2023-10-25 18:37:38,397 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:37:38,398 Model: "SequenceTagger( (embeddings): TransformerWordEmbeddings( (model): BertModel( (embeddings): BertEmbeddings( (word_embeddings): Embedding(64001, 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() )" 2023-10-25 18:37:38,398 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:37:38,398 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 2023-10-25 18:37:38,399 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:37:38,399 Train: 7142 sentences 2023-10-25 18:37:38,399 (train_with_dev=False, train_with_test=False) 2023-10-25 18:37:38,399 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:37:38,399 Training Params: 2023-10-25 18:37:38,399 - learning_rate: "5e-05" 2023-10-25 18:37:38,399 - mini_batch_size: "4" 2023-10-25 18:37:38,399 - max_epochs: "10" 2023-10-25 18:37:38,399 - shuffle: "True" 2023-10-25 18:37:38,399 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:37:38,399 Plugins: 2023-10-25 18:37:38,399 - TensorboardLogger 2023-10-25 18:37:38,399 - LinearScheduler | warmup_fraction: '0.1' 2023-10-25 18:37:38,399 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:37:38,399 Final evaluation on model from best epoch (best-model.pt) 2023-10-25 18:37:38,399 - metric: "('micro avg', 'f1-score')" 2023-10-25 18:37:38,399 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:37:38,399 Computation: 2023-10-25 18:37:38,399 - compute on device: cuda:0 2023-10-25 18:37:38,399 - embedding storage: none 2023-10-25 18:37:38,399 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:37:38,399 Model training base path: "hmbench-newseye/fr-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-5" 2023-10-25 18:37:38,399 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:37:38,399 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:37:38,399 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-25 18:37:47,542 epoch 1 - iter 178/1786 - loss 1.40453799 - time (sec): 9.14 - samples/sec: 2496.26 - lr: 0.000005 - momentum: 0.000000 2023-10-25 18:37:56,903 epoch 1 - iter 356/1786 - loss 0.90807035 - time (sec): 18.50 - samples/sec: 2525.74 - lr: 0.000010 - momentum: 0.000000 2023-10-25 18:38:06,238 epoch 1 - iter 534/1786 - loss 0.69820556 - time (sec): 27.84 - samples/sec: 2559.06 - lr: 0.000015 - momentum: 0.000000 2023-10-25 18:38:15,506 epoch 1 - iter 712/1786 - loss 0.56410304 - time (sec): 37.11 - samples/sec: 2650.91 - lr: 0.000020 - momentum: 0.000000 2023-10-25 18:38:24,825 epoch 1 - iter 890/1786 - loss 0.48580068 - time (sec): 46.42 - samples/sec: 2644.52 - lr: 0.000025 - momentum: 0.000000 2023-10-25 18:38:34,115 epoch 1 - iter 1068/1786 - loss 0.43280563 - time (sec): 55.71 - samples/sec: 2626.42 - lr: 0.000030 - momentum: 0.000000 2023-10-25 18:38:43,163 epoch 1 - iter 1246/1786 - loss 0.39270958 - time (sec): 64.76 - samples/sec: 2638.36 - lr: 0.000035 - momentum: 0.000000 2023-10-25 18:38:52,184 epoch 1 - iter 1424/1786 - loss 0.36098717 - time (sec): 73.78 - samples/sec: 2661.58 - lr: 0.000040 - momentum: 0.000000 2023-10-25 18:39:01,242 epoch 1 - iter 1602/1786 - loss 0.33608692 - time (sec): 82.84 - samples/sec: 2688.79 - lr: 0.000045 - momentum: 0.000000 2023-10-25 18:39:10,573 epoch 1 - iter 1780/1786 - loss 0.31930380 - time (sec): 92.17 - samples/sec: 2689.56 - lr: 0.000050 - momentum: 0.000000 2023-10-25 18:39:10,863 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:39:10,864 EPOCH 1 done: loss 0.3186 - lr: 0.000050 2023-10-25 18:39:15,004 DEV : loss 0.10816145688295364 - f1-score (micro avg) 0.7071 2023-10-25 18:39:15,026 saving best model 2023-10-25 18:39:15,535 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:39:25,230 epoch 2 - iter 178/1786 - loss 0.11968307 - time (sec): 9.69 - samples/sec: 2666.97 - lr: 0.000049 - momentum: 0.000000 2023-10-25 18:39:34,400 epoch 2 - iter 356/1786 - loss 0.12356003 - time (sec): 18.86 - samples/sec: 2568.18 - lr: 0.000049 - momentum: 0.000000 2023-10-25 18:39:43,423 epoch 2 - iter 534/1786 - loss 0.12183286 - time (sec): 27.89 - samples/sec: 2629.03 - lr: 0.000048 - momentum: 0.000000 2023-10-25 18:39:52,372 epoch 2 - iter 712/1786 - loss 0.12132289 - time (sec): 36.83 - samples/sec: 2689.32 - lr: 0.000048 - momentum: 0.000000 2023-10-25 18:40:01,037 epoch 2 - iter 890/1786 - loss 0.12094300 - time (sec): 45.50 - samples/sec: 2683.34 - lr: 0.000047 - momentum: 0.000000 2023-10-25 18:40:09,856 epoch 2 - iter 1068/1786 - loss 0.12136206 - time (sec): 54.32 - samples/sec: 2703.63 - lr: 0.000047 - momentum: 0.000000 2023-10-25 18:40:18,988 epoch 2 - iter 1246/1786 - loss 0.12150787 - time (sec): 63.45 - samples/sec: 2703.34 - lr: 0.000046 - momentum: 0.000000 2023-10-25 18:40:28,126 epoch 2 - iter 1424/1786 - loss 0.12125528 - time (sec): 72.59 - samples/sec: 2734.94 - lr: 0.000046 - momentum: 0.000000 2023-10-25 18:40:37,434 epoch 2 - iter 1602/1786 - loss 0.12087874 - time (sec): 81.90 - samples/sec: 2723.53 - lr: 0.000045 - momentum: 0.000000 2023-10-25 18:40:46,385 epoch 2 - iter 1780/1786 - loss 0.12089949 - time (sec): 90.85 - samples/sec: 2730.30 - lr: 0.000044 - momentum: 0.000000 2023-10-25 18:40:46,664 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:40:46,665 EPOCH 2 done: loss 0.1209 - lr: 0.000044 2023-10-25 18:40:50,825 DEV : loss 0.11196932196617126 - f1-score (micro avg) 0.7568 2023-10-25 18:40:50,846 saving best model 2023-10-25 18:40:51,544 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:41:00,933 epoch 3 - iter 178/1786 - loss 0.07216058 - time (sec): 9.39 - samples/sec: 2668.52 - lr: 0.000044 - momentum: 0.000000 2023-10-25 18:41:11,563 epoch 3 - iter 356/1786 - loss 0.08332331 - time (sec): 20.02 - samples/sec: 2458.48 - lr: 0.000043 - momentum: 0.000000 2023-10-25 18:41:21,065 epoch 3 - iter 534/1786 - loss 0.08767583 - time (sec): 29.52 - samples/sec: 2493.08 - lr: 0.000043 - momentum: 0.000000 2023-10-25 18:41:30,647 epoch 3 - iter 712/1786 - loss 0.08415718 - time (sec): 39.10 - samples/sec: 2524.84 - lr: 0.000042 - momentum: 0.000000 2023-10-25 18:41:40,016 epoch 3 - iter 890/1786 - loss 0.08739794 - time (sec): 48.47 - samples/sec: 2552.95 - lr: 0.000042 - momentum: 0.000000 2023-10-25 18:41:49,509 epoch 3 - iter 1068/1786 - loss 0.08687276 - time (sec): 57.96 - samples/sec: 2572.12 - lr: 0.000041 - momentum: 0.000000 2023-10-25 18:41:59,111 epoch 3 - iter 1246/1786 - loss 0.08597967 - time (sec): 67.57 - samples/sec: 2587.39 - lr: 0.000041 - momentum: 0.000000 2023-10-25 18:42:08,462 epoch 3 - iter 1424/1786 - loss 0.08636941 - time (sec): 76.92 - samples/sec: 2559.26 - lr: 0.000040 - momentum: 0.000000 2023-10-25 18:42:17,442 epoch 3 - iter 1602/1786 - loss 0.08593164 - time (sec): 85.90 - samples/sec: 2587.64 - lr: 0.000039 - momentum: 0.000000 2023-10-25 18:42:26,319 epoch 3 - iter 1780/1786 - loss 0.08507425 - time (sec): 94.77 - samples/sec: 2616.24 - lr: 0.000039 - momentum: 0.000000 2023-10-25 18:42:26,608 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:42:26,608 EPOCH 3 done: loss 0.0850 - lr: 0.000039 2023-10-25 18:42:30,919 DEV : loss 0.13416369259357452 - f1-score (micro avg) 0.7734 2023-10-25 18:42:30,941 saving best model 2023-10-25 18:42:31,602 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:42:40,853 epoch 4 - iter 178/1786 - loss 0.05392007 - time (sec): 9.25 - samples/sec: 2660.38 - lr: 0.000038 - momentum: 0.000000 2023-10-25 18:42:50,351 epoch 4 - iter 356/1786 - loss 0.05706668 - time (sec): 18.75 - samples/sec: 2686.77 - lr: 0.000038 - momentum: 0.000000 2023-10-25 18:42:59,263 epoch 4 - iter 534/1786 - loss 0.06080001 - time (sec): 27.66 - samples/sec: 2693.61 - lr: 0.000037 - momentum: 0.000000 2023-10-25 18:43:08,040 epoch 4 - iter 712/1786 - loss 0.06411629 - time (sec): 36.44 - samples/sec: 2714.05 - lr: 0.000037 - momentum: 0.000000 2023-10-25 18:43:17,331 epoch 4 - iter 890/1786 - loss 0.06293852 - time (sec): 45.73 - samples/sec: 2693.11 - lr: 0.000036 - momentum: 0.000000 2023-10-25 18:43:26,753 epoch 4 - iter 1068/1786 - loss 0.06522331 - time (sec): 55.15 - samples/sec: 2696.19 - lr: 0.000036 - momentum: 0.000000 2023-10-25 18:43:35,821 epoch 4 - iter 1246/1786 - loss 0.06635384 - time (sec): 64.22 - samples/sec: 2691.82 - lr: 0.000035 - momentum: 0.000000 2023-10-25 18:43:44,420 epoch 4 - iter 1424/1786 - loss 0.06654842 - time (sec): 72.82 - samples/sec: 2724.98 - lr: 0.000034 - momentum: 0.000000 2023-10-25 18:43:53,437 epoch 4 - iter 1602/1786 - loss 0.06525833 - time (sec): 81.83 - samples/sec: 2730.35 - lr: 0.000034 - momentum: 0.000000 2023-10-25 18:44:02,280 epoch 4 - iter 1780/1786 - loss 0.06417141 - time (sec): 90.68 - samples/sec: 2735.96 - lr: 0.000033 - momentum: 0.000000 2023-10-25 18:44:02,561 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:44:02,561 EPOCH 4 done: loss 0.0641 - lr: 0.000033 2023-10-25 18:44:07,999 DEV : loss 0.1622055321931839 - f1-score (micro avg) 0.7949 2023-10-25 18:44:08,021 saving best model 2023-10-25 18:44:08,751 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:44:18,039 epoch 5 - iter 178/1786 - loss 0.05453094 - time (sec): 9.29 - samples/sec: 2544.86 - lr: 0.000033 - momentum: 0.000000 2023-10-25 18:44:27,561 epoch 5 - iter 356/1786 - loss 0.04577426 - time (sec): 18.81 - samples/sec: 2676.28 - lr: 0.000032 - momentum: 0.000000 2023-10-25 18:44:37,038 epoch 5 - iter 534/1786 - loss 0.04807378 - time (sec): 28.28 - samples/sec: 2655.55 - lr: 0.000032 - momentum: 0.000000 2023-10-25 18:44:46,571 epoch 5 - iter 712/1786 - loss 0.04717079 - time (sec): 37.82 - samples/sec: 2653.43 - lr: 0.000031 - momentum: 0.000000 2023-10-25 18:44:56,134 epoch 5 - iter 890/1786 - loss 0.04803019 - time (sec): 47.38 - samples/sec: 2655.16 - lr: 0.000031 - momentum: 0.000000 2023-10-25 18:45:05,721 epoch 5 - iter 1068/1786 - loss 0.04817761 - time (sec): 56.97 - samples/sec: 2618.31 - lr: 0.000030 - momentum: 0.000000 2023-10-25 18:45:15,218 epoch 5 - iter 1246/1786 - loss 0.04776400 - time (sec): 66.46 - samples/sec: 2624.83 - lr: 0.000029 - momentum: 0.000000 2023-10-25 18:45:24,767 epoch 5 - iter 1424/1786 - loss 0.04756135 - time (sec): 76.01 - samples/sec: 2611.00 - lr: 0.000029 - momentum: 0.000000 2023-10-25 18:45:34,208 epoch 5 - iter 1602/1786 - loss 0.04690501 - time (sec): 85.46 - samples/sec: 2596.86 - lr: 0.000028 - momentum: 0.000000 2023-10-25 18:45:43,226 epoch 5 - iter 1780/1786 - loss 0.04701536 - time (sec): 94.47 - samples/sec: 2622.82 - lr: 0.000028 - momentum: 0.000000 2023-10-25 18:45:43,541 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:45:43,541 EPOCH 5 done: loss 0.0471 - lr: 0.000028 2023-10-25 18:45:47,551 DEV : loss 0.17839400470256805 - f1-score (micro avg) 0.7641 2023-10-25 18:45:47,571 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:45:57,008 epoch 6 - iter 178/1786 - loss 0.04739676 - time (sec): 9.44 - samples/sec: 2669.74 - lr: 0.000027 - momentum: 0.000000 2023-10-25 18:46:06,535 epoch 6 - iter 356/1786 - loss 0.03785218 - time (sec): 18.96 - samples/sec: 2692.43 - lr: 0.000027 - momentum: 0.000000 2023-10-25 18:46:15,980 epoch 6 - iter 534/1786 - loss 0.03634755 - time (sec): 28.41 - samples/sec: 2618.52 - lr: 0.000026 - momentum: 0.000000 2023-10-25 18:46:25,654 epoch 6 - iter 712/1786 - loss 0.04044514 - time (sec): 38.08 - samples/sec: 2607.08 - lr: 0.000026 - momentum: 0.000000 2023-10-25 18:46:35,381 epoch 6 - iter 890/1786 - loss 0.03838047 - time (sec): 47.81 - samples/sec: 2615.00 - lr: 0.000025 - momentum: 0.000000 2023-10-25 18:46:44,873 epoch 6 - iter 1068/1786 - loss 0.03827037 - time (sec): 57.30 - samples/sec: 2616.02 - lr: 0.000024 - momentum: 0.000000 2023-10-25 18:46:54,684 epoch 6 - iter 1246/1786 - loss 0.03744695 - time (sec): 67.11 - samples/sec: 2607.36 - lr: 0.000024 - momentum: 0.000000 2023-10-25 18:47:03,908 epoch 6 - iter 1424/1786 - loss 0.03713239 - time (sec): 76.34 - samples/sec: 2598.15 - lr: 0.000023 - momentum: 0.000000 2023-10-25 18:47:13,637 epoch 6 - iter 1602/1786 - loss 0.03788438 - time (sec): 86.06 - samples/sec: 2617.86 - lr: 0.000023 - momentum: 0.000000 2023-10-25 18:47:22,668 epoch 6 - iter 1780/1786 - loss 0.03775238 - time (sec): 95.10 - samples/sec: 2609.10 - lr: 0.000022 - momentum: 0.000000 2023-10-25 18:47:22,974 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:47:22,975 EPOCH 6 done: loss 0.0377 - lr: 0.000022 2023-10-25 18:47:27,370 DEV : loss 0.17886780202388763 - f1-score (micro avg) 0.7886 2023-10-25 18:47:27,392 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:47:37,024 epoch 7 - iter 178/1786 - loss 0.03077925 - time (sec): 9.63 - samples/sec: 2692.33 - lr: 0.000022 - momentum: 0.000000 2023-10-25 18:47:46,517 epoch 7 - iter 356/1786 - loss 0.02932877 - time (sec): 19.12 - samples/sec: 2574.46 - lr: 0.000021 - momentum: 0.000000 2023-10-25 18:47:55,977 epoch 7 - iter 534/1786 - loss 0.02829002 - time (sec): 28.58 - samples/sec: 2597.22 - lr: 0.000021 - momentum: 0.000000 2023-10-25 18:48:05,423 epoch 7 - iter 712/1786 - loss 0.02973574 - time (sec): 38.03 - samples/sec: 2597.81 - lr: 0.000020 - momentum: 0.000000 2023-10-25 18:48:14,909 epoch 7 - iter 890/1786 - loss 0.02980090 - time (sec): 47.52 - samples/sec: 2639.78 - lr: 0.000019 - momentum: 0.000000 2023-10-25 18:48:24,135 epoch 7 - iter 1068/1786 - loss 0.03007659 - time (sec): 56.74 - samples/sec: 2657.82 - lr: 0.000019 - momentum: 0.000000 2023-10-25 18:48:33,402 epoch 7 - iter 1246/1786 - loss 0.02862806 - time (sec): 66.01 - samples/sec: 2678.20 - lr: 0.000018 - momentum: 0.000000 2023-10-25 18:48:42,731 epoch 7 - iter 1424/1786 - loss 0.02826589 - time (sec): 75.34 - samples/sec: 2639.07 - lr: 0.000018 - momentum: 0.000000 2023-10-25 18:48:51,933 epoch 7 - iter 1602/1786 - loss 0.02797061 - time (sec): 84.54 - samples/sec: 2641.63 - lr: 0.000017 - momentum: 0.000000 2023-10-25 18:49:01,510 epoch 7 - iter 1780/1786 - loss 0.02819106 - time (sec): 94.12 - samples/sec: 2632.82 - lr: 0.000017 - momentum: 0.000000 2023-10-25 18:49:01,831 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:49:01,831 EPOCH 7 done: loss 0.0281 - lr: 0.000017 2023-10-25 18:49:06,709 DEV : loss 0.19223909080028534 - f1-score (micro avg) 0.7857 2023-10-25 18:49:06,731 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:49:16,260 epoch 8 - iter 178/1786 - loss 0.02174503 - time (sec): 9.53 - samples/sec: 2599.42 - lr: 0.000016 - momentum: 0.000000 2023-10-25 18:49:25,630 epoch 8 - iter 356/1786 - loss 0.01913432 - time (sec): 18.90 - samples/sec: 2564.11 - lr: 0.000016 - momentum: 0.000000 2023-10-25 18:49:35,140 epoch 8 - iter 534/1786 - loss 0.01746195 - time (sec): 28.41 - samples/sec: 2589.15 - lr: 0.000015 - momentum: 0.000000 2023-10-25 18:49:44,521 epoch 8 - iter 712/1786 - loss 0.01693117 - time (sec): 37.79 - samples/sec: 2623.00 - lr: 0.000014 - momentum: 0.000000 2023-10-25 18:49:53,894 epoch 8 - iter 890/1786 - loss 0.01671270 - time (sec): 47.16 - samples/sec: 2621.55 - lr: 0.000014 - momentum: 0.000000 2023-10-25 18:50:03,220 epoch 8 - iter 1068/1786 - loss 0.01731034 - time (sec): 56.49 - samples/sec: 2602.71 - lr: 0.000013 - momentum: 0.000000 2023-10-25 18:50:12,498 epoch 8 - iter 1246/1786 - loss 0.01886582 - time (sec): 65.77 - samples/sec: 2608.65 - lr: 0.000013 - momentum: 0.000000 2023-10-25 18:50:21,455 epoch 8 - iter 1424/1786 - loss 0.01893327 - time (sec): 74.72 - samples/sec: 2631.11 - lr: 0.000012 - momentum: 0.000000 2023-10-25 18:50:30,440 epoch 8 - iter 1602/1786 - loss 0.01901785 - time (sec): 83.71 - samples/sec: 2666.98 - lr: 0.000012 - momentum: 0.000000 2023-10-25 18:50:39,483 epoch 8 - iter 1780/1786 - loss 0.01903471 - time (sec): 92.75 - samples/sec: 2673.79 - lr: 0.000011 - momentum: 0.000000 2023-10-25 18:50:39,785 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:50:39,785 EPOCH 8 done: loss 0.0190 - lr: 0.000011 2023-10-25 18:50:43,730 DEV : loss 0.22295665740966797 - f1-score (micro avg) 0.7927 2023-10-25 18:50:43,754 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:50:53,379 epoch 9 - iter 178/1786 - loss 0.00894777 - time (sec): 9.62 - samples/sec: 2649.14 - lr: 0.000011 - momentum: 0.000000 2023-10-25 18:51:02,883 epoch 9 - iter 356/1786 - loss 0.01120950 - time (sec): 19.13 - samples/sec: 2527.08 - lr: 0.000010 - momentum: 0.000000 2023-10-25 18:51:12,372 epoch 9 - iter 534/1786 - loss 0.01367811 - time (sec): 28.62 - samples/sec: 2533.14 - lr: 0.000009 - momentum: 0.000000 2023-10-25 18:51:21,536 epoch 9 - iter 712/1786 - loss 0.01359022 - time (sec): 37.78 - samples/sec: 2567.40 - lr: 0.000009 - momentum: 0.000000 2023-10-25 18:51:30,710 epoch 9 - iter 890/1786 - loss 0.01350802 - time (sec): 46.95 - samples/sec: 2560.14 - lr: 0.000008 - momentum: 0.000000 2023-10-25 18:51:39,721 epoch 9 - iter 1068/1786 - loss 0.01419765 - time (sec): 55.97 - samples/sec: 2623.59 - lr: 0.000008 - momentum: 0.000000 2023-10-25 18:51:48,814 epoch 9 - iter 1246/1786 - loss 0.01486049 - time (sec): 65.06 - samples/sec: 2629.79 - lr: 0.000007 - momentum: 0.000000 2023-10-25 18:51:57,499 epoch 9 - iter 1424/1786 - loss 0.01410719 - time (sec): 73.74 - samples/sec: 2666.03 - lr: 0.000007 - momentum: 0.000000 2023-10-25 18:52:06,116 epoch 9 - iter 1602/1786 - loss 0.01367858 - time (sec): 82.36 - samples/sec: 2689.33 - lr: 0.000006 - momentum: 0.000000 2023-10-25 18:52:15,069 epoch 9 - iter 1780/1786 - loss 0.01309891 - time (sec): 91.31 - samples/sec: 2717.19 - lr: 0.000006 - momentum: 0.000000 2023-10-25 18:52:15,368 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:52:15,369 EPOCH 9 done: loss 0.0132 - lr: 0.000006 2023-10-25 18:52:20,439 DEV : loss 0.22367699444293976 - f1-score (micro avg) 0.7949 2023-10-25 18:52:20,459 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:52:29,298 epoch 10 - iter 178/1786 - loss 0.00880098 - time (sec): 8.84 - samples/sec: 2842.17 - lr: 0.000005 - momentum: 0.000000 2023-10-25 18:52:37,837 epoch 10 - iter 356/1786 - loss 0.00764413 - time (sec): 17.38 - samples/sec: 2803.77 - lr: 0.000004 - momentum: 0.000000 2023-10-25 18:52:46,910 epoch 10 - iter 534/1786 - loss 0.00803735 - time (sec): 26.45 - samples/sec: 2802.88 - lr: 0.000004 - momentum: 0.000000 2023-10-25 18:52:56,296 epoch 10 - iter 712/1786 - loss 0.00807527 - time (sec): 35.84 - samples/sec: 2775.80 - lr: 0.000003 - momentum: 0.000000 2023-10-25 18:53:05,399 epoch 10 - iter 890/1786 - loss 0.00741347 - time (sec): 44.94 - samples/sec: 2772.21 - lr: 0.000003 - momentum: 0.000000 2023-10-25 18:53:14,850 epoch 10 - iter 1068/1786 - loss 0.00783141 - time (sec): 54.39 - samples/sec: 2736.83 - lr: 0.000002 - momentum: 0.000000 2023-10-25 18:53:24,330 epoch 10 - iter 1246/1786 - loss 0.00870824 - time (sec): 63.87 - samples/sec: 2711.41 - lr: 0.000002 - momentum: 0.000000 2023-10-25 18:53:33,732 epoch 10 - iter 1424/1786 - loss 0.00816998 - time (sec): 73.27 - samples/sec: 2716.97 - lr: 0.000001 - momentum: 0.000000 2023-10-25 18:53:43,352 epoch 10 - iter 1602/1786 - loss 0.00820551 - time (sec): 82.89 - samples/sec: 2698.61 - lr: 0.000001 - momentum: 0.000000 2023-10-25 18:53:52,998 epoch 10 - iter 1780/1786 - loss 0.00861252 - time (sec): 92.54 - samples/sec: 2679.83 - lr: 0.000000 - momentum: 0.000000 2023-10-25 18:53:53,317 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:53:53,317 EPOCH 10 done: loss 0.0086 - lr: 0.000000 2023-10-25 18:53:58,167 DEV : loss 0.22940541803836823 - f1-score (micro avg) 0.792 2023-10-25 18:53:58,699 ---------------------------------------------------------------------------------------------------- 2023-10-25 18:53:58,701 Loading model from best epoch ... 2023-10-25 18:54:00,626 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 2023-10-25 18:54:13,345 Results: - F-score (micro) 0.6526 - F-score (macro) 0.5814 - Accuracy 0.5009 By class: precision recall f1-score support LOC 0.6415 0.6521 0.6467 1095 PER 0.7798 0.7312 0.7547 1012 ORG 0.3913 0.5294 0.4500 357 HumanProd 0.3594 0.6970 0.4742 33 micro avg 0.6386 0.6672 0.6526 2497 macro avg 0.5430 0.6524 0.5814 2497 weighted avg 0.6580 0.6672 0.6601 2497 2023-10-25 18:54:13,345 ----------------------------------------------------------------------------------------------------