2023-10-17 21:12:15,815 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:12:15,816 Model: "SequenceTagger( (embeddings): TransformerWordEmbeddings( (model): ElectraModel( (embeddings): ElectraEmbeddings( (word_embeddings): Embedding(32001, 768) (position_embeddings): Embedding(512, 768) (token_type_embeddings): Embedding(2, 768) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) (encoder): ElectraEncoder( (layer): ModuleList( (0-11): 12 x ElectraLayer( (attention): ElectraAttention( (self): ElectraSelfAttention( (query): Linear(in_features=768, out_features=768, bias=True) (key): Linear(in_features=768, out_features=768, bias=True) (value): Linear(in_features=768, out_features=768, bias=True) (dropout): Dropout(p=0.1, inplace=False) ) (output): ElectraSelfOutput( (dense): Linear(in_features=768, out_features=768, bias=True) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) (intermediate): ElectraIntermediate( (dense): Linear(in_features=768, out_features=3072, bias=True) (intermediate_act_fn): GELUActivation() ) (output): ElectraOutput( (dense): Linear(in_features=3072, out_features=768, bias=True) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) ) ) ) (locked_dropout): LockedDropout(p=0.5) (linear): Linear(in_features=768, out_features=21, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-17 21:12:15,816 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:12:15,816 MultiCorpus: 5901 train + 1287 dev + 1505 test sentences - NER_HIPE_2022 Corpus: 5901 train + 1287 dev + 1505 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/fr/with_doc_seperator 2023-10-17 21:12:15,816 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:12:15,816 Train: 5901 sentences 2023-10-17 21:12:15,816 (train_with_dev=False, train_with_test=False) 2023-10-17 21:12:15,816 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:12:15,816 Training Params: 2023-10-17 21:12:15,816 - learning_rate: "5e-05" 2023-10-17 21:12:15,816 - mini_batch_size: "4" 2023-10-17 21:12:15,816 - max_epochs: "10" 2023-10-17 21:12:15,816 - shuffle: "True" 2023-10-17 21:12:15,816 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:12:15,816 Plugins: 2023-10-17 21:12:15,816 - TensorboardLogger 2023-10-17 21:12:15,816 - LinearScheduler | warmup_fraction: '0.1' 2023-10-17 21:12:15,817 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:12:15,817 Final evaluation on model from best epoch (best-model.pt) 2023-10-17 21:12:15,817 - metric: "('micro avg', 'f1-score')" 2023-10-17 21:12:15,817 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:12:15,817 Computation: 2023-10-17 21:12:15,817 - compute on device: cuda:0 2023-10-17 21:12:15,817 - embedding storage: none 2023-10-17 21:12:15,817 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:12:15,817 Model training base path: "hmbench-hipe2020/fr-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2" 2023-10-17 21:12:15,817 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:12:15,817 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:12:15,817 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-17 21:12:22,757 epoch 1 - iter 147/1476 - loss 2.71106741 - time (sec): 6.94 - samples/sec: 2321.76 - lr: 0.000005 - momentum: 0.000000 2023-10-17 21:12:30,110 epoch 1 - iter 294/1476 - loss 1.53519710 - time (sec): 14.29 - samples/sec: 2468.89 - lr: 0.000010 - momentum: 0.000000 2023-10-17 21:12:37,137 epoch 1 - iter 441/1476 - loss 1.19741914 - time (sec): 21.32 - samples/sec: 2353.25 - lr: 0.000015 - momentum: 0.000000 2023-10-17 21:12:44,328 epoch 1 - iter 588/1476 - loss 0.98125061 - time (sec): 28.51 - samples/sec: 2350.64 - lr: 0.000020 - momentum: 0.000000 2023-10-17 21:12:51,162 epoch 1 - iter 735/1476 - loss 0.84406033 - time (sec): 35.34 - samples/sec: 2352.31 - lr: 0.000025 - momentum: 0.000000 2023-10-17 21:12:58,205 epoch 1 - iter 882/1476 - loss 0.74180713 - time (sec): 42.39 - samples/sec: 2351.66 - lr: 0.000030 - momentum: 0.000000 2023-10-17 21:13:05,428 epoch 1 - iter 1029/1476 - loss 0.66397153 - time (sec): 49.61 - samples/sec: 2348.85 - lr: 0.000035 - momentum: 0.000000 2023-10-17 21:13:12,269 epoch 1 - iter 1176/1476 - loss 0.60399966 - time (sec): 56.45 - samples/sec: 2341.54 - lr: 0.000040 - momentum: 0.000000 2023-10-17 21:13:19,290 epoch 1 - iter 1323/1476 - loss 0.55840279 - time (sec): 63.47 - samples/sec: 2347.10 - lr: 0.000045 - momentum: 0.000000 2023-10-17 21:13:26,560 epoch 1 - iter 1470/1476 - loss 0.52004827 - time (sec): 70.74 - samples/sec: 2342.55 - lr: 0.000050 - momentum: 0.000000 2023-10-17 21:13:26,880 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:13:26,880 EPOCH 1 done: loss 0.5189 - lr: 0.000050 2023-10-17 21:13:33,183 DEV : loss 0.15781794488430023 - f1-score (micro avg) 0.7134 2023-10-17 21:13:33,228 saving best model 2023-10-17 21:13:33,643 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:13:40,926 epoch 2 - iter 147/1476 - loss 0.13618022 - time (sec): 7.28 - samples/sec: 2059.47 - lr: 0.000049 - momentum: 0.000000 2023-10-17 21:13:48,382 epoch 2 - iter 294/1476 - loss 0.14828668 - time (sec): 14.74 - samples/sec: 2265.73 - lr: 0.000049 - momentum: 0.000000 2023-10-17 21:13:55,425 epoch 2 - iter 441/1476 - loss 0.15814149 - time (sec): 21.78 - samples/sec: 2319.49 - lr: 0.000048 - momentum: 0.000000 2023-10-17 21:14:02,701 epoch 2 - iter 588/1476 - loss 0.15346701 - time (sec): 29.06 - samples/sec: 2360.96 - lr: 0.000048 - momentum: 0.000000 2023-10-17 21:14:09,918 epoch 2 - iter 735/1476 - loss 0.15116044 - time (sec): 36.27 - samples/sec: 2382.74 - lr: 0.000047 - momentum: 0.000000 2023-10-17 21:14:16,991 epoch 2 - iter 882/1476 - loss 0.14674645 - time (sec): 43.35 - samples/sec: 2377.52 - lr: 0.000047 - momentum: 0.000000 2023-10-17 21:14:23,920 epoch 2 - iter 1029/1476 - loss 0.14578389 - time (sec): 50.27 - samples/sec: 2352.17 - lr: 0.000046 - momentum: 0.000000 2023-10-17 21:14:30,697 epoch 2 - iter 1176/1476 - loss 0.14549586 - time (sec): 57.05 - samples/sec: 2339.49 - lr: 0.000046 - momentum: 0.000000 2023-10-17 21:14:37,676 epoch 2 - iter 1323/1476 - loss 0.14491984 - time (sec): 64.03 - samples/sec: 2336.70 - lr: 0.000045 - momentum: 0.000000 2023-10-17 21:14:44,608 epoch 2 - iter 1470/1476 - loss 0.14348353 - time (sec): 70.96 - samples/sec: 2336.48 - lr: 0.000044 - momentum: 0.000000 2023-10-17 21:14:44,869 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:14:44,870 EPOCH 2 done: loss 0.1433 - lr: 0.000044 2023-10-17 21:14:56,908 DEV : loss 0.13238579034805298 - f1-score (micro avg) 0.8057 2023-10-17 21:14:56,944 saving best model 2023-10-17 21:14:57,416 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:15:04,234 epoch 3 - iter 147/1476 - loss 0.09099243 - time (sec): 6.81 - samples/sec: 2221.06 - lr: 0.000044 - momentum: 0.000000 2023-10-17 21:15:10,879 epoch 3 - iter 294/1476 - loss 0.09767210 - time (sec): 13.46 - samples/sec: 2276.72 - lr: 0.000043 - momentum: 0.000000 2023-10-17 21:15:17,857 epoch 3 - iter 441/1476 - loss 0.09619969 - time (sec): 20.43 - samples/sec: 2337.51 - lr: 0.000043 - momentum: 0.000000 2023-10-17 21:15:25,059 epoch 3 - iter 588/1476 - loss 0.09884334 - time (sec): 27.64 - samples/sec: 2358.50 - lr: 0.000042 - momentum: 0.000000 2023-10-17 21:15:32,052 epoch 3 - iter 735/1476 - loss 0.09130138 - time (sec): 34.63 - samples/sec: 2329.09 - lr: 0.000042 - momentum: 0.000000 2023-10-17 21:15:39,127 epoch 3 - iter 882/1476 - loss 0.09409109 - time (sec): 41.70 - samples/sec: 2324.32 - lr: 0.000041 - momentum: 0.000000 2023-10-17 21:15:46,434 epoch 3 - iter 1029/1476 - loss 0.09210907 - time (sec): 49.01 - samples/sec: 2359.09 - lr: 0.000041 - momentum: 0.000000 2023-10-17 21:15:53,396 epoch 3 - iter 1176/1476 - loss 0.09301213 - time (sec): 55.97 - samples/sec: 2370.49 - lr: 0.000040 - momentum: 0.000000 2023-10-17 21:16:00,199 epoch 3 - iter 1323/1476 - loss 0.09078236 - time (sec): 62.78 - samples/sec: 2380.47 - lr: 0.000039 - momentum: 0.000000 2023-10-17 21:16:07,142 epoch 3 - iter 1470/1476 - loss 0.09325701 - time (sec): 69.72 - samples/sec: 2378.16 - lr: 0.000039 - momentum: 0.000000 2023-10-17 21:16:07,408 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:16:07,408 EPOCH 3 done: loss 0.0935 - lr: 0.000039 2023-10-17 21:16:18,751 DEV : loss 0.17908549308776855 - f1-score (micro avg) 0.8022 2023-10-17 21:16:18,780 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:16:26,621 epoch 4 - iter 147/1476 - loss 0.05955487 - time (sec): 7.84 - samples/sec: 2174.55 - lr: 0.000038 - momentum: 0.000000 2023-10-17 21:16:34,138 epoch 4 - iter 294/1476 - loss 0.06464650 - time (sec): 15.36 - samples/sec: 2300.33 - lr: 0.000038 - momentum: 0.000000 2023-10-17 21:16:41,169 epoch 4 - iter 441/1476 - loss 0.06441669 - time (sec): 22.39 - samples/sec: 2318.35 - lr: 0.000037 - momentum: 0.000000 2023-10-17 21:16:48,222 epoch 4 - iter 588/1476 - loss 0.06965834 - time (sec): 29.44 - samples/sec: 2329.10 - lr: 0.000037 - momentum: 0.000000 2023-10-17 21:16:55,135 epoch 4 - iter 735/1476 - loss 0.07027131 - time (sec): 36.35 - samples/sec: 2311.10 - lr: 0.000036 - momentum: 0.000000 2023-10-17 21:17:02,168 epoch 4 - iter 882/1476 - loss 0.06940446 - time (sec): 43.39 - samples/sec: 2315.69 - lr: 0.000036 - momentum: 0.000000 2023-10-17 21:17:08,988 epoch 4 - iter 1029/1476 - loss 0.06906629 - time (sec): 50.21 - samples/sec: 2311.07 - lr: 0.000035 - momentum: 0.000000 2023-10-17 21:17:16,069 epoch 4 - iter 1176/1476 - loss 0.06960608 - time (sec): 57.29 - samples/sec: 2318.23 - lr: 0.000034 - momentum: 0.000000 2023-10-17 21:17:23,415 epoch 4 - iter 1323/1476 - loss 0.06848630 - time (sec): 64.63 - samples/sec: 2335.06 - lr: 0.000034 - momentum: 0.000000 2023-10-17 21:17:30,550 epoch 4 - iter 1470/1476 - loss 0.06867505 - time (sec): 71.77 - samples/sec: 2310.89 - lr: 0.000033 - momentum: 0.000000 2023-10-17 21:17:30,826 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:17:30,827 EPOCH 4 done: loss 0.0686 - lr: 0.000033 2023-10-17 21:17:42,009 DEV : loss 0.17883314192295074 - f1-score (micro avg) 0.8315 2023-10-17 21:17:42,038 saving best model 2023-10-17 21:17:42,543 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:17:49,944 epoch 5 - iter 147/1476 - loss 0.04503414 - time (sec): 7.40 - samples/sec: 2375.63 - lr: 0.000033 - momentum: 0.000000 2023-10-17 21:17:57,061 epoch 5 - iter 294/1476 - loss 0.04077908 - time (sec): 14.51 - samples/sec: 2363.98 - lr: 0.000032 - momentum: 0.000000 2023-10-17 21:18:04,459 epoch 5 - iter 441/1476 - loss 0.04485968 - time (sec): 21.91 - samples/sec: 2394.21 - lr: 0.000032 - momentum: 0.000000 2023-10-17 21:18:11,546 epoch 5 - iter 588/1476 - loss 0.04489000 - time (sec): 29.00 - samples/sec: 2379.88 - lr: 0.000031 - momentum: 0.000000 2023-10-17 21:18:18,929 epoch 5 - iter 735/1476 - loss 0.04378441 - time (sec): 36.38 - samples/sec: 2359.70 - lr: 0.000031 - momentum: 0.000000 2023-10-17 21:18:26,184 epoch 5 - iter 882/1476 - loss 0.04187352 - time (sec): 43.64 - samples/sec: 2361.29 - lr: 0.000030 - momentum: 0.000000 2023-10-17 21:18:33,052 epoch 5 - iter 1029/1476 - loss 0.04478438 - time (sec): 50.50 - samples/sec: 2331.91 - lr: 0.000029 - momentum: 0.000000 2023-10-17 21:18:40,065 epoch 5 - iter 1176/1476 - loss 0.04667311 - time (sec): 57.52 - samples/sec: 2309.59 - lr: 0.000029 - momentum: 0.000000 2023-10-17 21:18:47,060 epoch 5 - iter 1323/1476 - loss 0.04707730 - time (sec): 64.51 - samples/sec: 2314.55 - lr: 0.000028 - momentum: 0.000000 2023-10-17 21:18:54,289 epoch 5 - iter 1470/1476 - loss 0.04804846 - time (sec): 71.74 - samples/sec: 2312.31 - lr: 0.000028 - momentum: 0.000000 2023-10-17 21:18:54,581 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:18:54,581 EPOCH 5 done: loss 0.0479 - lr: 0.000028 2023-10-17 21:19:05,740 DEV : loss 0.1932452768087387 - f1-score (micro avg) 0.8216 2023-10-17 21:19:05,770 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:19:13,041 epoch 6 - iter 147/1476 - loss 0.02692959 - time (sec): 7.27 - samples/sec: 2251.36 - lr: 0.000027 - momentum: 0.000000 2023-10-17 21:19:20,095 epoch 6 - iter 294/1476 - loss 0.03039685 - time (sec): 14.32 - samples/sec: 2258.57 - lr: 0.000027 - momentum: 0.000000 2023-10-17 21:19:26,924 epoch 6 - iter 441/1476 - loss 0.02969535 - time (sec): 21.15 - samples/sec: 2240.64 - lr: 0.000026 - momentum: 0.000000 2023-10-17 21:19:34,018 epoch 6 - iter 588/1476 - loss 0.02976128 - time (sec): 28.25 - samples/sec: 2269.04 - lr: 0.000026 - momentum: 0.000000 2023-10-17 21:19:41,263 epoch 6 - iter 735/1476 - loss 0.03299645 - time (sec): 35.49 - samples/sec: 2272.63 - lr: 0.000025 - momentum: 0.000000 2023-10-17 21:19:48,189 epoch 6 - iter 882/1476 - loss 0.03080088 - time (sec): 42.42 - samples/sec: 2277.34 - lr: 0.000024 - momentum: 0.000000 2023-10-17 21:19:55,156 epoch 6 - iter 1029/1476 - loss 0.03111720 - time (sec): 49.39 - samples/sec: 2274.80 - lr: 0.000024 - momentum: 0.000000 2023-10-17 21:20:02,179 epoch 6 - iter 1176/1476 - loss 0.03136825 - time (sec): 56.41 - samples/sec: 2267.49 - lr: 0.000023 - momentum: 0.000000 2023-10-17 21:20:09,961 epoch 6 - iter 1323/1476 - loss 0.03283562 - time (sec): 64.19 - samples/sec: 2315.08 - lr: 0.000023 - momentum: 0.000000 2023-10-17 21:20:17,596 epoch 6 - iter 1470/1476 - loss 0.03154629 - time (sec): 71.82 - samples/sec: 2298.45 - lr: 0.000022 - momentum: 0.000000 2023-10-17 21:20:18,023 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:20:18,023 EPOCH 6 done: loss 0.0318 - lr: 0.000022 2023-10-17 21:20:29,225 DEV : loss 0.20342403650283813 - f1-score (micro avg) 0.829 2023-10-17 21:20:29,256 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:20:36,413 epoch 7 - iter 147/1476 - loss 0.02891200 - time (sec): 7.16 - samples/sec: 2142.34 - lr: 0.000022 - momentum: 0.000000 2023-10-17 21:20:43,330 epoch 7 - iter 294/1476 - loss 0.02394202 - time (sec): 14.07 - samples/sec: 2229.98 - lr: 0.000021 - momentum: 0.000000 2023-10-17 21:20:50,316 epoch 7 - iter 441/1476 - loss 0.02015639 - time (sec): 21.06 - samples/sec: 2184.29 - lr: 0.000021 - momentum: 0.000000 2023-10-17 21:20:57,474 epoch 7 - iter 588/1476 - loss 0.02007236 - time (sec): 28.22 - samples/sec: 2211.82 - lr: 0.000020 - momentum: 0.000000 2023-10-17 21:21:04,503 epoch 7 - iter 735/1476 - loss 0.02231409 - time (sec): 35.25 - samples/sec: 2232.96 - lr: 0.000019 - momentum: 0.000000 2023-10-17 21:21:11,792 epoch 7 - iter 882/1476 - loss 0.02414592 - time (sec): 42.54 - samples/sec: 2270.14 - lr: 0.000019 - momentum: 0.000000 2023-10-17 21:21:19,804 epoch 7 - iter 1029/1476 - loss 0.02619634 - time (sec): 50.55 - samples/sec: 2320.73 - lr: 0.000018 - momentum: 0.000000 2023-10-17 21:21:27,164 epoch 7 - iter 1176/1476 - loss 0.02513161 - time (sec): 57.91 - samples/sec: 2299.35 - lr: 0.000018 - momentum: 0.000000 2023-10-17 21:21:34,416 epoch 7 - iter 1323/1476 - loss 0.02524479 - time (sec): 65.16 - samples/sec: 2297.30 - lr: 0.000017 - momentum: 0.000000 2023-10-17 21:21:41,691 epoch 7 - iter 1470/1476 - loss 0.02525241 - time (sec): 72.43 - samples/sec: 2292.05 - lr: 0.000017 - momentum: 0.000000 2023-10-17 21:21:41,978 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:21:41,978 EPOCH 7 done: loss 0.0252 - lr: 0.000017 2023-10-17 21:21:53,349 DEV : loss 0.2042791247367859 - f1-score (micro avg) 0.8458 2023-10-17 21:21:53,382 saving best model 2023-10-17 21:21:53,867 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:22:00,762 epoch 8 - iter 147/1476 - loss 0.00937919 - time (sec): 6.89 - samples/sec: 2283.57 - lr: 0.000016 - momentum: 0.000000 2023-10-17 21:22:07,881 epoch 8 - iter 294/1476 - loss 0.00853811 - time (sec): 14.01 - samples/sec: 2258.90 - lr: 0.000016 - momentum: 0.000000 2023-10-17 21:22:14,881 epoch 8 - iter 441/1476 - loss 0.01070567 - time (sec): 21.01 - samples/sec: 2269.68 - lr: 0.000015 - momentum: 0.000000 2023-10-17 21:22:22,036 epoch 8 - iter 588/1476 - loss 0.01179930 - time (sec): 28.17 - samples/sec: 2246.93 - lr: 0.000014 - momentum: 0.000000 2023-10-17 21:22:30,116 epoch 8 - iter 735/1476 - loss 0.01355360 - time (sec): 36.24 - samples/sec: 2315.08 - lr: 0.000014 - momentum: 0.000000 2023-10-17 21:22:37,569 epoch 8 - iter 882/1476 - loss 0.01288325 - time (sec): 43.70 - samples/sec: 2340.72 - lr: 0.000013 - momentum: 0.000000 2023-10-17 21:22:44,608 epoch 8 - iter 1029/1476 - loss 0.01292629 - time (sec): 50.74 - samples/sec: 2330.05 - lr: 0.000013 - momentum: 0.000000 2023-10-17 21:22:51,814 epoch 8 - iter 1176/1476 - loss 0.01385110 - time (sec): 57.94 - samples/sec: 2329.28 - lr: 0.000012 - momentum: 0.000000 2023-10-17 21:22:58,993 epoch 8 - iter 1323/1476 - loss 0.01415473 - time (sec): 65.12 - samples/sec: 2309.57 - lr: 0.000012 - momentum: 0.000000 2023-10-17 21:23:05,874 epoch 8 - iter 1470/1476 - loss 0.01408906 - time (sec): 72.00 - samples/sec: 2299.77 - lr: 0.000011 - momentum: 0.000000 2023-10-17 21:23:06,189 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:23:06,190 EPOCH 8 done: loss 0.0140 - lr: 0.000011 2023-10-17 21:23:17,357 DEV : loss 0.23180881142616272 - f1-score (micro avg) 0.8352 2023-10-17 21:23:17,387 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:23:24,663 epoch 9 - iter 147/1476 - loss 0.01659959 - time (sec): 7.27 - samples/sec: 2475.68 - lr: 0.000011 - momentum: 0.000000 2023-10-17 21:23:31,742 epoch 9 - iter 294/1476 - loss 0.01094948 - time (sec): 14.35 - samples/sec: 2413.12 - lr: 0.000010 - momentum: 0.000000 2023-10-17 21:23:38,527 epoch 9 - iter 441/1476 - loss 0.00931768 - time (sec): 21.14 - samples/sec: 2376.05 - lr: 0.000009 - momentum: 0.000000 2023-10-17 21:23:45,851 epoch 9 - iter 588/1476 - loss 0.00991294 - time (sec): 28.46 - samples/sec: 2373.51 - lr: 0.000009 - momentum: 0.000000 2023-10-17 21:23:52,800 epoch 9 - iter 735/1476 - loss 0.01019933 - time (sec): 35.41 - samples/sec: 2348.46 - lr: 0.000008 - momentum: 0.000000 2023-10-17 21:24:00,113 epoch 9 - iter 882/1476 - loss 0.01047522 - time (sec): 42.72 - samples/sec: 2354.51 - lr: 0.000008 - momentum: 0.000000 2023-10-17 21:24:07,361 epoch 9 - iter 1029/1476 - loss 0.01030206 - time (sec): 49.97 - samples/sec: 2347.60 - lr: 0.000007 - momentum: 0.000000 2023-10-17 21:24:14,651 epoch 9 - iter 1176/1476 - loss 0.01068040 - time (sec): 57.26 - samples/sec: 2353.18 - lr: 0.000007 - momentum: 0.000000 2023-10-17 21:24:21,700 epoch 9 - iter 1323/1476 - loss 0.01029783 - time (sec): 64.31 - samples/sec: 2344.88 - lr: 0.000006 - momentum: 0.000000 2023-10-17 21:24:28,634 epoch 9 - iter 1470/1476 - loss 0.00979050 - time (sec): 71.25 - samples/sec: 2328.95 - lr: 0.000006 - momentum: 0.000000 2023-10-17 21:24:28,905 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:24:28,905 EPOCH 9 done: loss 0.0098 - lr: 0.000006 2023-10-17 21:24:40,143 DEV : loss 0.2325468510389328 - f1-score (micro avg) 0.8472 2023-10-17 21:24:40,174 saving best model 2023-10-17 21:24:40,670 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:24:47,921 epoch 10 - iter 147/1476 - loss 0.00447119 - time (sec): 7.25 - samples/sec: 2284.31 - lr: 0.000005 - momentum: 0.000000 2023-10-17 21:24:55,320 epoch 10 - iter 294/1476 - loss 0.00607937 - time (sec): 14.64 - samples/sec: 2399.71 - lr: 0.000004 - momentum: 0.000000 2023-10-17 21:25:02,482 epoch 10 - iter 441/1476 - loss 0.00594049 - time (sec): 21.81 - samples/sec: 2332.13 - lr: 0.000004 - momentum: 0.000000 2023-10-17 21:25:09,895 epoch 10 - iter 588/1476 - loss 0.00571948 - time (sec): 29.22 - samples/sec: 2316.57 - lr: 0.000003 - momentum: 0.000000 2023-10-17 21:25:17,081 epoch 10 - iter 735/1476 - loss 0.00519373 - time (sec): 36.41 - samples/sec: 2296.15 - lr: 0.000003 - momentum: 0.000000 2023-10-17 21:25:23,922 epoch 10 - iter 882/1476 - loss 0.00678593 - time (sec): 43.25 - samples/sec: 2294.83 - lr: 0.000002 - momentum: 0.000000 2023-10-17 21:25:31,170 epoch 10 - iter 1029/1476 - loss 0.00637890 - time (sec): 50.49 - samples/sec: 2285.07 - lr: 0.000002 - momentum: 0.000000 2023-10-17 21:25:38,140 epoch 10 - iter 1176/1476 - loss 0.00583236 - time (sec): 57.47 - samples/sec: 2296.51 - lr: 0.000001 - momentum: 0.000000 2023-10-17 21:25:45,142 epoch 10 - iter 1323/1476 - loss 0.00563599 - time (sec): 64.47 - samples/sec: 2301.07 - lr: 0.000001 - momentum: 0.000000 2023-10-17 21:25:52,319 epoch 10 - iter 1470/1476 - loss 0.00529100 - time (sec): 71.64 - samples/sec: 2315.70 - lr: 0.000000 - momentum: 0.000000 2023-10-17 21:25:52,584 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:25:52,584 EPOCH 10 done: loss 0.0053 - lr: 0.000000 2023-10-17 21:26:03,720 DEV : loss 0.2309638112783432 - f1-score (micro avg) 0.8473 2023-10-17 21:26:03,750 saving best model 2023-10-17 21:26:04,626 ---------------------------------------------------------------------------------------------------- 2023-10-17 21:26:04,627 Loading model from best epoch ... 2023-10-17 21:26:05,984 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-time, B-time, E-time, I-time, S-prod, B-prod, E-prod, I-prod 2023-10-17 21:26:12,598 Results: - F-score (micro) 0.805 - F-score (macro) 0.7103 - Accuracy 0.6948 By class: precision recall f1-score support loc 0.8445 0.8800 0.8619 858 pers 0.7666 0.8194 0.7921 537 org 0.6532 0.6136 0.6328 132 prod 0.6885 0.6885 0.6885 61 time 0.5312 0.6296 0.5763 54 micro avg 0.7874 0.8234 0.8050 1642 macro avg 0.6968 0.7262 0.7103 1642 weighted avg 0.7875 0.8234 0.8048 1642 2023-10-17 21:26:12,598 ----------------------------------------------------------------------------------------------------