2023-10-16 08:59:27,927 ---------------------------------------------------------------------------------------------------- 2023-10-16 08:59:27,928 Model: "SequenceTagger( (embeddings): TransformerWordEmbeddings( (model): BertModel( (embeddings): BertEmbeddings( (word_embeddings): Embedding(32001, 768) (position_embeddings): Embedding(512, 768) (token_type_embeddings): Embedding(2, 768) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) (encoder): BertEncoder( (layer): ModuleList( (0-11): 12 x BertLayer( (attention): BertAttention( (self): BertSelfAttention( (query): Linear(in_features=768, out_features=768, bias=True) (key): Linear(in_features=768, out_features=768, bias=True) (value): Linear(in_features=768, out_features=768, bias=True) (dropout): Dropout(p=0.1, inplace=False) ) (output): BertSelfOutput( (dense): Linear(in_features=768, out_features=768, bias=True) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) (intermediate): BertIntermediate( (dense): Linear(in_features=768, out_features=3072, bias=True) (intermediate_act_fn): GELUActivation() ) (output): BertOutput( (dense): Linear(in_features=3072, out_features=768, bias=True) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) ) (pooler): BertPooler( (dense): Linear(in_features=768, out_features=768, bias=True) (activation): Tanh() ) ) ) (locked_dropout): LockedDropout(p=0.5) (linear): Linear(in_features=768, out_features=17, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-16 08:59:27,928 ---------------------------------------------------------------------------------------------------- 2023-10-16 08:59:27,928 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 2023-10-16 08:59:27,928 ---------------------------------------------------------------------------------------------------- 2023-10-16 08:59:27,928 Train: 7142 sentences 2023-10-16 08:59:27,928 (train_with_dev=False, train_with_test=False) 2023-10-16 08:59:27,928 ---------------------------------------------------------------------------------------------------- 2023-10-16 08:59:27,928 Training Params: 2023-10-16 08:59:27,928 - learning_rate: "5e-05" 2023-10-16 08:59:27,928 - mini_batch_size: "4" 2023-10-16 08:59:27,928 - max_epochs: "10" 2023-10-16 08:59:27,928 - shuffle: "True" 2023-10-16 08:59:27,929 ---------------------------------------------------------------------------------------------------- 2023-10-16 08:59:27,929 Plugins: 2023-10-16 08:59:27,929 - LinearScheduler | warmup_fraction: '0.1' 2023-10-16 08:59:27,929 ---------------------------------------------------------------------------------------------------- 2023-10-16 08:59:27,929 Final evaluation on model from best epoch (best-model.pt) 2023-10-16 08:59:27,929 - metric: "('micro avg', 'f1-score')" 2023-10-16 08:59:27,929 ---------------------------------------------------------------------------------------------------- 2023-10-16 08:59:27,929 Computation: 2023-10-16 08:59:27,929 - compute on device: cuda:0 2023-10-16 08:59:27,929 - embedding storage: none 2023-10-16 08:59:27,929 ---------------------------------------------------------------------------------------------------- 2023-10-16 08:59:27,929 Model training base path: "hmbench-newseye/fr-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1" 2023-10-16 08:59:27,929 ---------------------------------------------------------------------------------------------------- 2023-10-16 08:59:27,929 ---------------------------------------------------------------------------------------------------- 2023-10-16 08:59:36,701 epoch 1 - iter 178/1786 - loss 1.93950853 - time (sec): 8.77 - samples/sec: 2806.48 - lr: 0.000005 - momentum: 0.000000 2023-10-16 08:59:45,303 epoch 1 - iter 356/1786 - loss 1.19684096 - time (sec): 17.37 - samples/sec: 2873.23 - lr: 0.000010 - momentum: 0.000000 2023-10-16 08:59:53,918 epoch 1 - iter 534/1786 - loss 0.90350754 - time (sec): 25.99 - samples/sec: 2868.02 - lr: 0.000015 - momentum: 0.000000 2023-10-16 09:00:02,649 epoch 1 - iter 712/1786 - loss 0.73207711 - time (sec): 34.72 - samples/sec: 2899.46 - lr: 0.000020 - momentum: 0.000000 2023-10-16 09:00:11,223 epoch 1 - iter 890/1786 - loss 0.63211784 - time (sec): 43.29 - samples/sec: 2877.60 - lr: 0.000025 - momentum: 0.000000 2023-10-16 09:00:19,866 epoch 1 - iter 1068/1786 - loss 0.55770395 - time (sec): 51.94 - samples/sec: 2860.36 - lr: 0.000030 - momentum: 0.000000 2023-10-16 09:00:28,458 epoch 1 - iter 1246/1786 - loss 0.50479715 - time (sec): 60.53 - samples/sec: 2864.56 - lr: 0.000035 - momentum: 0.000000 2023-10-16 09:00:37,259 epoch 1 - iter 1424/1786 - loss 0.45844199 - time (sec): 69.33 - samples/sec: 2856.58 - lr: 0.000040 - momentum: 0.000000 2023-10-16 09:00:46,035 epoch 1 - iter 1602/1786 - loss 0.42492404 - time (sec): 78.11 - samples/sec: 2872.10 - lr: 0.000045 - momentum: 0.000000 2023-10-16 09:00:54,534 epoch 1 - iter 1780/1786 - loss 0.40035384 - time (sec): 86.60 - samples/sec: 2862.69 - lr: 0.000050 - momentum: 0.000000 2023-10-16 09:00:54,814 ---------------------------------------------------------------------------------------------------- 2023-10-16 09:00:54,815 EPOCH 1 done: loss 0.3994 - lr: 0.000050 2023-10-16 09:00:57,828 DEV : loss 0.164632648229599 - f1-score (micro avg) 0.6901 2023-10-16 09:00:57,844 saving best model 2023-10-16 09:00:58,206 ---------------------------------------------------------------------------------------------------- 2023-10-16 09:01:06,863 epoch 2 - iter 178/1786 - loss 0.11487137 - time (sec): 8.66 - samples/sec: 2797.96 - lr: 0.000049 - momentum: 0.000000 2023-10-16 09:01:15,536 epoch 2 - iter 356/1786 - loss 0.10865483 - time (sec): 17.33 - samples/sec: 2865.27 - lr: 0.000049 - momentum: 0.000000 2023-10-16 09:01:24,138 epoch 2 - iter 534/1786 - loss 0.11685553 - time (sec): 25.93 - samples/sec: 2839.41 - lr: 0.000048 - momentum: 0.000000 2023-10-16 09:01:32,735 epoch 2 - iter 712/1786 - loss 0.11885426 - time (sec): 34.53 - samples/sec: 2838.78 - lr: 0.000048 - momentum: 0.000000 2023-10-16 09:01:41,455 epoch 2 - iter 890/1786 - loss 0.11920458 - time (sec): 43.25 - samples/sec: 2830.37 - lr: 0.000047 - momentum: 0.000000 2023-10-16 09:01:50,204 epoch 2 - iter 1068/1786 - loss 0.11979459 - time (sec): 52.00 - samples/sec: 2832.05 - lr: 0.000047 - momentum: 0.000000 2023-10-16 09:01:58,827 epoch 2 - iter 1246/1786 - loss 0.12470555 - time (sec): 60.62 - samples/sec: 2847.73 - lr: 0.000046 - momentum: 0.000000 2023-10-16 09:02:07,725 epoch 2 - iter 1424/1786 - loss 0.12324524 - time (sec): 69.52 - samples/sec: 2848.49 - lr: 0.000046 - momentum: 0.000000 2023-10-16 09:02:16,897 epoch 2 - iter 1602/1786 - loss 0.12369002 - time (sec): 78.69 - samples/sec: 2834.51 - lr: 0.000045 - momentum: 0.000000 2023-10-16 09:02:25,680 epoch 2 - iter 1780/1786 - loss 0.12238966 - time (sec): 87.47 - samples/sec: 2833.26 - lr: 0.000044 - momentum: 0.000000 2023-10-16 09:02:25,944 ---------------------------------------------------------------------------------------------------- 2023-10-16 09:02:25,944 EPOCH 2 done: loss 0.1228 - lr: 0.000044 2023-10-16 09:02:30,616 DEV : loss 0.1199851781129837 - f1-score (micro avg) 0.7637 2023-10-16 09:02:30,633 saving best model 2023-10-16 09:02:31,073 ---------------------------------------------------------------------------------------------------- 2023-10-16 09:02:39,971 epoch 3 - iter 178/1786 - loss 0.08422764 - time (sec): 8.90 - samples/sec: 2650.63 - lr: 0.000044 - momentum: 0.000000 2023-10-16 09:02:48,902 epoch 3 - iter 356/1786 - loss 0.08832799 - time (sec): 17.83 - samples/sec: 2666.78 - lr: 0.000043 - momentum: 0.000000 2023-10-16 09:02:57,286 epoch 3 - iter 534/1786 - loss 0.08603336 - time (sec): 26.21 - samples/sec: 2720.96 - lr: 0.000043 - momentum: 0.000000 2023-10-16 09:03:06,353 epoch 3 - iter 712/1786 - loss 0.08439697 - time (sec): 35.28 - samples/sec: 2768.02 - lr: 0.000042 - momentum: 0.000000 2023-10-16 09:03:15,079 epoch 3 - iter 890/1786 - loss 0.08403027 - time (sec): 44.00 - samples/sec: 2776.45 - lr: 0.000042 - momentum: 0.000000 2023-10-16 09:03:23,832 epoch 3 - iter 1068/1786 - loss 0.08339811 - time (sec): 52.76 - samples/sec: 2796.66 - lr: 0.000041 - momentum: 0.000000 2023-10-16 09:03:32,601 epoch 3 - iter 1246/1786 - loss 0.08366626 - time (sec): 61.53 - samples/sec: 2796.00 - lr: 0.000041 - momentum: 0.000000 2023-10-16 09:03:41,648 epoch 3 - iter 1424/1786 - loss 0.08520626 - time (sec): 70.57 - samples/sec: 2811.10 - lr: 0.000040 - momentum: 0.000000 2023-10-16 09:03:50,239 epoch 3 - iter 1602/1786 - loss 0.08861563 - time (sec): 79.16 - samples/sec: 2795.22 - lr: 0.000039 - momentum: 0.000000 2023-10-16 09:03:59,241 epoch 3 - iter 1780/1786 - loss 0.08777026 - time (sec): 88.17 - samples/sec: 2814.81 - lr: 0.000039 - momentum: 0.000000 2023-10-16 09:03:59,514 ---------------------------------------------------------------------------------------------------- 2023-10-16 09:03:59,514 EPOCH 3 done: loss 0.0876 - lr: 0.000039 2023-10-16 09:04:03,562 DEV : loss 0.12482591718435287 - f1-score (micro avg) 0.7656 2023-10-16 09:04:03,577 saving best model 2023-10-16 09:04:04,030 ---------------------------------------------------------------------------------------------------- 2023-10-16 09:04:12,850 epoch 4 - iter 178/1786 - loss 0.07490226 - time (sec): 8.82 - samples/sec: 2826.41 - lr: 0.000038 - momentum: 0.000000 2023-10-16 09:04:21,987 epoch 4 - iter 356/1786 - loss 0.07388713 - time (sec): 17.96 - samples/sec: 2819.39 - lr: 0.000038 - momentum: 0.000000 2023-10-16 09:04:30,657 epoch 4 - iter 534/1786 - loss 0.07199900 - time (sec): 26.63 - samples/sec: 2796.78 - lr: 0.000037 - momentum: 0.000000 2023-10-16 09:04:39,443 epoch 4 - iter 712/1786 - loss 0.06917859 - time (sec): 35.41 - samples/sec: 2848.90 - lr: 0.000037 - momentum: 0.000000 2023-10-16 09:04:48,062 epoch 4 - iter 890/1786 - loss 0.06579901 - time (sec): 44.03 - samples/sec: 2843.02 - lr: 0.000036 - momentum: 0.000000 2023-10-16 09:04:57,041 epoch 4 - iter 1068/1786 - loss 0.06377113 - time (sec): 53.01 - samples/sec: 2843.47 - lr: 0.000036 - momentum: 0.000000 2023-10-16 09:05:05,388 epoch 4 - iter 1246/1786 - loss 0.06597385 - time (sec): 61.36 - samples/sec: 2819.29 - lr: 0.000035 - momentum: 0.000000 2023-10-16 09:05:14,250 epoch 4 - iter 1424/1786 - loss 0.06743665 - time (sec): 70.22 - samples/sec: 2821.07 - lr: 0.000034 - momentum: 0.000000 2023-10-16 09:05:23,540 epoch 4 - iter 1602/1786 - loss 0.06657675 - time (sec): 79.51 - samples/sec: 2802.30 - lr: 0.000034 - momentum: 0.000000 2023-10-16 09:05:32,491 epoch 4 - iter 1780/1786 - loss 0.06664923 - time (sec): 88.46 - samples/sec: 2801.56 - lr: 0.000033 - momentum: 0.000000 2023-10-16 09:05:32,817 ---------------------------------------------------------------------------------------------------- 2023-10-16 09:05:32,817 EPOCH 4 done: loss 0.0669 - lr: 0.000033 2023-10-16 09:05:36,817 DEV : loss 0.17001311480998993 - f1-score (micro avg) 0.7773 2023-10-16 09:05:36,833 saving best model 2023-10-16 09:05:37,282 ---------------------------------------------------------------------------------------------------- 2023-10-16 09:05:46,244 epoch 5 - iter 178/1786 - loss 0.05321277 - time (sec): 8.96 - samples/sec: 2962.54 - lr: 0.000033 - momentum: 0.000000 2023-10-16 09:05:55,007 epoch 5 - iter 356/1786 - loss 0.05027400 - time (sec): 17.72 - samples/sec: 2914.67 - lr: 0.000032 - momentum: 0.000000 2023-10-16 09:06:03,991 epoch 5 - iter 534/1786 - loss 0.05055258 - time (sec): 26.70 - samples/sec: 2883.74 - lr: 0.000032 - momentum: 0.000000 2023-10-16 09:06:12,573 epoch 5 - iter 712/1786 - loss 0.04836095 - time (sec): 35.29 - samples/sec: 2930.97 - lr: 0.000031 - momentum: 0.000000 2023-10-16 09:06:21,516 epoch 5 - iter 890/1786 - loss 0.04806037 - time (sec): 44.23 - samples/sec: 2907.81 - lr: 0.000031 - momentum: 0.000000 2023-10-16 09:06:30,294 epoch 5 - iter 1068/1786 - loss 0.04703404 - time (sec): 53.01 - samples/sec: 2880.93 - lr: 0.000030 - momentum: 0.000000 2023-10-16 09:06:38,968 epoch 5 - iter 1246/1786 - loss 0.04922091 - time (sec): 61.68 - samples/sec: 2876.27 - lr: 0.000029 - momentum: 0.000000 2023-10-16 09:06:47,627 epoch 5 - iter 1424/1786 - loss 0.05007639 - time (sec): 70.34 - samples/sec: 2846.92 - lr: 0.000029 - momentum: 0.000000 2023-10-16 09:06:56,313 epoch 5 - iter 1602/1786 - loss 0.05063967 - time (sec): 79.03 - samples/sec: 2824.96 - lr: 0.000028 - momentum: 0.000000 2023-10-16 09:07:04,982 epoch 5 - iter 1780/1786 - loss 0.05136262 - time (sec): 87.70 - samples/sec: 2821.84 - lr: 0.000028 - momentum: 0.000000 2023-10-16 09:07:05,332 ---------------------------------------------------------------------------------------------------- 2023-10-16 09:07:05,332 EPOCH 5 done: loss 0.0514 - lr: 0.000028 2023-10-16 09:07:09,890 DEV : loss 0.1479008048772812 - f1-score (micro avg) 0.7853 2023-10-16 09:07:09,906 saving best model 2023-10-16 09:07:10,355 ---------------------------------------------------------------------------------------------------- 2023-10-16 09:07:19,191 epoch 6 - iter 178/1786 - loss 0.04084234 - time (sec): 8.83 - samples/sec: 2787.69 - lr: 0.000027 - momentum: 0.000000 2023-10-16 09:07:27,857 epoch 6 - iter 356/1786 - loss 0.03988235 - time (sec): 17.50 - samples/sec: 2819.19 - lr: 0.000027 - momentum: 0.000000 2023-10-16 09:07:36,931 epoch 6 - iter 534/1786 - loss 0.04230182 - time (sec): 26.57 - samples/sec: 2812.14 - lr: 0.000026 - momentum: 0.000000 2023-10-16 09:07:45,917 epoch 6 - iter 712/1786 - loss 0.04315507 - time (sec): 35.56 - samples/sec: 2844.22 - lr: 0.000026 - momentum: 0.000000 2023-10-16 09:07:54,708 epoch 6 - iter 890/1786 - loss 0.03981327 - time (sec): 44.35 - samples/sec: 2841.28 - lr: 0.000025 - momentum: 0.000000 2023-10-16 09:08:03,319 epoch 6 - iter 1068/1786 - loss 0.04083548 - time (sec): 52.96 - samples/sec: 2847.09 - lr: 0.000024 - momentum: 0.000000 2023-10-16 09:08:12,027 epoch 6 - iter 1246/1786 - loss 0.03919229 - time (sec): 61.67 - samples/sec: 2864.30 - lr: 0.000024 - momentum: 0.000000 2023-10-16 09:08:20,916 epoch 6 - iter 1424/1786 - loss 0.03877892 - time (sec): 70.56 - samples/sec: 2859.52 - lr: 0.000023 - momentum: 0.000000 2023-10-16 09:08:29,628 epoch 6 - iter 1602/1786 - loss 0.03867605 - time (sec): 79.27 - samples/sec: 2826.83 - lr: 0.000023 - momentum: 0.000000 2023-10-16 09:08:38,418 epoch 6 - iter 1780/1786 - loss 0.03727970 - time (sec): 88.06 - samples/sec: 2818.19 - lr: 0.000022 - momentum: 0.000000 2023-10-16 09:08:38,691 ---------------------------------------------------------------------------------------------------- 2023-10-16 09:08:38,692 EPOCH 6 done: loss 0.0374 - lr: 0.000022 2023-10-16 09:08:42,752 DEV : loss 0.17757560312747955 - f1-score (micro avg) 0.7965 2023-10-16 09:08:42,768 saving best model 2023-10-16 09:08:43,226 ---------------------------------------------------------------------------------------------------- 2023-10-16 09:08:52,089 epoch 7 - iter 178/1786 - loss 0.02636464 - time (sec): 8.86 - samples/sec: 2939.18 - lr: 0.000022 - momentum: 0.000000 2023-10-16 09:09:00,968 epoch 7 - iter 356/1786 - loss 0.02439338 - time (sec): 17.74 - samples/sec: 2863.45 - lr: 0.000021 - momentum: 0.000000 2023-10-16 09:09:10,107 epoch 7 - iter 534/1786 - loss 0.02868364 - time (sec): 26.88 - samples/sec: 2854.02 - lr: 0.000021 - momentum: 0.000000 2023-10-16 09:09:19,060 epoch 7 - iter 712/1786 - loss 0.02820433 - time (sec): 35.83 - samples/sec: 2819.85 - lr: 0.000020 - momentum: 0.000000 2023-10-16 09:09:27,628 epoch 7 - iter 890/1786 - loss 0.02740708 - time (sec): 44.40 - samples/sec: 2785.70 - lr: 0.000019 - momentum: 0.000000 2023-10-16 09:09:36,262 epoch 7 - iter 1068/1786 - loss 0.02931505 - time (sec): 53.03 - samples/sec: 2806.18 - lr: 0.000019 - momentum: 0.000000 2023-10-16 09:09:45,216 epoch 7 - iter 1246/1786 - loss 0.03044116 - time (sec): 61.98 - samples/sec: 2800.05 - lr: 0.000018 - momentum: 0.000000 2023-10-16 09:09:54,031 epoch 7 - iter 1424/1786 - loss 0.02978701 - time (sec): 70.80 - samples/sec: 2799.97 - lr: 0.000018 - momentum: 0.000000 2023-10-16 09:10:02,873 epoch 7 - iter 1602/1786 - loss 0.02989961 - time (sec): 79.64 - samples/sec: 2806.68 - lr: 0.000017 - momentum: 0.000000 2023-10-16 09:10:11,596 epoch 7 - iter 1780/1786 - loss 0.02875032 - time (sec): 88.37 - samples/sec: 2808.03 - lr: 0.000017 - momentum: 0.000000 2023-10-16 09:10:11,893 ---------------------------------------------------------------------------------------------------- 2023-10-16 09:10:11,893 EPOCH 7 done: loss 0.0287 - lr: 0.000017 2023-10-16 09:10:16,471 DEV : loss 0.18708880245685577 - f1-score (micro avg) 0.7881 2023-10-16 09:10:16,487 ---------------------------------------------------------------------------------------------------- 2023-10-16 09:10:25,496 epoch 8 - iter 178/1786 - loss 0.02251176 - time (sec): 9.01 - samples/sec: 2740.85 - lr: 0.000016 - momentum: 0.000000 2023-10-16 09:10:34,741 epoch 8 - iter 356/1786 - loss 0.02431309 - time (sec): 18.25 - samples/sec: 2821.65 - lr: 0.000016 - momentum: 0.000000 2023-10-16 09:10:43,326 epoch 8 - iter 534/1786 - loss 0.02293182 - time (sec): 26.84 - samples/sec: 2799.53 - lr: 0.000015 - momentum: 0.000000 2023-10-16 09:10:52,052 epoch 8 - iter 712/1786 - loss 0.02282164 - time (sec): 35.56 - samples/sec: 2771.58 - lr: 0.000014 - momentum: 0.000000 2023-10-16 09:11:00,770 epoch 8 - iter 890/1786 - loss 0.02376710 - time (sec): 44.28 - samples/sec: 2737.99 - lr: 0.000014 - momentum: 0.000000 2023-10-16 09:11:10,034 epoch 8 - iter 1068/1786 - loss 0.02298881 - time (sec): 53.55 - samples/sec: 2757.59 - lr: 0.000013 - momentum: 0.000000 2023-10-16 09:11:18,807 epoch 8 - iter 1246/1786 - loss 0.02269837 - time (sec): 62.32 - samples/sec: 2768.66 - lr: 0.000013 - momentum: 0.000000 2023-10-16 09:11:27,722 epoch 8 - iter 1424/1786 - loss 0.02263600 - time (sec): 71.23 - samples/sec: 2798.75 - lr: 0.000012 - momentum: 0.000000 2023-10-16 09:11:36,520 epoch 8 - iter 1602/1786 - loss 0.02277881 - time (sec): 80.03 - samples/sec: 2808.31 - lr: 0.000012 - momentum: 0.000000 2023-10-16 09:11:44,784 epoch 8 - iter 1780/1786 - loss 0.02229267 - time (sec): 88.30 - samples/sec: 2811.57 - lr: 0.000011 - momentum: 0.000000 2023-10-16 09:11:45,035 ---------------------------------------------------------------------------------------------------- 2023-10-16 09:11:45,036 EPOCH 8 done: loss 0.0224 - lr: 0.000011 2023-10-16 09:11:49,077 DEV : loss 0.2031172513961792 - f1-score (micro avg) 0.7976 2023-10-16 09:11:49,093 saving best model 2023-10-16 09:11:49,555 ---------------------------------------------------------------------------------------------------- 2023-10-16 09:11:58,992 epoch 9 - iter 178/1786 - loss 0.02189545 - time (sec): 9.43 - samples/sec: 2629.64 - lr: 0.000011 - momentum: 0.000000 2023-10-16 09:12:07,517 epoch 9 - iter 356/1786 - loss 0.01747601 - time (sec): 17.96 - samples/sec: 2762.19 - lr: 0.000010 - momentum: 0.000000 2023-10-16 09:12:16,272 epoch 9 - iter 534/1786 - loss 0.01786279 - time (sec): 26.71 - samples/sec: 2828.56 - lr: 0.000009 - momentum: 0.000000 2023-10-16 09:12:25,209 epoch 9 - iter 712/1786 - loss 0.01550724 - time (sec): 35.65 - samples/sec: 2797.98 - lr: 0.000009 - momentum: 0.000000 2023-10-16 09:12:33,988 epoch 9 - iter 890/1786 - loss 0.01545366 - time (sec): 44.43 - samples/sec: 2794.14 - lr: 0.000008 - momentum: 0.000000 2023-10-16 09:12:42,895 epoch 9 - iter 1068/1786 - loss 0.01489816 - time (sec): 53.34 - samples/sec: 2812.03 - lr: 0.000008 - momentum: 0.000000 2023-10-16 09:12:51,587 epoch 9 - iter 1246/1786 - loss 0.01528282 - time (sec): 62.03 - samples/sec: 2832.17 - lr: 0.000007 - momentum: 0.000000 2023-10-16 09:13:00,247 epoch 9 - iter 1424/1786 - loss 0.01453460 - time (sec): 70.69 - samples/sec: 2840.93 - lr: 0.000007 - momentum: 0.000000 2023-10-16 09:13:08,659 epoch 9 - iter 1602/1786 - loss 0.01464221 - time (sec): 79.10 - samples/sec: 2832.27 - lr: 0.000006 - momentum: 0.000000 2023-10-16 09:13:17,327 epoch 9 - iter 1780/1786 - loss 0.01504874 - time (sec): 87.77 - samples/sec: 2825.63 - lr: 0.000006 - momentum: 0.000000 2023-10-16 09:13:17,625 ---------------------------------------------------------------------------------------------------- 2023-10-16 09:13:17,625 EPOCH 9 done: loss 0.0150 - lr: 0.000006 2023-10-16 09:13:21,657 DEV : loss 0.18724995851516724 - f1-score (micro avg) 0.8027 2023-10-16 09:13:21,673 saving best model 2023-10-16 09:13:22,121 ---------------------------------------------------------------------------------------------------- 2023-10-16 09:13:31,015 epoch 10 - iter 178/1786 - loss 0.00571772 - time (sec): 8.89 - samples/sec: 2914.90 - lr: 0.000005 - momentum: 0.000000 2023-10-16 09:13:39,977 epoch 10 - iter 356/1786 - loss 0.00917164 - time (sec): 17.85 - samples/sec: 2941.77 - lr: 0.000004 - momentum: 0.000000 2023-10-16 09:13:49,087 epoch 10 - iter 534/1786 - loss 0.00931697 - time (sec): 26.96 - samples/sec: 2873.81 - lr: 0.000004 - momentum: 0.000000 2023-10-16 09:13:57,666 epoch 10 - iter 712/1786 - loss 0.00909777 - time (sec): 35.54 - samples/sec: 2879.62 - lr: 0.000003 - momentum: 0.000000 2023-10-16 09:14:06,392 epoch 10 - iter 890/1786 - loss 0.00911553 - time (sec): 44.27 - samples/sec: 2859.28 - lr: 0.000003 - momentum: 0.000000 2023-10-16 09:14:14,995 epoch 10 - iter 1068/1786 - loss 0.00852015 - time (sec): 52.87 - samples/sec: 2838.41 - lr: 0.000002 - momentum: 0.000000 2023-10-16 09:14:23,665 epoch 10 - iter 1246/1786 - loss 0.00864843 - time (sec): 61.54 - samples/sec: 2817.84 - lr: 0.000002 - momentum: 0.000000 2023-10-16 09:14:32,625 epoch 10 - iter 1424/1786 - loss 0.00874912 - time (sec): 70.50 - samples/sec: 2815.35 - lr: 0.000001 - momentum: 0.000000 2023-10-16 09:14:41,364 epoch 10 - iter 1602/1786 - loss 0.00888492 - time (sec): 79.24 - samples/sec: 2821.53 - lr: 0.000001 - momentum: 0.000000 2023-10-16 09:14:50,107 epoch 10 - iter 1780/1786 - loss 0.00921080 - time (sec): 87.98 - samples/sec: 2818.57 - lr: 0.000000 - momentum: 0.000000 2023-10-16 09:14:50,383 ---------------------------------------------------------------------------------------------------- 2023-10-16 09:14:50,383 EPOCH 10 done: loss 0.0092 - lr: 0.000000 2023-10-16 09:14:55,010 DEV : loss 0.2006855458021164 - f1-score (micro avg) 0.8054 2023-10-16 09:14:55,026 saving best model 2023-10-16 09:14:55,864 ---------------------------------------------------------------------------------------------------- 2023-10-16 09:14:55,865 Loading model from best epoch ... 2023-10-16 09:14:57,658 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 2023-10-16 09:15:06,928 Results: - F-score (micro) 0.6898 - F-score (macro) 0.6075 - Accuracy 0.5436 By class: precision recall f1-score support LOC 0.7223 0.6721 0.6963 1095 PER 0.7563 0.7668 0.7615 1012 ORG 0.4680 0.5742 0.5157 357 HumanProd 0.3559 0.6364 0.4565 33 micro avg 0.6837 0.6960 0.6898 2497 macro avg 0.5756 0.6624 0.6075 2497 weighted avg 0.6949 0.6960 0.6938 2497 2023-10-16 09:15:06,928 ----------------------------------------------------------------------------------------------------