 - loss 0.33395568 - time (sec): 90.10 - samples/sec: 3354.90 - lr: 0.000024 - momentum: 0.000000 2023-10-15 02:15:41,643 epoch 1 - iter 1620/1809 - loss 0.30862678 - time (sec): 101.57 - samples/sec: 3345.11 - lr: 0.000027 - momentum: 0.000000 2023-10-15 02:15:53,267 epoch 1 - iter 1800/1809 - loss 0.28855675 - time (sec): 113.19 - samples/sec: 3342.88 - lr: 0.000030 - momentum: 0.000000 2023-10-15 02:15:53,844 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:15:53,844 EPOCH 1 done: loss 0.2879 - lr: 0.000030 2023-10-15 02:15:58,741 DEV : loss 0.13172951340675354 - f1-score (micro avg) 0.619 2023-10-15 02:15:58,782 saving best model 2023-10-15 02:15:59,172 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:16:11,817 epoch 2 - iter 180/1809 - loss 0.08947028 - time (sec): 12.64 - samples/sec: 3016.49 - lr: 0.000030 - momentum: 0.000000 2023-10-15 02:16:23,203 epoch 2 - iter 360/1809 - loss 0.08337000 - time (sec): 24.03 - samples/sec: 3124.59 - lr: 0.000029 - momentum: 0.000000 2023-10-15 02:16:34,060 epoch 2 - iter 540/1809 - loss 0.08264446 - time (sec): 34.89 - samples/sec: 3153.05 - lr: 0.000029 - momentum: 0.000000 2023-10-15 02:16:45,345 epoch 2 - iter 720/1809 - loss 0.08177670 - time (sec): 46.17 - samples/sec: 3219.93 - lr: 0.000029 - momentum: 0.000000 2023-10-15 02:16:56,562 epoch 2 - iter 900/1809 - loss 0.08349985 - time (sec): 57.39 - samples/sec: 3268.13 - lr: 0.000028 - momentum: 0.000000 2023-10-15 02:17:07,800 epoch 2 - iter 1080/1809 - loss 0.08456116 - time (sec): 68.63 - samples/sec: 3302.75 - lr: 0.000028 - momentum: 0.000000 2023-10-15 02:17:19,132 epoch 2 - iter 1260/1809 - loss 0.08295867 - time (sec): 79.96 - samples/sec: 3308.07 - lr: 0.000028 - momentum: 0.000000 2023-10-15 02:17:30,433 epoch 2 - iter 1440/1809 - loss 0.08216898 - time (sec): 91.26 - samples/sec: 3314.28 - lr: 0.000027 - momentum: 0.000000 2023-10-15 02:17:42,075 epoch 2 - iter 1620/1809 - loss 0.08149738 - time (sec): 102.90 - samples/sec: 3308.35 - lr: 0.000027 - momentum: 0.000000 2023-10-15 02:17:53,256 epoch 2 - iter 1800/1809 - loss 0.08048420 - time (sec): 114.08 - samples/sec: 3314.18 - lr: 0.000027 - momentum: 0.000000 2023-10-15 02:17:53,802 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:17:53,803 EPOCH 2 done: loss 0.0807 - lr: 0.000027 2023-10-15 02:17:59,385 DEV : loss 0.12593407928943634 - f1-score (micro avg) 0.6396 2023-10-15 02:17:59,416 saving best model 2023-10-15 02:17:59,915 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:18:11,666 epoch 3 - iter 180/1809 - loss 0.05780327 - time (sec): 11.75 - samples/sec: 3344.57 - lr: 0.000026 - momentum: 0.000000 2023-10-15 02:18:23,153 epoch 3 - iter 360/1809 - loss 0.05701266 - time (sec): 23.24 - samples/sec: 3305.08 - lr: 0.000026 - momentum: 0.000000 2023-10-15 02:18:34,765 epoch 3 - iter 540/1809 - loss 0.05674387 - time (sec): 34.85 - samples/sec: 3283.83 - lr: 0.000026 - momentum: 0.000000 2023-10-15 02:18:46,421 epoch 3 - iter 720/1809 - loss 0.05547901 - time (sec): 46.50 - samples/sec: 3296.20 - lr: 0.000025 - momentum: 0.000000 2023-10-15 02:18:59,135 epoch 3 - iter 900/1809 - loss 0.05680081 - time (sec): 59.22 - samples/sec: 3220.71 - lr: 0.000025 - momentum: 0.000000 2023-10-15 02:19:10,559 epoch 3 - iter 1080/1809 - loss 0.05816395 - time (sec): 70.64 - samples/sec: 3227.10 - lr: 0.000025 - momentum: 0.000000 2023-10-15 02:19:22,063 epoch 3 - iter 1260/1809 - loss 0.05871572 - time (sec): 82.15 - samples/sec: 3224.99 - lr: 0.000024 - momentum: 0.000000 2023-10-15 02:19:33,639 epoch 3 - iter 1440/1809 - loss 0.05780555 - time (sec): 93.72 - samples/sec: 3223.82 - lr: 0.000024 - momentum: 0.000000 2023-10-15 02:19:45,305 epoch 3 - iter 1620/1809 - loss 0.05757005 - time (sec): 105.39 - samples/sec: 3225.36 - lr: 0.000024 - momentum: 0.000000 2023-10-15 02:19:56,661 epoch 3 - iter 1800/1809 - loss 0.05749567 - time (sec): 116.74 - samples/sec: 3239.30 - lr: 0.000023 - momentum: 0.000000 2023-10-15 02:19:57,152 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:19:57,152 EPOCH 3 done: loss 0.0575 - lr: 0.000023 2023-10-15 02:20:02,874 DEV : loss 0.1452324539422989 - f1-score (micro avg) 0.6325 2023-10-15 02:20:02,916 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:20:14,468 epoch 4 - iter 180/1809 - loss 0.03496327 - time (sec): 11.55 - samples/sec: 3129.42 - lr: 0.000023 - momentum: 0.000000 2023-10-15 02:20:26,389 epoch 4 - iter 360/1809 - loss 0.04079733 - time (sec): 23.47 - samples/sec: 3147.05 - lr: 0.000023 - momentum: 0.000000 2023-10-15 02:20:38,078 epoch 4 - iter 540/1809 - loss 0.03823883 - time (sec): 35.16 - samples/sec: 3217.16 - lr: 0.000022 - momentum: 0.000000 2023-10-15 02:20:49,395 epoch 4 - iter 720/1809 - loss 0.04151284 - time (sec): 46.48 - samples/sec: 3230.29 - lr: 0.000022 - momentum: 0.000000 2023-10-15 02:21:01,210 epoch 4 - iter 900/1809 - loss 0.04052903 - time (sec): 58.29 - samples/sec: 3250.78 - lr: 0.000022 - momentum: 0.000000 2023-10-15 02:21:12,762 epoch 4 - iter 1080/1809 - loss 0.03971197 - time (sec): 69.84 - samples/sec: 3252.29 - lr: 0.000021 - momentum: 0.000000 2023-10-15 02:21:23,785 epoch 4 - iter 1260/1809 - loss 0.04066587 - time (sec): 80.87 - samples/sec: 3260.03 - lr: 0.000021 - momentum: 0.000000 2023-10-15 02:21:34,819 epoch 4 - iter 1440/1809 - loss 0.04096194 - time (sec): 91.90 - samples/sec: 3285.20 - lr: 0.000021 - momentum: 0.000000 2023-10-15 02:21:46,289 epoch 4 - iter 1620/1809 - loss 0.04138910 - time (sec): 103.37 - samples/sec: 3289.06 - lr: 0.000020 - momentum: 0.000000 2023-10-15 02:21:57,449 epoch 4 - iter 1800/1809 - loss 0.04070492 - time (sec): 114.53 - samples/sec: 3299.55 - lr: 0.000020 - momentum: 0.000000 2023-10-15 02:21:58,939 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:21:58,939 EPOCH 4 done: loss 0.0408 - lr: 0.000020 2023-10-15 02:22:04,577 DEV : loss 0.20669673383235931 - f1-score (micro avg) 0.6421 2023-10-15 02:22:04,608 saving best model 2023-10-15 02:22:05,114 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:22:16,531 epoch 5 - iter 180/1809 - loss 0.02389214 - time (sec): 11.41 - samples/sec: 3214.97 - lr: 0.000020 - momentum: 0.000000 2023-10-15 02:22:28,115 epoch 5 - iter 360/1809 - loss 0.02800805 - time (sec): 23.00 - samples/sec: 3206.97 - lr: 0.000019 - momentum: 0.000000 2023-10-15 02:22:39,847 epoch 5 - iter 540/1809 - loss 0.02827358 - time (sec): 34.73 - samples/sec: 3185.08 - lr: 0.000019 - momentum: 0.000000 2023-10-15 02:22:51,849 epoch 5 - iter 720/1809 - loss 0.02901182 - time (sec): 46.73 - samples/sec: 3194.78 - lr: 0.000019 - momentum: 0.000000 2023-10-15 02:23:03,331 epoch 5 - iter 900/1809 - loss 0.02864829 - time (sec): 58.22 - samples/sec: 3204.14 - lr: 0.000018 - momentum: 0.000000 2023-10-15 02:23:14,829 epoch 5 - iter 1080/1809 - loss 0.02941320 - time (sec): 69.71 - samples/sec: 3227.23 - lr: 0.000018 - momentum: 0.000000 2023-10-15 02:23:26,496 epoch 5 - iter 1260/1809 - loss 0.02925040 - time (sec): 81.38 - samples/sec: 3223.63 - lr: 0.000018 - momentum: 0.000000 2023-10-15 02:23:38,853 epoch 5 - iter 1440/1809 - loss 0.03024850 - time (sec): 93.74 - samples/sec: 3224.36 - lr: 0.000017 - momentum: 0.000000 2023-10-15 02:23:50,326 epoch 5 - iter 1620/1809 - loss 0.03013904 - time (sec): 105.21 - samples/sec: 3242.00 - lr: 0.000017 - momentum: 0.000000 2023-10-15 02:24:01,655 epoch 5 - iter 1800/1809 - loss 0.03031792 - time (sec): 116.54 - samples/sec: 3245.16 - lr: 0.000017 - momentum: 0.000000 2023-10-15 02:24:02,192 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:24:02,192 EPOCH 5 done: loss 0.0302 - lr: 0.000017 2023-10-15 02:24:09,030 DEV : loss 0.3110675513744354 - f1-score (micro avg) 0.6318 2023-10-15 02:24:09,071 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:24:20,379 epoch 6 - iter 180/1809 - loss 0.01816067 - time (sec): 11.31 - samples/sec: 3292.12 - lr: 0.000016 - momentum: 0.000000 2023-10-15 02:24:31,904 epoch 6 - iter 360/1809 - loss 0.02260225 - time (sec): 22.83 - samples/sec: 3258.83 - lr: 0.000016 - momentum: 0.000000 2023-10-15 02:24:43,547 epoch 6 - iter 540/1809 - loss 0.02107577 - time (sec): 34.47 - samples/sec: 3226.15 - lr: 0.000016 - momentum: 0.000000 2023-10-15 02:24:55,447 epoch 6 - iter 720/1809 - loss 0.02218085 - time (sec): 46.37 - samples/sec: 3219.97 - lr: 0.000015 - momentum: 0.000000 2023-10-15 02:25:07,299 epoch 6 - iter 900/1809 - loss 0.02197810 - time (sec): 58.23 - samples/sec: 3230.90 - lr: 0.000015 - momentum: 0.000000 2023-10-15 02:25:18,759 epoch 6 - iter 1080/1809 - loss 0.02231769 - time (sec): 69.69 - samples/sec: 3242.11 - lr: 0.000015 - momentum: 0.000000 2023-10-15 02:25:30,458 epoch 6 - iter 1260/1809 - loss 0.02186519 - time (sec): 81.39 - samples/sec: 3259.09 - lr: 0.000014 - momentum: 0.000000 2023-10-15 02:25:41,915 epoch 6 - iter 1440/1809 - loss 0.02158008 - time (sec): 92.84 - samples/sec: 3260.83 - lr: 0.000014 - momentum: 0.000000 2023-10-15 02:25:53,268 epoch 6 - iter 1620/1809 - loss 0.02185088 - time (sec): 104.20 - samples/sec: 3259.37 - lr: 0.000014 - momentum: 0.000000 2023-10-15 02:26:04,662 epoch 6 - iter 1800/1809 - loss 0.02134928 - time (sec): 115.59 - samples/sec: 3269.94 - lr: 0.000013 - momentum: 0.000000 2023-10-15 02:26:05,244 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:26:05,244 EPOCH 6 done: loss 0.0214 - lr: 0.000013 2023-10-15 02:26:11,783 DEV : loss 0.3325505256652832 - f1-score (micro avg) 0.6493 2023-10-15 02:26:11,813 saving best model 2023-10-15 02:26:12,331 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:26:23,804 epoch 7 - iter 180/1809 - loss 0.01455509 - time (sec): 11.47 - samples/sec: 3329.38 - lr: 0.000013 - momentum: 0.000000 2023-10-15 02:26:35,078 epoch 7 - iter 360/1809 - loss 0.01121576 - time (sec): 22.75 - samples/sec: 3272.00 - lr: 0.000013 - momentum: 0.000000 2023-10-15 02:26:46,891 epoch 7 - iter 540/1809 - loss 0.01362276 - time (sec): 34.56 - samples/sec: 3260.25 - lr: 0.000012 - momentum: 0.000000 2023-10-15 02:26:58,682 epoch 7 - iter 720/1809 - loss 0.01404265 - time (sec): 46.35 - samples/sec: 3245.06 - lr: 0.000012 - momentum: 0.000000 2023-10-15 02:27:09,857 epoch 7 - iter 900/1809 - loss 0.01380675 - time (sec): 57.52 - samples/sec: 3266.88 - lr: 0.000012 - momentum: 0.000000 2023-10-15 02:27:21,560 epoch 7 - iter 1080/1809 - loss 0.01462766 - time (sec): 69.23 - samples/sec: 3276.78 - lr: 0.000011 - momentum: 0.000000 2023-10-15 02:27:32,867 epoch 7 - iter 1260/1809 - loss 0.01549142 - time (sec): 80.53 - samples/sec: 3283.47 - lr: 0.000011 - momentum: 0.000000 2023-10-15 02:27:44,136 epoch 7 - iter 1440/1809 - loss 0.01554337 - time (sec): 91.80 - samples/sec: 3284.27 - lr: 0.000011 - momentum: 0.000000 2023-10-15 02:27:56,136 epoch 7 - iter 1620/1809 - loss 0.01548596 - time (sec): 103.80 - samples/sec: 3271.47 - lr: 0.000010 - momentum: 0.000000 2023-10-15 02:28:07,652 epoch 7 - iter 1800/1809 - loss 0.01536064 - time (sec): 115.32 - samples/sec: 3279.88 - lr: 0.000010 - momentum: 0.000000 2023-10-15 02:28:08,163 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:28:08,164 EPOCH 7 done: loss 0.0154 - lr: 0.000010 2023-10-15 02:28:15,762 DEV : loss 0.37081876397132874 - f1-score (micro avg) 0.6315 2023-10-15 02:28:15,800 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:28:27,190 epoch 8 - iter 180/1809 - loss 0.00878142 - time (sec): 11.39 - samples/sec: 3326.19 - lr: 0.000010 - momentum: 0.000000 2023-10-15 02:28:38,168 epoch 8 - iter 360/1809 - loss 0.00883600 - time (sec): 22.37 - samples/sec: 3309.93 - lr: 0.000009 - momentum: 0.000000 2023-10-15 02:28:49,182 epoch 8 - iter 540/1809 - loss 0.00912758 - time (sec): 33.38 - samples/sec: 3313.93 - lr: 0.000009 - momentum: 0.000000 2023-10-15 02:29:00,302 epoch 8 - iter 720/1809 - loss 0.00987622 - time (sec): 44.50 - samples/sec: 3351.66 - lr: 0.000009 - momentum: 0.000000 2023-10-15 02:29:11,250 epoch 8 - iter 900/1809 - loss 0.01048489 - time (sec): 55.45 - samples/sec: 3345.57 - lr: 0.000008 - momentum: 0.000000 2023-10-15 02:29:22,244 epoch 8 - iter 1080/1809 - loss 0.01059545 - time (sec): 66.44 - samples/sec: 3354.63 - lr: 0.000008 - momentum: 0.000000 2023-10-15 02:29:33,312 epoch 8 - iter 1260/1809 - loss 0.01035550 - time (sec): 77.51 - samples/sec: 3362.51 - lr: 0.000008 - momentum: 0.000000 2023-10-15 02:29:44,498 epoch 8 - iter 1440/1809 - loss 0.01013690 - time (sec): 88.70 - samples/sec: 3378.06 - lr: 0.000007 - momentum: 0.000000 2023-10-15 02:29:56,057 epoch 8 - iter 1620/1809 - loss 0.00979165 - time (sec): 100.26 - samples/sec: 3380.91 - lr: 0.000007 - momentum: 0.000000 2023-10-15 02:30:07,595 epoch 8 - iter 1800/1809 - loss 0.00963702 - time (sec): 111.79 - samples/sec: 3384.29 - lr: 0.000007 - momentum: 0.000000 2023-10-15 02:30:08,146 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:30:08,146 EPOCH 8 done: loss 0.0097 - lr: 0.000007 2023-10-15 02:30:13,762 DEV : loss 0.4017082154750824 - f1-score (micro avg) 0.6415 2023-10-15 02:30:13,803 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:30:25,977 epoch 9 - iter 180/1809 - loss 0.00765845 - time (sec): 12.17 - samples/sec: 3256.59 - lr: 0.000006 - momentum: 0.000000 2023-10-15 02:30:36,972 epoch 9 - iter 360/1809 - loss 0.00675346 - time (sec): 23.17 - samples/sec: 3322.71 - lr: 0.000006 - momentum: 0.000000 2023-10-15 02:30:47,836 epoch 9 - iter 540/1809 - loss 0.00672088 - time (sec): 34.03 - samples/sec: 3343.95 - lr: 0.000006 - momentum: 0.000000 2023-10-15 02:30:58,928 epoch 9 - iter 720/1809 - loss 0.00646790 - time (sec): 45.12 - samples/sec: 3399.01 - lr: 0.000005 - momentum: 0.000000 2023-10-15 02:31:09,758 epoch 9 - iter 900/1809 - loss 0.00676701 - time (sec): 55.95 - samples/sec: 3406.90 - lr: 0.000005 - momentum: 0.000000 2023-10-15 02:31:20,723 epoch 9 - iter 1080/1809 - loss 0.00669838 - time (sec): 66.92 - samples/sec: 3400.98 - lr: 0.000005 - momentum: 0.000000 2023-10-15 02:31:32,136 epoch 9 - iter 1260/1809 - loss 0.00698570 - time (sec): 78.33 - samples/sec: 3395.23 - lr: 0.000004 - momentum: 0.000000 2023-10-15 02:31:43,046 epoch 9 - iter 1440/1809 - loss 0.00706185 - time (sec): 89.24 - samples/sec: 3404.64 - lr: 0.000004 - momentum: 0.000000 2023-10-15 02:31:54,348 epoch 9 - iter 1620/1809 - loss 0.00720346 - time (sec): 100.54 - samples/sec: 3398.13 - lr: 0.000004 - momentum: 0.000000 2023-10-15 02:32:05,269 epoch 9 - iter 1800/1809 - loss 0.00700774 - time (sec): 111.46 - samples/sec: 3392.28 - lr: 0.000003 - momentum: 0.000000 2023-10-15 02:32:05,812 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:32:05,812 EPOCH 9 done: loss 0.0070 - lr: 0.000003 2023-10-15 02:32:11,497 DEV : loss 0.4086902439594269 - f1-score (micro avg) 0.6492 2023-10-15 02:32:11,544 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:32:23,632 epoch 10 - iter 180/1809 - loss 0.00472486 - time (sec): 12.09 - samples/sec: 3177.79 - lr: 0.000003 - momentum: 0.000000 2023-10-15 02:32:35,272 epoch 10 - iter 360/1809 - loss 0.00533860 - time (sec): 23.73 - samples/sec: 3171.01 - lr: 0.000003 - momentum: 0.000000 2023-10-15 02:32:46,964 epoch 10 - iter 540/1809 - loss 0.00499615 - time (sec): 35.42 - samples/sec: 3218.25 - lr: 0.000002 - momentum: 0.000000 2023-10-15 02:32:59,939 epoch 10 - iter 720/1809 - loss 0.00489132 - time (sec): 48.39 - samples/sec: 3122.81 - lr: 0.000002 - momentum: 0.000000 2023-10-15 02:33:11,964 epoch 10 - iter 900/1809 - loss 0.00443154 - time (sec): 60.42 - samples/sec: 3144.19 - lr: 0.000002 - momentum: 0.000000 2023-10-15 02:33:23,384 epoch 10 - iter 1080/1809 - loss 0.00446488 - time (sec): 71.84 - samples/sec: 3161.77 - lr: 0.000001 - momentum: 0.000000 2023-10-15 02:33:34,855 epoch 10 - iter 1260/1809 - loss 0.00421573 - time (sec): 83.31 - samples/sec: 3181.83 - lr: 0.000001 - momentum: 0.000000 2023-10-15 02:33:46,612 epoch 10 - iter 1440/1809 - loss 0.00414519 - time (sec): 95.07 - samples/sec: 3175.58 - lr: 0.000001 - momentum: 0.000000 2023-10-15 02:33:58,058 epoch 10 - iter 1620/1809 - loss 0.00475133 - time (sec): 106.51 - samples/sec: 3196.13 - lr: 0.000000 - momentum: 0.000000 2023-10-15 02:34:09,631 epoch 10 - iter 1800/1809 - loss 0.00456537 - time (sec): 118.08 - samples/sec: 3203.33 - lr: 0.000000 - momentum: 0.000000 2023-10-15 02:34:10,184 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:34:10,184 EPOCH 10 done: loss 0.0046 - lr: 0.000000 2023-10-15 02:34:15,872 DEV : loss 0.4162753224372864 - f1-score (micro avg) 0.6466 2023-10-15 02:34:16,294 ---------------------------------------------------------------------------------------------------- 2023-10-15 02:34:16,295 Loading model from best epoch ... 2023-10-15 02:34:17,937 SequenceTagger predicts: Dictionary with 13 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 2023-10-15 02:34:25,749 Results: - F-score (micro) 0.6541 - F-score (macro) 0.51 - Accuracy 0.5007 By class: precision recall f1-score support loc 0.6394 0.7800 0.7027 591 pers 0.5851 0.7703 0.6651 357 org 0.1739 0.1519 0.1622 79 micro avg 0.5937 0.7283 0.6541 1027 macro avg 0.4661 0.5674 0.5100 1027 weighted avg 0.5847 0.7283 0.6481 1027 2023-10-15 02:34:25,749 ----------------------------------------------------------------------------------------------------