2023-10-11 11:42:52,153 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:42:52,155 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 11:42:52,155 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:42:52,156 MultiCorpus: 1085 train + 148 dev + 364 test sentences - NER_HIPE_2022 Corpus: 1085 train + 148 dev + 364 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/sv/with_doc_seperator 2023-10-11 11:42:52,156 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:42:52,156 Train: 1085 sentences 2023-10-11 11:42:52,156 (train_with_dev=False, train_with_test=False) 2023-10-11 11:42:52,156 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:42:52,156 Training Params: 2023-10-11 11:42:52,156 - learning_rate: "0.00016" 2023-10-11 11:42:52,156 - mini_batch_size: "8" 2023-10-11 11:42:52,156 - max_epochs: "10" 2023-10-11 11:42:52,156 - shuffle: "True" 2023-10-11 11:42:52,156 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:42:52,156 Plugins: 2023-10-11 11:42:52,156 - TensorboardLogger 2023-10-11 11:42:52,157 - LinearScheduler | warmup_fraction: '0.1' 2023-10-11 11:42:52,157 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:42:52,157 Final evaluation on model from best epoch (best-model.pt) 2023-10-11 11:42:52,157 - metric: "('micro avg', 'f1-score')" 2023-10-11 11:42:52,157 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:42:52,157 Computation: 2023-10-11 11:42:52,157 - compute on device: cuda:0 2023-10-11 11:42:52,157 - embedding storage: none 2023-10-11 11:42:52,157 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:42:52,157 Model training base path: "hmbench-newseye/sv-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-4" 2023-10-11 11:42:52,157 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:42:52,157 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:42:52,157 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-11 11:43:00,927 epoch 1 - iter 13/136 - loss 2.84089892 - time (sec): 8.77 - samples/sec: 615.10 - lr: 0.000014 - momentum: 0.000000 2023-10-11 11:43:10,129 epoch 1 - iter 26/136 - loss 2.83498049 - time (sec): 17.97 - samples/sec: 615.85 - lr: 0.000029 - momentum: 0.000000 2023-10-11 11:43:18,889 epoch 1 - iter 39/136 - loss 2.82406903 - time (sec): 26.73 - samples/sec: 601.04 - lr: 0.000045 - momentum: 0.000000 2023-10-11 11:43:27,568 epoch 1 - iter 52/136 - loss 2.80264022 - time (sec): 35.41 - samples/sec: 593.72 - lr: 0.000060 - momentum: 0.000000 2023-10-11 11:43:35,878 epoch 1 - iter 65/136 - loss 2.76592580 - time (sec): 43.72 - samples/sec: 582.84 - lr: 0.000075 - momentum: 0.000000 2023-10-11 11:43:44,185 epoch 1 - iter 78/136 - loss 2.71169051 - time (sec): 52.03 - samples/sec: 577.85 - lr: 0.000091 - momentum: 0.000000 2023-10-11 11:43:52,478 epoch 1 - iter 91/136 - loss 2.64797024 - time (sec): 60.32 - samples/sec: 571.71 - lr: 0.000106 - momentum: 0.000000 2023-10-11 11:44:01,392 epoch 1 - iter 104/136 - loss 2.56715392 - time (sec): 69.23 - samples/sec: 576.23 - lr: 0.000121 - momentum: 0.000000 2023-10-11 11:44:09,241 epoch 1 - iter 117/136 - loss 2.50180341 - time (sec): 77.08 - samples/sec: 572.72 - lr: 0.000136 - momentum: 0.000000 2023-10-11 11:44:18,110 epoch 1 - iter 130/136 - loss 2.41593244 - time (sec): 85.95 - samples/sec: 571.04 - lr: 0.000152 - momentum: 0.000000 2023-10-11 11:44:22,531 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:44:22,532 EPOCH 1 done: loss 2.3622 - lr: 0.000152 2023-10-11 11:44:27,578 DEV : loss 1.331149935722351 - f1-score (micro avg) 0.0 2023-10-11 11:44:27,587 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:44:36,368 epoch 2 - iter 13/136 - loss 1.32887985 - time (sec): 8.78 - samples/sec: 602.57 - lr: 0.000158 - momentum: 0.000000 2023-10-11 11:44:44,636 epoch 2 - iter 26/136 - loss 1.26733038 - time (sec): 17.05 - samples/sec: 574.22 - lr: 0.000157 - momentum: 0.000000 2023-10-11 11:44:52,303 epoch 2 - iter 39/136 - loss 1.20119028 - time (sec): 24.71 - samples/sec: 550.00 - lr: 0.000155 - momentum: 0.000000 2023-10-11 11:45:01,401 epoch 2 - iter 52/136 - loss 1.10723903 - time (sec): 33.81 - samples/sec: 560.91 - lr: 0.000153 - momentum: 0.000000 2023-10-11 11:45:11,107 epoch 2 - iter 65/136 - loss 1.02852991 - time (sec): 43.52 - samples/sec: 563.78 - lr: 0.000152 - momentum: 0.000000 2023-10-11 11:45:20,058 epoch 2 - iter 78/136 - loss 0.96147737 - time (sec): 52.47 - samples/sec: 564.12 - lr: 0.000150 - momentum: 0.000000 2023-10-11 11:45:28,380 epoch 2 - iter 91/136 - loss 0.90827453 - time (sec): 60.79 - samples/sec: 560.43 - lr: 0.000148 - momentum: 0.000000 2023-10-11 11:45:36,838 epoch 2 - iter 104/136 - loss 0.85911401 - time (sec): 69.25 - samples/sec: 560.82 - lr: 0.000147 - momentum: 0.000000 2023-10-11 11:45:44,748 epoch 2 - iter 117/136 - loss 0.83120687 - time (sec): 77.16 - samples/sec: 559.18 - lr: 0.000145 - momentum: 0.000000 2023-10-11 11:45:54,150 epoch 2 - iter 130/136 - loss 0.79282972 - time (sec): 86.56 - samples/sec: 570.10 - lr: 0.000143 - momentum: 0.000000 2023-10-11 11:45:58,234 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:45:58,235 EPOCH 2 done: loss 0.7761 - lr: 0.000143 2023-10-11 11:46:03,660 DEV : loss 0.3906053304672241 - f1-score (micro avg) 0.0 2023-10-11 11:46:03,668 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:46:12,037 epoch 3 - iter 13/136 - loss 0.34739828 - time (sec): 8.37 - samples/sec: 556.71 - lr: 0.000141 - momentum: 0.000000 2023-10-11 11:46:21,281 epoch 3 - iter 26/136 - loss 0.39816021 - time (sec): 17.61 - samples/sec: 600.01 - lr: 0.000139 - momentum: 0.000000 2023-10-11 11:46:29,696 epoch 3 - iter 39/136 - loss 0.40218390 - time (sec): 26.03 - samples/sec: 601.13 - lr: 0.000137 - momentum: 0.000000 2023-10-11 11:46:38,051 epoch 3 - iter 52/136 - loss 0.40576799 - time (sec): 34.38 - samples/sec: 595.88 - lr: 0.000136 - momentum: 0.000000 2023-10-11 11:46:46,295 epoch 3 - iter 65/136 - loss 0.39917278 - time (sec): 42.63 - samples/sec: 592.30 - lr: 0.000134 - momentum: 0.000000 2023-10-11 11:46:54,812 epoch 3 - iter 78/136 - loss 0.39934212 - time (sec): 51.14 - samples/sec: 592.05 - lr: 0.000132 - momentum: 0.000000 2023-10-11 11:47:03,667 epoch 3 - iter 91/136 - loss 0.39725495 - time (sec): 60.00 - samples/sec: 596.39 - lr: 0.000131 - momentum: 0.000000 2023-10-11 11:47:11,817 epoch 3 - iter 104/136 - loss 0.39381516 - time (sec): 68.15 - samples/sec: 589.08 - lr: 0.000129 - momentum: 0.000000 2023-10-11 11:47:20,374 epoch 3 - iter 117/136 - loss 0.38541873 - time (sec): 76.70 - samples/sec: 590.97 - lr: 0.000127 - momentum: 0.000000 2023-10-11 11:47:28,386 epoch 3 - iter 130/136 - loss 0.38481317 - time (sec): 84.72 - samples/sec: 587.10 - lr: 0.000126 - momentum: 0.000000 2023-10-11 11:47:32,136 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:47:32,136 EPOCH 3 done: loss 0.3826 - lr: 0.000126 2023-10-11 11:47:37,658 DEV : loss 0.2841358780860901 - f1-score (micro avg) 0.303 2023-10-11 11:47:37,666 saving best model 2023-10-11 11:47:38,500 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:47:46,931 epoch 4 - iter 13/136 - loss 0.30981951 - time (sec): 8.43 - samples/sec: 565.84 - lr: 0.000123 - momentum: 0.000000 2023-10-11 11:47:55,313 epoch 4 - iter 26/136 - loss 0.26209209 - time (sec): 16.81 - samples/sec: 572.91 - lr: 0.000121 - momentum: 0.000000 2023-10-11 11:48:04,432 epoch 4 - iter 39/136 - loss 0.29742145 - time (sec): 25.93 - samples/sec: 602.75 - lr: 0.000120 - momentum: 0.000000 2023-10-11 11:48:12,946 epoch 4 - iter 52/136 - loss 0.30087458 - time (sec): 34.44 - samples/sec: 600.55 - lr: 0.000118 - momentum: 0.000000 2023-10-11 11:48:21,575 epoch 4 - iter 65/136 - loss 0.29630507 - time (sec): 43.07 - samples/sec: 597.25 - lr: 0.000116 - momentum: 0.000000 2023-10-11 11:48:30,045 epoch 4 - iter 78/136 - loss 0.29026697 - time (sec): 51.54 - samples/sec: 593.86 - lr: 0.000115 - momentum: 0.000000 2023-10-11 11:48:38,662 epoch 4 - iter 91/136 - loss 0.29000751 - time (sec): 60.16 - samples/sec: 596.55 - lr: 0.000113 - momentum: 0.000000 2023-10-11 11:48:46,927 epoch 4 - iter 104/136 - loss 0.28862524 - time (sec): 68.42 - samples/sec: 591.68 - lr: 0.000111 - momentum: 0.000000 2023-10-11 11:48:55,506 epoch 4 - iter 117/136 - loss 0.29291057 - time (sec): 77.00 - samples/sec: 588.96 - lr: 0.000109 - momentum: 0.000000 2023-10-11 11:49:03,284 epoch 4 - iter 130/136 - loss 0.29357331 - time (sec): 84.78 - samples/sec: 580.03 - lr: 0.000108 - momentum: 0.000000 2023-10-11 11:49:07,528 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:49:07,528 EPOCH 4 done: loss 0.2924 - lr: 0.000108 2023-10-11 11:49:13,047 DEV : loss 0.24194754660129547 - f1-score (micro avg) 0.361 2023-10-11 11:49:13,055 saving best model 2023-10-11 11:49:15,756 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:49:24,571 epoch 5 - iter 13/136 - loss 0.25599446 - time (sec): 8.81 - samples/sec: 607.84 - lr: 0.000105 - momentum: 0.000000 2023-10-11 11:49:33,041 epoch 5 - iter 26/136 - loss 0.26637160 - time (sec): 17.28 - samples/sec: 592.68 - lr: 0.000104 - momentum: 0.000000 2023-10-11 11:49:41,925 epoch 5 - iter 39/136 - loss 0.26414833 - time (sec): 26.16 - samples/sec: 599.21 - lr: 0.000102 - momentum: 0.000000 2023-10-11 11:49:51,071 epoch 5 - iter 52/136 - loss 0.25660484 - time (sec): 35.31 - samples/sec: 606.04 - lr: 0.000100 - momentum: 0.000000 2023-10-11 11:49:59,386 epoch 5 - iter 65/136 - loss 0.25490717 - time (sec): 43.63 - samples/sec: 600.20 - lr: 0.000099 - momentum: 0.000000 2023-10-11 11:50:07,873 epoch 5 - iter 78/136 - loss 0.25012610 - time (sec): 52.11 - samples/sec: 591.76 - lr: 0.000097 - momentum: 0.000000 2023-10-11 11:50:16,091 epoch 5 - iter 91/136 - loss 0.24408605 - time (sec): 60.33 - samples/sec: 585.86 - lr: 0.000095 - momentum: 0.000000 2023-10-11 11:50:24,451 epoch 5 - iter 104/136 - loss 0.24082052 - time (sec): 68.69 - samples/sec: 583.79 - lr: 0.000093 - momentum: 0.000000 2023-10-11 11:50:32,782 epoch 5 - iter 117/136 - loss 0.23933921 - time (sec): 77.02 - samples/sec: 581.82 - lr: 0.000092 - momentum: 0.000000 2023-10-11 11:50:41,368 epoch 5 - iter 130/136 - loss 0.23509442 - time (sec): 85.61 - samples/sec: 581.75 - lr: 0.000090 - momentum: 0.000000 2023-10-11 11:50:45,040 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:50:45,041 EPOCH 5 done: loss 0.2334 - lr: 0.000090 2023-10-11 11:50:50,569 DEV : loss 0.2023283690214157 - f1-score (micro avg) 0.5352 2023-10-11 11:50:50,577 saving best model 2023-10-11 11:50:53,072 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:51:01,826 epoch 6 - iter 13/136 - loss 0.18517967 - time (sec): 8.75 - samples/sec: 571.85 - lr: 0.000088 - momentum: 0.000000 2023-10-11 11:51:10,387 epoch 6 - iter 26/136 - loss 0.18845994 - time (sec): 17.31 - samples/sec: 580.12 - lr: 0.000086 - momentum: 0.000000 2023-10-11 11:51:19,918 epoch 6 - iter 39/136 - loss 0.18613737 - time (sec): 26.84 - samples/sec: 598.85 - lr: 0.000084 - momentum: 0.000000 2023-10-11 11:51:27,768 epoch 6 - iter 52/136 - loss 0.18257001 - time (sec): 34.69 - samples/sec: 587.24 - lr: 0.000083 - momentum: 0.000000 2023-10-11 11:51:35,920 epoch 6 - iter 65/136 - loss 0.17864499 - time (sec): 42.85 - samples/sec: 585.99 - lr: 0.000081 - momentum: 0.000000 2023-10-11 11:51:44,330 epoch 6 - iter 78/136 - loss 0.17689256 - time (sec): 51.26 - samples/sec: 584.99 - lr: 0.000079 - momentum: 0.000000 2023-10-11 11:51:52,541 epoch 6 - iter 91/136 - loss 0.18469652 - time (sec): 59.47 - samples/sec: 586.24 - lr: 0.000077 - momentum: 0.000000 2023-10-11 11:52:00,605 epoch 6 - iter 104/136 - loss 0.18264529 - time (sec): 67.53 - samples/sec: 582.48 - lr: 0.000076 - momentum: 0.000000 2023-10-11 11:52:09,270 epoch 6 - iter 117/136 - loss 0.18078056 - time (sec): 76.19 - samples/sec: 585.17 - lr: 0.000074 - momentum: 0.000000 2023-10-11 11:52:18,305 epoch 6 - iter 130/136 - loss 0.17744831 - time (sec): 85.23 - samples/sec: 589.03 - lr: 0.000072 - momentum: 0.000000 2023-10-11 11:52:21,741 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:52:21,741 EPOCH 6 done: loss 0.1785 - lr: 0.000072 2023-10-11 11:52:27,298 DEV : loss 0.18196691572666168 - f1-score (micro avg) 0.6184 2023-10-11 11:52:27,306 saving best model 2023-10-11 11:52:29,802 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:52:38,343 epoch 7 - iter 13/136 - loss 0.13573850 - time (sec): 8.54 - samples/sec: 550.59 - lr: 0.000070 - momentum: 0.000000 2023-10-11 11:52:46,362 epoch 7 - iter 26/136 - loss 0.15154756 - time (sec): 16.56 - samples/sec: 539.09 - lr: 0.000068 - momentum: 0.000000 2023-10-11 11:52:54,607 epoch 7 - iter 39/136 - loss 0.14147926 - time (sec): 24.80 - samples/sec: 545.99 - lr: 0.000067 - momentum: 0.000000 2023-10-11 11:53:02,804 epoch 7 - iter 52/136 - loss 0.14239010 - time (sec): 33.00 - samples/sec: 544.29 - lr: 0.000065 - momentum: 0.000000 2023-10-11 11:53:13,375 epoch 7 - iter 65/136 - loss 0.13702337 - time (sec): 43.57 - samples/sec: 546.13 - lr: 0.000063 - momentum: 0.000000 2023-10-11 11:53:22,841 epoch 7 - iter 78/136 - loss 0.13307308 - time (sec): 53.03 - samples/sec: 556.81 - lr: 0.000061 - momentum: 0.000000 2023-10-11 11:53:31,166 epoch 7 - iter 91/136 - loss 0.13621321 - time (sec): 61.36 - samples/sec: 553.56 - lr: 0.000060 - momentum: 0.000000 2023-10-11 11:53:39,831 epoch 7 - iter 104/136 - loss 0.13813726 - time (sec): 70.02 - samples/sec: 553.89 - lr: 0.000058 - momentum: 0.000000 2023-10-11 11:53:48,877 epoch 7 - iter 117/136 - loss 0.13858223 - time (sec): 79.07 - samples/sec: 560.08 - lr: 0.000056 - momentum: 0.000000 2023-10-11 11:53:57,820 epoch 7 - iter 130/136 - loss 0.13942011 - time (sec): 88.01 - samples/sec: 563.86 - lr: 0.000055 - momentum: 0.000000 2023-10-11 11:54:01,781 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:54:01,781 EPOCH 7 done: loss 0.1413 - lr: 0.000055 2023-10-11 11:54:07,495 DEV : loss 0.1657494753599167 - f1-score (micro avg) 0.6118 2023-10-11 11:54:07,504 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:54:16,368 epoch 8 - iter 13/136 - loss 0.09943176 - time (sec): 8.86 - samples/sec: 569.40 - lr: 0.000052 - momentum: 0.000000 2023-10-11 11:54:25,369 epoch 8 - iter 26/136 - loss 0.12241690 - time (sec): 17.86 - samples/sec: 583.50 - lr: 0.000051 - momentum: 0.000000 2023-10-11 11:54:34,298 epoch 8 - iter 39/136 - loss 0.12160070 - time (sec): 26.79 - samples/sec: 591.13 - lr: 0.000049 - momentum: 0.000000 2023-10-11 11:54:42,211 epoch 8 - iter 52/136 - loss 0.11904427 - time (sec): 34.71 - samples/sec: 571.27 - lr: 0.000047 - momentum: 0.000000 2023-10-11 11:54:51,653 epoch 8 - iter 65/136 - loss 0.11543585 - time (sec): 44.15 - samples/sec: 581.03 - lr: 0.000045 - momentum: 0.000000 2023-10-11 11:55:01,140 epoch 8 - iter 78/136 - loss 0.11519079 - time (sec): 53.63 - samples/sec: 590.70 - lr: 0.000044 - momentum: 0.000000 2023-10-11 11:55:09,607 epoch 8 - iter 91/136 - loss 0.11768833 - time (sec): 62.10 - samples/sec: 583.56 - lr: 0.000042 - momentum: 0.000000 2023-10-11 11:55:17,593 epoch 8 - iter 104/136 - loss 0.11737793 - time (sec): 70.09 - samples/sec: 577.78 - lr: 0.000040 - momentum: 0.000000 2023-10-11 11:55:26,457 epoch 8 - iter 117/136 - loss 0.11880016 - time (sec): 78.95 - samples/sec: 580.16 - lr: 0.000039 - momentum: 0.000000 2023-10-11 11:55:34,585 epoch 8 - iter 130/136 - loss 0.11635025 - time (sec): 87.08 - samples/sec: 577.11 - lr: 0.000037 - momentum: 0.000000 2023-10-11 11:55:37,961 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:55:37,961 EPOCH 8 done: loss 0.1166 - lr: 0.000037 2023-10-11 11:55:43,939 DEV : loss 0.1629943549633026 - f1-score (micro avg) 0.6403 2023-10-11 11:55:43,947 saving best model 2023-10-11 11:55:46,470 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:55:54,803 epoch 9 - iter 13/136 - loss 0.12409007 - time (sec): 8.33 - samples/sec: 568.60 - lr: 0.000034 - momentum: 0.000000 2023-10-11 11:56:02,461 epoch 9 - iter 26/136 - loss 0.12338939 - time (sec): 15.99 - samples/sec: 549.75 - lr: 0.000033 - momentum: 0.000000 2023-10-11 11:56:10,655 epoch 9 - iter 39/136 - loss 0.11349678 - time (sec): 24.18 - samples/sec: 555.76 - lr: 0.000031 - momentum: 0.000000 2023-10-11 11:56:19,197 epoch 9 - iter 52/136 - loss 0.10693226 - time (sec): 32.72 - samples/sec: 566.11 - lr: 0.000029 - momentum: 0.000000 2023-10-11 11:56:28,242 epoch 9 - iter 65/136 - loss 0.10839615 - time (sec): 41.77 - samples/sec: 576.28 - lr: 0.000028 - momentum: 0.000000 2023-10-11 11:56:36,534 epoch 9 - iter 78/136 - loss 0.10371986 - time (sec): 50.06 - samples/sec: 575.29 - lr: 0.000026 - momentum: 0.000000 2023-10-11 11:56:45,697 epoch 9 - iter 91/136 - loss 0.09994810 - time (sec): 59.22 - samples/sec: 578.63 - lr: 0.000024 - momentum: 0.000000 2023-10-11 11:56:54,262 epoch 9 - iter 104/136 - loss 0.10282509 - time (sec): 67.79 - samples/sec: 577.37 - lr: 0.000023 - momentum: 0.000000 2023-10-11 11:57:03,462 epoch 9 - iter 117/136 - loss 0.10471070 - time (sec): 76.99 - samples/sec: 580.27 - lr: 0.000021 - momentum: 0.000000 2023-10-11 11:57:12,384 epoch 9 - iter 130/136 - loss 0.10371980 - time (sec): 85.91 - samples/sec: 582.84 - lr: 0.000019 - momentum: 0.000000 2023-10-11 11:57:15,903 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:57:15,903 EPOCH 9 done: loss 0.1028 - lr: 0.000019 2023-10-11 11:57:21,970 DEV : loss 0.15900003910064697 - f1-score (micro avg) 0.6486 2023-10-11 11:57:21,979 saving best model 2023-10-11 11:57:24,526 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:57:33,259 epoch 10 - iter 13/136 - loss 0.08673914 - time (sec): 8.73 - samples/sec: 577.96 - lr: 0.000017 - momentum: 0.000000 2023-10-11 11:57:41,971 epoch 10 - iter 26/136 - loss 0.10283299 - time (sec): 17.44 - samples/sec: 581.89 - lr: 0.000015 - momentum: 0.000000 2023-10-11 11:57:50,068 epoch 10 - iter 39/136 - loss 0.10096336 - time (sec): 25.54 - samples/sec: 576.52 - lr: 0.000013 - momentum: 0.000000 2023-10-11 11:57:58,268 epoch 10 - iter 52/136 - loss 0.10508712 - time (sec): 33.74 - samples/sec: 576.09 - lr: 0.000012 - momentum: 0.000000 2023-10-11 11:58:07,497 epoch 10 - iter 65/136 - loss 0.09663908 - time (sec): 42.97 - samples/sec: 582.26 - lr: 0.000010 - momentum: 0.000000 2023-10-11 11:58:15,641 epoch 10 - iter 78/136 - loss 0.09515246 - time (sec): 51.11 - samples/sec: 576.29 - lr: 0.000008 - momentum: 0.000000 2023-10-11 11:58:24,729 epoch 10 - iter 91/136 - loss 0.09381065 - time (sec): 60.20 - samples/sec: 577.73 - lr: 0.000007 - momentum: 0.000000 2023-10-11 11:58:33,569 epoch 10 - iter 104/136 - loss 0.09477746 - time (sec): 69.04 - samples/sec: 575.19 - lr: 0.000005 - momentum: 0.000000 2023-10-11 11:58:42,433 epoch 10 - iter 117/136 - loss 0.09519411 - time (sec): 77.90 - samples/sec: 577.00 - lr: 0.000003 - momentum: 0.000000 2023-10-11 11:58:51,065 epoch 10 - iter 130/136 - loss 0.09571016 - time (sec): 86.54 - samples/sec: 576.79 - lr: 0.000002 - momentum: 0.000000 2023-10-11 11:58:54,747 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:58:54,747 EPOCH 10 done: loss 0.0960 - lr: 0.000002 2023-10-11 11:59:00,673 DEV : loss 0.16201113164424896 - f1-score (micro avg) 0.6306 2023-10-11 11:59:01,568 ---------------------------------------------------------------------------------------------------- 2023-10-11 11:59:01,570 Loading model from best epoch ... 2023-10-11 11:59:05,259 SequenceTagger predicts: Dictionary with 17 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd, S-ORG, B-ORG, E-ORG, I-ORG 2023-10-11 11:59:17,559 Results: - F-score (micro) 0.6285 - F-score (macro) 0.4257 - Accuracy 0.5154 By class: precision recall f1-score support LOC 0.6256 0.8462 0.7193 312 PER 0.6557 0.5769 0.6138 208 HumanProd 0.2429 0.7727 0.3696 22 ORG 0.0000 0.0000 0.0000 55 micro avg 0.5906 0.6717 0.6285 597 macro avg 0.3810 0.5490 0.4257 597 weighted avg 0.5644 0.6717 0.6034 597 2023-10-11 11:59:17,559 ----------------------------------------------------------------------------------------------------