2023-10-11 02:58:49,354 ---------------------------------------------------------------------------------------------------- 2023-10-11 02:58:49,356 Model: "SequenceTagger( (embeddings): ByT5Embeddings( (model): T5EncoderModel( (shared): Embedding(384, 1472) (encoder): T5Stack( (embed_tokens): Embedding(384, 1472) (block): ModuleList( (0): T5Block( (layer): ModuleList( (0): T5LayerSelfAttention( (SelfAttention): T5Attention( (q): Linear(in_features=1472, out_features=384, bias=False) (k): Linear(in_features=1472, out_features=384, bias=False) (v): Linear(in_features=1472, out_features=384, bias=False) (o): Linear(in_features=384, out_features=1472, bias=False) (relative_attention_bias): Embedding(32, 6) ) (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) (1): T5LayerFF( (DenseReluDense): T5DenseGatedActDense( (wi_0): Linear(in_features=1472, out_features=3584, bias=False) (wi_1): Linear(in_features=1472, out_features=3584, bias=False) (wo): Linear(in_features=3584, out_features=1472, bias=False) (dropout): Dropout(p=0.1, inplace=False) (act): NewGELUActivation() ) (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) (1-11): 11 x T5Block( (layer): ModuleList( (0): T5LayerSelfAttention( (SelfAttention): T5Attention( (q): Linear(in_features=1472, out_features=384, bias=False) (k): Linear(in_features=1472, out_features=384, bias=False) (v): Linear(in_features=1472, out_features=384, bias=False) (o): Linear(in_features=384, out_features=1472, bias=False) ) (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) (1): T5LayerFF( (DenseReluDense): T5DenseGatedActDense( (wi_0): Linear(in_features=1472, out_features=3584, bias=False) (wi_1): Linear(in_features=1472, out_features=3584, bias=False) (wo): Linear(in_features=3584, out_features=1472, bias=False) (dropout): Dropout(p=0.1, inplace=False) (act): NewGELUActivation() ) (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) ) (final_layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) (locked_dropout): LockedDropout(p=0.5) (linear): Linear(in_features=1472, out_features=17, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-11 02:58:49,356 ---------------------------------------------------------------------------------------------------- 2023-10-11 02:58:49,356 MultiCorpus: 1166 train + 165 dev + 415 test sentences - NER_HIPE_2022 Corpus: 1166 train + 165 dev + 415 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fi/with_doc_seperator 2023-10-11 02:58:49,357 ---------------------------------------------------------------------------------------------------- 2023-10-11 02:58:49,357 Train: 1166 sentences 2023-10-11 02:58:49,357 (train_with_dev=False, train_with_test=False) 2023-10-11 02:58:49,357 ---------------------------------------------------------------------------------------------------- 2023-10-11 02:58:49,357 Training Params: 2023-10-11 02:58:49,357 - learning_rate: "0.00016" 2023-10-11 02:58:49,357 - mini_batch_size: "8" 2023-10-11 02:58:49,357 - max_epochs: "10" 2023-10-11 02:58:49,357 - shuffle: "True" 2023-10-11 02:58:49,357 ---------------------------------------------------------------------------------------------------- 2023-10-11 02:58:49,357 Plugins: 2023-10-11 02:58:49,357 - TensorboardLogger 2023-10-11 02:58:49,357 - LinearScheduler | warmup_fraction: '0.1' 2023-10-11 02:58:49,357 ---------------------------------------------------------------------------------------------------- 2023-10-11 02:58:49,357 Final evaluation on model from best epoch (best-model.pt) 2023-10-11 02:58:49,357 - metric: "('micro avg', 'f1-score')" 2023-10-11 02:58:49,358 ---------------------------------------------------------------------------------------------------- 2023-10-11 02:58:49,358 Computation: 2023-10-11 02:58:49,358 - compute on device: cuda:0 2023-10-11 02:58:49,358 - embedding storage: none 2023-10-11 02:58:49,358 ---------------------------------------------------------------------------------------------------- 2023-10-11 02:58:49,358 Model training base path: "hmbench-newseye/fi-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-5" 2023-10-11 02:58:49,358 ---------------------------------------------------------------------------------------------------- 2023-10-11 02:58:49,358 ---------------------------------------------------------------------------------------------------- 2023-10-11 02:58:49,358 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-11 02:58:58,372 epoch 1 - iter 14/146 - loss 2.82156082 - time (sec): 9.01 - samples/sec: 470.06 - lr: 0.000014 - momentum: 0.000000 2023-10-11 02:59:06,819 epoch 1 - iter 28/146 - loss 2.81328962 - time (sec): 17.46 - samples/sec: 467.31 - lr: 0.000030 - momentum: 0.000000 2023-10-11 02:59:15,404 epoch 1 - iter 42/146 - loss 2.80236709 - time (sec): 26.04 - samples/sec: 455.58 - lr: 0.000045 - momentum: 0.000000 2023-10-11 02:59:25,395 epoch 1 - iter 56/146 - loss 2.77560051 - time (sec): 36.03 - samples/sec: 460.86 - lr: 0.000060 - momentum: 0.000000 2023-10-11 02:59:34,647 epoch 1 - iter 70/146 - loss 2.73504332 - time (sec): 45.29 - samples/sec: 461.08 - lr: 0.000076 - momentum: 0.000000 2023-10-11 02:59:44,395 epoch 1 - iter 84/146 - loss 2.66639699 - time (sec): 55.04 - samples/sec: 469.45 - lr: 0.000091 - momentum: 0.000000 2023-10-11 02:59:53,771 epoch 1 - iter 98/146 - loss 2.58655394 - time (sec): 64.41 - samples/sec: 475.67 - lr: 0.000106 - momentum: 0.000000 2023-10-11 03:00:02,636 epoch 1 - iter 112/146 - loss 2.52547117 - time (sec): 73.28 - samples/sec: 475.66 - lr: 0.000122 - momentum: 0.000000 2023-10-11 03:00:12,006 epoch 1 - iter 126/146 - loss 2.43072282 - time (sec): 82.65 - samples/sec: 478.44 - lr: 0.000137 - momentum: 0.000000 2023-10-11 03:00:20,230 epoch 1 - iter 140/146 - loss 2.35869693 - time (sec): 90.87 - samples/sec: 475.36 - lr: 0.000152 - momentum: 0.000000 2023-10-11 03:00:23,460 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:00:23,460 EPOCH 1 done: loss 2.3326 - lr: 0.000152 2023-10-11 03:00:28,956 DEV : loss 1.2650508880615234 - f1-score (micro avg) 0.0 2023-10-11 03:00:28,965 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:00:37,834 epoch 2 - iter 14/146 - loss 1.28850784 - time (sec): 8.87 - samples/sec: 483.05 - lr: 0.000158 - momentum: 0.000000 2023-10-11 03:00:46,400 epoch 2 - iter 28/146 - loss 1.17475751 - time (sec): 17.43 - samples/sec: 488.80 - lr: 0.000157 - momentum: 0.000000 2023-10-11 03:00:54,577 epoch 2 - iter 42/146 - loss 1.11252222 - time (sec): 25.61 - samples/sec: 479.42 - lr: 0.000155 - momentum: 0.000000 2023-10-11 03:01:03,357 epoch 2 - iter 56/146 - loss 1.00841290 - time (sec): 34.39 - samples/sec: 481.62 - lr: 0.000153 - momentum: 0.000000 2023-10-11 03:01:12,621 epoch 2 - iter 70/146 - loss 0.99023783 - time (sec): 43.65 - samples/sec: 488.32 - lr: 0.000152 - momentum: 0.000000 2023-10-11 03:01:21,672 epoch 2 - iter 84/146 - loss 0.92269285 - time (sec): 52.71 - samples/sec: 490.27 - lr: 0.000150 - momentum: 0.000000 2023-10-11 03:01:30,342 epoch 2 - iter 98/146 - loss 0.87677537 - time (sec): 61.38 - samples/sec: 490.80 - lr: 0.000148 - momentum: 0.000000 2023-10-11 03:01:38,580 epoch 2 - iter 112/146 - loss 0.83163574 - time (sec): 69.61 - samples/sec: 487.38 - lr: 0.000147 - momentum: 0.000000 2023-10-11 03:01:46,714 epoch 2 - iter 126/146 - loss 0.81335178 - time (sec): 77.75 - samples/sec: 483.21 - lr: 0.000145 - momentum: 0.000000 2023-10-11 03:01:56,170 epoch 2 - iter 140/146 - loss 0.78893900 - time (sec): 87.20 - samples/sec: 486.32 - lr: 0.000143 - momentum: 0.000000 2023-10-11 03:02:00,042 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:02:00,043 EPOCH 2 done: loss 0.7803 - lr: 0.000143 2023-10-11 03:02:05,705 DEV : loss 0.4041951894760132 - f1-score (micro avg) 0.0 2023-10-11 03:02:05,714 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:02:14,895 epoch 3 - iter 14/146 - loss 0.53427675 - time (sec): 9.18 - samples/sec: 419.66 - lr: 0.000141 - momentum: 0.000000 2023-10-11 03:02:23,336 epoch 3 - iter 28/146 - loss 0.51483169 - time (sec): 17.62 - samples/sec: 439.28 - lr: 0.000139 - momentum: 0.000000 2023-10-11 03:02:32,172 epoch 3 - iter 42/146 - loss 0.48292516 - time (sec): 26.46 - samples/sec: 457.02 - lr: 0.000137 - momentum: 0.000000 2023-10-11 03:02:40,999 epoch 3 - iter 56/146 - loss 0.45342483 - time (sec): 35.28 - samples/sec: 471.31 - lr: 0.000136 - momentum: 0.000000 2023-10-11 03:02:49,374 epoch 3 - iter 70/146 - loss 0.44551814 - time (sec): 43.66 - samples/sec: 471.21 - lr: 0.000134 - momentum: 0.000000 2023-10-11 03:02:57,741 epoch 3 - iter 84/146 - loss 0.42893312 - time (sec): 52.03 - samples/sec: 474.40 - lr: 0.000132 - momentum: 0.000000 2023-10-11 03:03:07,151 epoch 3 - iter 98/146 - loss 0.44060923 - time (sec): 61.43 - samples/sec: 482.77 - lr: 0.000131 - momentum: 0.000000 2023-10-11 03:03:16,212 epoch 3 - iter 112/146 - loss 0.43204748 - time (sec): 70.50 - samples/sec: 472.07 - lr: 0.000129 - momentum: 0.000000 2023-10-11 03:03:26,562 epoch 3 - iter 126/146 - loss 0.42357750 - time (sec): 80.85 - samples/sec: 472.37 - lr: 0.000127 - momentum: 0.000000 2023-10-11 03:03:36,128 epoch 3 - iter 140/146 - loss 0.42216673 - time (sec): 90.41 - samples/sec: 470.41 - lr: 0.000125 - momentum: 0.000000 2023-10-11 03:03:40,267 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:03:40,268 EPOCH 3 done: loss 0.4162 - lr: 0.000125 2023-10-11 03:03:45,974 DEV : loss 0.2766019403934479 - f1-score (micro avg) 0.0 2023-10-11 03:03:45,983 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:03:56,215 epoch 4 - iter 14/146 - loss 0.28148112 - time (sec): 10.23 - samples/sec: 450.55 - lr: 0.000123 - momentum: 0.000000 2023-10-11 03:04:06,418 epoch 4 - iter 28/146 - loss 0.26087007 - time (sec): 20.43 - samples/sec: 466.80 - lr: 0.000121 - momentum: 0.000000 2023-10-11 03:04:15,473 epoch 4 - iter 42/146 - loss 0.27268962 - time (sec): 29.49 - samples/sec: 473.85 - lr: 0.000120 - momentum: 0.000000 2023-10-11 03:04:24,961 epoch 4 - iter 56/146 - loss 0.32422208 - time (sec): 38.98 - samples/sec: 488.20 - lr: 0.000118 - momentum: 0.000000 2023-10-11 03:04:33,577 epoch 4 - iter 70/146 - loss 0.32316287 - time (sec): 47.59 - samples/sec: 486.47 - lr: 0.000116 - momentum: 0.000000 2023-10-11 03:04:42,328 epoch 4 - iter 84/146 - loss 0.31830554 - time (sec): 56.34 - samples/sec: 483.60 - lr: 0.000115 - momentum: 0.000000 2023-10-11 03:04:50,674 epoch 4 - iter 98/146 - loss 0.32027813 - time (sec): 64.69 - samples/sec: 479.57 - lr: 0.000113 - momentum: 0.000000 2023-10-11 03:04:59,746 epoch 4 - iter 112/146 - loss 0.30992976 - time (sec): 73.76 - samples/sec: 468.27 - lr: 0.000111 - momentum: 0.000000 2023-10-11 03:05:08,557 epoch 4 - iter 126/146 - loss 0.30555368 - time (sec): 82.57 - samples/sec: 469.25 - lr: 0.000109 - momentum: 0.000000 2023-10-11 03:05:17,259 epoch 4 - iter 140/146 - loss 0.30611365 - time (sec): 91.27 - samples/sec: 471.05 - lr: 0.000108 - momentum: 0.000000 2023-10-11 03:05:20,441 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:05:20,442 EPOCH 4 done: loss 0.3084 - lr: 0.000108 2023-10-11 03:05:26,153 DEV : loss 0.2286217361688614 - f1-score (micro avg) 0.3883 2023-10-11 03:05:26,162 saving best model 2023-10-11 03:05:27,094 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:05:36,127 epoch 5 - iter 14/146 - loss 0.25889454 - time (sec): 9.03 - samples/sec: 470.83 - lr: 0.000105 - momentum: 0.000000 2023-10-11 03:05:46,332 epoch 5 - iter 28/146 - loss 0.23524679 - time (sec): 19.24 - samples/sec: 485.34 - lr: 0.000104 - momentum: 0.000000 2023-10-11 03:05:55,679 epoch 5 - iter 42/146 - loss 0.24651874 - time (sec): 28.58 - samples/sec: 487.14 - lr: 0.000102 - momentum: 0.000000 2023-10-11 03:06:05,138 epoch 5 - iter 56/146 - loss 0.25126634 - time (sec): 38.04 - samples/sec: 487.88 - lr: 0.000100 - momentum: 0.000000 2023-10-11 03:06:14,789 epoch 5 - iter 70/146 - loss 0.27351281 - time (sec): 47.69 - samples/sec: 477.24 - lr: 0.000099 - momentum: 0.000000 2023-10-11 03:06:23,908 epoch 5 - iter 84/146 - loss 0.26826607 - time (sec): 56.81 - samples/sec: 474.74 - lr: 0.000097 - momentum: 0.000000 2023-10-11 03:06:32,783 epoch 5 - iter 98/146 - loss 0.25700981 - time (sec): 65.69 - samples/sec: 472.31 - lr: 0.000095 - momentum: 0.000000 2023-10-11 03:06:40,955 epoch 5 - iter 112/146 - loss 0.25377327 - time (sec): 73.86 - samples/sec: 465.88 - lr: 0.000093 - momentum: 0.000000 2023-10-11 03:06:50,055 epoch 5 - iter 126/146 - loss 0.24593692 - time (sec): 82.96 - samples/sec: 467.80 - lr: 0.000092 - momentum: 0.000000 2023-10-11 03:06:58,949 epoch 5 - iter 140/146 - loss 0.24041901 - time (sec): 91.85 - samples/sec: 467.84 - lr: 0.000090 - momentum: 0.000000 2023-10-11 03:07:02,398 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:07:02,398 EPOCH 5 done: loss 0.2400 - lr: 0.000090 2023-10-11 03:07:08,709 DEV : loss 0.1938922256231308 - f1-score (micro avg) 0.5096 2023-10-11 03:07:08,719 saving best model 2023-10-11 03:07:11,341 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:07:19,792 epoch 6 - iter 14/146 - loss 0.17003425 - time (sec): 8.45 - samples/sec: 421.76 - lr: 0.000088 - momentum: 0.000000 2023-10-11 03:07:29,419 epoch 6 - iter 28/146 - loss 0.21222092 - time (sec): 18.07 - samples/sec: 424.62 - lr: 0.000086 - momentum: 0.000000 2023-10-11 03:07:38,670 epoch 6 - iter 42/146 - loss 0.21021534 - time (sec): 27.32 - samples/sec: 425.94 - lr: 0.000084 - momentum: 0.000000 2023-10-11 03:07:48,456 epoch 6 - iter 56/146 - loss 0.20003050 - time (sec): 37.11 - samples/sec: 436.31 - lr: 0.000083 - momentum: 0.000000 2023-10-11 03:07:57,208 epoch 6 - iter 70/146 - loss 0.19157871 - time (sec): 45.86 - samples/sec: 443.75 - lr: 0.000081 - momentum: 0.000000 2023-10-11 03:08:06,928 epoch 6 - iter 84/146 - loss 0.17913338 - time (sec): 55.58 - samples/sec: 459.50 - lr: 0.000079 - momentum: 0.000000 2023-10-11 03:08:15,299 epoch 6 - iter 98/146 - loss 0.17940064 - time (sec): 63.95 - samples/sec: 461.31 - lr: 0.000077 - momentum: 0.000000 2023-10-11 03:08:23,678 epoch 6 - iter 112/146 - loss 0.18121308 - time (sec): 72.33 - samples/sec: 463.21 - lr: 0.000076 - momentum: 0.000000 2023-10-11 03:08:31,936 epoch 6 - iter 126/146 - loss 0.17892881 - time (sec): 80.59 - samples/sec: 461.57 - lr: 0.000074 - momentum: 0.000000 2023-10-11 03:08:41,917 epoch 6 - iter 140/146 - loss 0.18585573 - time (sec): 90.57 - samples/sec: 469.98 - lr: 0.000072 - momentum: 0.000000 2023-10-11 03:08:45,984 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:08:45,984 EPOCH 6 done: loss 0.1851 - lr: 0.000072 2023-10-11 03:08:51,662 DEV : loss 0.17115506529808044 - f1-score (micro avg) 0.566 2023-10-11 03:08:51,671 saving best model 2023-10-11 03:08:54,392 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:09:04,084 epoch 7 - iter 14/146 - loss 0.15701219 - time (sec): 9.69 - samples/sec: 391.69 - lr: 0.000070 - momentum: 0.000000 2023-10-11 03:09:12,536 epoch 7 - iter 28/146 - loss 0.15510761 - time (sec): 18.14 - samples/sec: 423.97 - lr: 0.000068 - momentum: 0.000000 2023-10-11 03:09:20,257 epoch 7 - iter 42/146 - loss 0.14629892 - time (sec): 25.86 - samples/sec: 427.51 - lr: 0.000067 - momentum: 0.000000 2023-10-11 03:09:30,238 epoch 7 - iter 56/146 - loss 0.14700964 - time (sec): 35.84 - samples/sec: 460.86 - lr: 0.000065 - momentum: 0.000000 2023-10-11 03:09:39,544 epoch 7 - iter 70/146 - loss 0.14185422 - time (sec): 45.15 - samples/sec: 455.37 - lr: 0.000063 - momentum: 0.000000 2023-10-11 03:09:48,635 epoch 7 - iter 84/146 - loss 0.14513215 - time (sec): 54.24 - samples/sec: 454.63 - lr: 0.000061 - momentum: 0.000000 2023-10-11 03:09:57,352 epoch 7 - iter 98/146 - loss 0.15171915 - time (sec): 62.96 - samples/sec: 460.05 - lr: 0.000060 - momentum: 0.000000 2023-10-11 03:10:06,611 epoch 7 - iter 112/146 - loss 0.14705100 - time (sec): 72.22 - samples/sec: 460.11 - lr: 0.000058 - momentum: 0.000000 2023-10-11 03:10:15,921 epoch 7 - iter 126/146 - loss 0.14658088 - time (sec): 81.53 - samples/sec: 462.85 - lr: 0.000056 - momentum: 0.000000 2023-10-11 03:10:25,015 epoch 7 - iter 140/146 - loss 0.14531914 - time (sec): 90.62 - samples/sec: 468.17 - lr: 0.000055 - momentum: 0.000000 2023-10-11 03:10:28,779 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:10:28,780 EPOCH 7 done: loss 0.1463 - lr: 0.000055 2023-10-11 03:10:34,791 DEV : loss 0.16687284409999847 - f1-score (micro avg) 0.617 2023-10-11 03:10:34,801 saving best model 2023-10-11 03:10:37,399 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:10:45,847 epoch 8 - iter 14/146 - loss 0.12928267 - time (sec): 8.44 - samples/sec: 470.41 - lr: 0.000052 - momentum: 0.000000 2023-10-11 03:10:54,199 epoch 8 - iter 28/146 - loss 0.12679178 - time (sec): 16.80 - samples/sec: 486.18 - lr: 0.000051 - momentum: 0.000000 2023-10-11 03:11:02,512 epoch 8 - iter 42/146 - loss 0.15315931 - time (sec): 25.11 - samples/sec: 482.89 - lr: 0.000049 - momentum: 0.000000 2023-10-11 03:11:11,760 epoch 8 - iter 56/146 - loss 0.15510801 - time (sec): 34.36 - samples/sec: 496.46 - lr: 0.000047 - momentum: 0.000000 2023-10-11 03:11:21,747 epoch 8 - iter 70/146 - loss 0.14964513 - time (sec): 44.34 - samples/sec: 479.51 - lr: 0.000045 - momentum: 0.000000 2023-10-11 03:11:30,788 epoch 8 - iter 84/146 - loss 0.14419529 - time (sec): 53.38 - samples/sec: 464.01 - lr: 0.000044 - momentum: 0.000000 2023-10-11 03:11:40,389 epoch 8 - iter 98/146 - loss 0.14165465 - time (sec): 62.99 - samples/sec: 468.01 - lr: 0.000042 - momentum: 0.000000 2023-10-11 03:11:49,920 epoch 8 - iter 112/146 - loss 0.13718145 - time (sec): 72.52 - samples/sec: 473.75 - lr: 0.000040 - momentum: 0.000000 2023-10-11 03:11:58,832 epoch 8 - iter 126/146 - loss 0.13421700 - time (sec): 81.43 - samples/sec: 467.13 - lr: 0.000039 - momentum: 0.000000 2023-10-11 03:12:08,287 epoch 8 - iter 140/146 - loss 0.12891406 - time (sec): 90.88 - samples/sec: 468.80 - lr: 0.000037 - momentum: 0.000000 2023-10-11 03:12:12,150 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:12:12,151 EPOCH 8 done: loss 0.1265 - lr: 0.000037 2023-10-11 03:12:17,999 DEV : loss 0.15703192353248596 - f1-score (micro avg) 0.6863 2023-10-11 03:12:18,008 saving best model 2023-10-11 03:12:20,577 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:12:29,531 epoch 9 - iter 14/146 - loss 0.10771080 - time (sec): 8.95 - samples/sec: 474.12 - lr: 0.000035 - momentum: 0.000000 2023-10-11 03:12:37,859 epoch 9 - iter 28/146 - loss 0.11279535 - time (sec): 17.28 - samples/sec: 460.24 - lr: 0.000033 - momentum: 0.000000 2023-10-11 03:12:46,480 epoch 9 - iter 42/146 - loss 0.12798759 - time (sec): 25.90 - samples/sec: 467.63 - lr: 0.000031 - momentum: 0.000000 2023-10-11 03:12:54,631 epoch 9 - iter 56/146 - loss 0.13087974 - time (sec): 34.05 - samples/sec: 469.99 - lr: 0.000029 - momentum: 0.000000 2023-10-11 03:13:03,571 epoch 9 - iter 70/146 - loss 0.12721638 - time (sec): 42.99 - samples/sec: 476.40 - lr: 0.000028 - momentum: 0.000000 2023-10-11 03:13:13,313 epoch 9 - iter 84/146 - loss 0.12112059 - time (sec): 52.73 - samples/sec: 482.65 - lr: 0.000026 - momentum: 0.000000 2023-10-11 03:13:21,939 epoch 9 - iter 98/146 - loss 0.11598258 - time (sec): 61.36 - samples/sec: 482.96 - lr: 0.000024 - momentum: 0.000000 2023-10-11 03:13:30,792 epoch 9 - iter 112/146 - loss 0.11367321 - time (sec): 70.21 - samples/sec: 487.31 - lr: 0.000023 - momentum: 0.000000 2023-10-11 03:13:39,412 epoch 9 - iter 126/146 - loss 0.11114356 - time (sec): 78.83 - samples/sec: 491.30 - lr: 0.000021 - momentum: 0.000000 2023-10-11 03:13:48,050 epoch 9 - iter 140/146 - loss 0.11106486 - time (sec): 87.47 - samples/sec: 490.13 - lr: 0.000019 - momentum: 0.000000 2023-10-11 03:13:51,530 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:13:51,530 EPOCH 9 done: loss 0.1103 - lr: 0.000019 2023-10-11 03:13:57,096 DEV : loss 0.1517515331506729 - f1-score (micro avg) 0.7064 2023-10-11 03:13:57,105 saving best model 2023-10-11 03:13:59,616 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:14:09,366 epoch 10 - iter 14/146 - loss 0.09910516 - time (sec): 9.75 - samples/sec: 425.43 - lr: 0.000017 - momentum: 0.000000 2023-10-11 03:14:18,329 epoch 10 - iter 28/146 - loss 0.11261227 - time (sec): 18.71 - samples/sec: 425.22 - lr: 0.000015 - momentum: 0.000000 2023-10-11 03:14:28,387 epoch 10 - iter 42/146 - loss 0.10216965 - time (sec): 28.77 - samples/sec: 443.50 - lr: 0.000013 - momentum: 0.000000 2023-10-11 03:14:37,707 epoch 10 - iter 56/146 - loss 0.09500174 - time (sec): 38.09 - samples/sec: 450.40 - lr: 0.000012 - momentum: 0.000000 2023-10-11 03:14:46,518 epoch 10 - iter 70/146 - loss 0.09562900 - time (sec): 46.90 - samples/sec: 454.01 - lr: 0.000010 - momentum: 0.000000 2023-10-11 03:14:55,546 epoch 10 - iter 84/146 - loss 0.09501906 - time (sec): 55.93 - samples/sec: 452.85 - lr: 0.000008 - momentum: 0.000000 2023-10-11 03:15:04,659 epoch 10 - iter 98/146 - loss 0.09724605 - time (sec): 65.04 - samples/sec: 457.59 - lr: 0.000007 - momentum: 0.000000 2023-10-11 03:15:15,133 epoch 10 - iter 112/146 - loss 0.10037234 - time (sec): 75.51 - samples/sec: 450.82 - lr: 0.000005 - momentum: 0.000000 2023-10-11 03:15:24,660 epoch 10 - iter 126/146 - loss 0.10048535 - time (sec): 85.04 - samples/sec: 449.83 - lr: 0.000003 - momentum: 0.000000 2023-10-11 03:15:33,705 epoch 10 - iter 140/146 - loss 0.10441213 - time (sec): 94.08 - samples/sec: 450.04 - lr: 0.000002 - momentum: 0.000000 2023-10-11 03:15:37,647 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:15:37,647 EPOCH 10 done: loss 0.1026 - lr: 0.000002 2023-10-11 03:15:43,591 DEV : loss 0.14971290528774261 - f1-score (micro avg) 0.7176 2023-10-11 03:15:43,600 saving best model 2023-10-11 03:15:47,074 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:15:47,076 Loading model from best epoch ... 2023-10-11 03:15:51,855 SequenceTagger predicts: Dictionary with 17 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd 2023-10-11 03:16:05,085 Results: - F-score (micro) 0.6815 - F-score (macro) 0.6332 - Accuracy 0.5518 By class: precision recall f1-score support PER 0.7686 0.7730 0.7708 348 LOC 0.5942 0.7011 0.6432 261 ORG 0.2763 0.4038 0.3281 52 HumanProd 0.8095 0.7727 0.7907 22 micro avg 0.6490 0.7174 0.6815 683 macro avg 0.6121 0.6627 0.6332 683 weighted avg 0.6658 0.7174 0.6890 683 2023-10-11 03:16:05,086 ----------------------------------------------------------------------------------------------------