2023-10-11 03:35:43,703 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:35:43,706 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 03:35:43,706 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:35:43,707 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 03:35:43,707 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:35:43,707 Train: 1166 sentences 2023-10-11 03:35:43,707 (train_with_dev=False, train_with_test=False) 2023-10-11 03:35:43,707 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:35:43,707 Training Params: 2023-10-11 03:35:43,707 - learning_rate: "0.00016" 2023-10-11 03:35:43,707 - mini_batch_size: "4" 2023-10-11 03:35:43,707 - max_epochs: "10" 2023-10-11 03:35:43,707 - shuffle: "True" 2023-10-11 03:35:43,707 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:35:43,707 Plugins: 2023-10-11 03:35:43,707 - TensorboardLogger 2023-10-11 03:35:43,707 - LinearScheduler | warmup_fraction: '0.1' 2023-10-11 03:35:43,708 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:35:43,708 Final evaluation on model from best epoch (best-model.pt) 2023-10-11 03:35:43,708 - metric: "('micro avg', 'f1-score')" 2023-10-11 03:35:43,708 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:35:43,708 Computation: 2023-10-11 03:35:43,708 - compute on device: cuda:0 2023-10-11 03:35:43,708 - embedding storage: none 2023-10-11 03:35:43,708 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:35:43,708 Model training base path: "hmbench-newseye/fi-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-5" 2023-10-11 03:35:43,708 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:35:43,708 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:35:43,708 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-11 03:35:54,860 epoch 1 - iter 29/292 - loss 2.81957970 - time (sec): 11.15 - samples/sec: 386.73 - lr: 0.000015 - momentum: 0.000000 2023-10-11 03:36:04,709 epoch 1 - iter 58/292 - loss 2.80589817 - time (sec): 21.00 - samples/sec: 393.07 - lr: 0.000031 - momentum: 0.000000 2023-10-11 03:36:14,324 epoch 1 - iter 87/292 - loss 2.77951816 - time (sec): 30.61 - samples/sec: 399.20 - lr: 0.000047 - momentum: 0.000000 2023-10-11 03:36:26,204 epoch 1 - iter 116/292 - loss 2.70206612 - time (sec): 42.49 - samples/sec: 411.23 - lr: 0.000063 - momentum: 0.000000 2023-10-11 03:36:35,857 epoch 1 - iter 145/292 - loss 2.61368049 - time (sec): 52.15 - samples/sec: 418.15 - lr: 0.000079 - momentum: 0.000000 2023-10-11 03:36:46,142 epoch 1 - iter 174/292 - loss 2.50306929 - time (sec): 62.43 - samples/sec: 428.31 - lr: 0.000095 - momentum: 0.000000 2023-10-11 03:36:57,372 epoch 1 - iter 203/292 - loss 2.37448119 - time (sec): 73.66 - samples/sec: 438.29 - lr: 0.000111 - momentum: 0.000000 2023-10-11 03:37:06,746 epoch 1 - iter 232/292 - loss 2.26594986 - time (sec): 83.04 - samples/sec: 438.73 - lr: 0.000127 - momentum: 0.000000 2023-10-11 03:37:16,335 epoch 1 - iter 261/292 - loss 2.14370781 - time (sec): 92.63 - samples/sec: 439.30 - lr: 0.000142 - momentum: 0.000000 2023-10-11 03:37:25,653 epoch 1 - iter 290/292 - loss 2.03951691 - time (sec): 101.94 - samples/sec: 434.93 - lr: 0.000158 - momentum: 0.000000 2023-10-11 03:37:26,081 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:37:26,082 EPOCH 1 done: loss 2.0362 - lr: 0.000158 2023-10-11 03:37:31,472 DEV : loss 0.6394248604774475 - f1-score (micro avg) 0.0 2023-10-11 03:37:31,482 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:37:41,271 epoch 2 - iter 29/292 - loss 0.68301608 - time (sec): 9.79 - samples/sec: 447.23 - lr: 0.000158 - momentum: 0.000000 2023-10-11 03:37:50,950 epoch 2 - iter 58/292 - loss 0.63527545 - time (sec): 19.47 - samples/sec: 440.97 - lr: 0.000157 - momentum: 0.000000 2023-10-11 03:38:00,592 epoch 2 - iter 87/292 - loss 0.63132550 - time (sec): 29.11 - samples/sec: 436.41 - lr: 0.000155 - momentum: 0.000000 2023-10-11 03:38:10,670 epoch 2 - iter 116/292 - loss 0.58196696 - time (sec): 39.19 - samples/sec: 443.97 - lr: 0.000153 - momentum: 0.000000 2023-10-11 03:38:20,932 epoch 2 - iter 145/292 - loss 0.60854308 - time (sec): 49.45 - samples/sec: 450.09 - lr: 0.000151 - momentum: 0.000000 2023-10-11 03:38:31,538 epoch 2 - iter 174/292 - loss 0.57661614 - time (sec): 60.05 - samples/sec: 445.20 - lr: 0.000149 - momentum: 0.000000 2023-10-11 03:38:41,306 epoch 2 - iter 203/292 - loss 0.54824891 - time (sec): 69.82 - samples/sec: 447.23 - lr: 0.000148 - momentum: 0.000000 2023-10-11 03:38:50,123 epoch 2 - iter 232/292 - loss 0.52349347 - time (sec): 78.64 - samples/sec: 443.56 - lr: 0.000146 - momentum: 0.000000 2023-10-11 03:38:59,450 epoch 2 - iter 261/292 - loss 0.51475348 - time (sec): 87.97 - samples/sec: 443.06 - lr: 0.000144 - momentum: 0.000000 2023-10-11 03:39:09,649 epoch 2 - iter 290/292 - loss 0.49892401 - time (sec): 98.16 - samples/sec: 450.27 - lr: 0.000142 - momentum: 0.000000 2023-10-11 03:39:10,190 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:39:10,190 EPOCH 2 done: loss 0.4983 - lr: 0.000142 2023-10-11 03:39:16,070 DEV : loss 0.29269105195999146 - f1-score (micro avg) 0.0 2023-10-11 03:39:16,079 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:39:25,841 epoch 3 - iter 29/292 - loss 0.35386531 - time (sec): 9.76 - samples/sec: 408.81 - lr: 0.000141 - momentum: 0.000000 2023-10-11 03:39:35,418 epoch 3 - iter 58/292 - loss 0.35190101 - time (sec): 19.34 - samples/sec: 411.64 - lr: 0.000139 - momentum: 0.000000 2023-10-11 03:39:45,503 epoch 3 - iter 87/292 - loss 0.33548860 - time (sec): 29.42 - samples/sec: 425.46 - lr: 0.000137 - momentum: 0.000000 2023-10-11 03:39:55,406 epoch 3 - iter 116/292 - loss 0.31617685 - time (sec): 39.32 - samples/sec: 437.00 - lr: 0.000135 - momentum: 0.000000 2023-10-11 03:40:05,107 epoch 3 - iter 145/292 - loss 0.30835528 - time (sec): 49.03 - samples/sec: 434.75 - lr: 0.000133 - momentum: 0.000000 2023-10-11 03:40:15,147 epoch 3 - iter 174/292 - loss 0.29080066 - time (sec): 59.07 - samples/sec: 441.22 - lr: 0.000132 - momentum: 0.000000 2023-10-11 03:40:25,273 epoch 3 - iter 203/292 - loss 0.30465050 - time (sec): 69.19 - samples/sec: 443.61 - lr: 0.000130 - momentum: 0.000000 2023-10-11 03:40:34,384 epoch 3 - iter 232/292 - loss 0.30244144 - time (sec): 78.30 - samples/sec: 442.32 - lr: 0.000128 - momentum: 0.000000 2023-10-11 03:40:45,048 epoch 3 - iter 261/292 - loss 0.29585035 - time (sec): 88.97 - samples/sec: 446.66 - lr: 0.000126 - momentum: 0.000000 2023-10-11 03:40:55,141 epoch 3 - iter 290/292 - loss 0.29281659 - time (sec): 99.06 - samples/sec: 445.33 - lr: 0.000125 - momentum: 0.000000 2023-10-11 03:40:55,768 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:40:55,768 EPOCH 3 done: loss 0.2912 - lr: 0.000125 2023-10-11 03:41:01,875 DEV : loss 0.19170591235160828 - f1-score (micro avg) 0.477 2023-10-11 03:41:01,885 saving best model 2023-10-11 03:41:02,794 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:41:12,772 epoch 4 - iter 29/292 - loss 0.19259566 - time (sec): 9.98 - samples/sec: 473.64 - lr: 0.000123 - momentum: 0.000000 2023-10-11 03:41:23,316 epoch 4 - iter 58/292 - loss 0.16708108 - time (sec): 20.52 - samples/sec: 490.02 - lr: 0.000121 - momentum: 0.000000 2023-10-11 03:41:33,104 epoch 4 - iter 87/292 - loss 0.19487845 - time (sec): 30.31 - samples/sec: 484.26 - lr: 0.000119 - momentum: 0.000000 2023-10-11 03:41:42,943 epoch 4 - iter 116/292 - loss 0.19940283 - time (sec): 40.15 - samples/sec: 483.38 - lr: 0.000117 - momentum: 0.000000 2023-10-11 03:41:52,477 epoch 4 - iter 145/292 - loss 0.20218637 - time (sec): 49.68 - samples/sec: 479.92 - lr: 0.000116 - momentum: 0.000000 2023-10-11 03:42:01,818 epoch 4 - iter 174/292 - loss 0.19572403 - time (sec): 59.02 - samples/sec: 475.35 - lr: 0.000114 - momentum: 0.000000 2023-10-11 03:42:10,954 epoch 4 - iter 203/292 - loss 0.19603038 - time (sec): 68.16 - samples/sec: 468.68 - lr: 0.000112 - momentum: 0.000000 2023-10-11 03:42:20,751 epoch 4 - iter 232/292 - loss 0.19062971 - time (sec): 77.95 - samples/sec: 463.96 - lr: 0.000110 - momentum: 0.000000 2023-10-11 03:42:29,640 epoch 4 - iter 261/292 - loss 0.18725691 - time (sec): 86.84 - samples/sec: 455.07 - lr: 0.000109 - momentum: 0.000000 2023-10-11 03:42:39,637 epoch 4 - iter 290/292 - loss 0.18640083 - time (sec): 96.84 - samples/sec: 457.68 - lr: 0.000107 - momentum: 0.000000 2023-10-11 03:42:40,049 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:42:40,050 EPOCH 4 done: loss 0.1868 - lr: 0.000107 2023-10-11 03:42:45,788 DEV : loss 0.1456957757472992 - f1-score (micro avg) 0.6624 2023-10-11 03:42:45,798 saving best model 2023-10-11 03:42:46,754 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:42:56,570 epoch 5 - iter 29/292 - loss 0.13994720 - time (sec): 9.81 - samples/sec: 456.00 - lr: 0.000105 - momentum: 0.000000 2023-10-11 03:43:06,990 epoch 5 - iter 58/292 - loss 0.11776309 - time (sec): 20.23 - samples/sec: 472.13 - lr: 0.000103 - momentum: 0.000000 2023-10-11 03:43:17,064 epoch 5 - iter 87/292 - loss 0.13368301 - time (sec): 30.31 - samples/sec: 472.38 - lr: 0.000101 - momentum: 0.000000 2023-10-11 03:43:27,262 epoch 5 - iter 116/292 - loss 0.13287921 - time (sec): 40.51 - samples/sec: 471.51 - lr: 0.000100 - momentum: 0.000000 2023-10-11 03:43:36,983 epoch 5 - iter 145/292 - loss 0.13471707 - time (sec): 50.23 - samples/sec: 469.55 - lr: 0.000098 - momentum: 0.000000 2023-10-11 03:43:46,317 epoch 5 - iter 174/292 - loss 0.13352509 - time (sec): 59.56 - samples/sec: 467.45 - lr: 0.000096 - momentum: 0.000000 2023-10-11 03:43:56,006 epoch 5 - iter 203/292 - loss 0.13150760 - time (sec): 69.25 - samples/sec: 458.57 - lr: 0.000094 - momentum: 0.000000 2023-10-11 03:44:04,831 epoch 5 - iter 232/292 - loss 0.12973404 - time (sec): 78.07 - samples/sec: 454.17 - lr: 0.000093 - momentum: 0.000000 2023-10-11 03:44:14,924 epoch 5 - iter 261/292 - loss 0.12458178 - time (sec): 88.17 - samples/sec: 456.46 - lr: 0.000091 - momentum: 0.000000 2023-10-11 03:44:24,115 epoch 5 - iter 290/292 - loss 0.12278117 - time (sec): 97.36 - samples/sec: 455.52 - lr: 0.000089 - momentum: 0.000000 2023-10-11 03:44:24,506 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:44:24,506 EPOCH 5 done: loss 0.1229 - lr: 0.000089 2023-10-11 03:44:30,384 DEV : loss 0.12754610180854797 - f1-score (micro avg) 0.7325 2023-10-11 03:44:30,395 saving best model 2023-10-11 03:44:33,053 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:44:41,909 epoch 6 - iter 29/292 - loss 0.08298524 - time (sec): 8.85 - samples/sec: 421.75 - lr: 0.000087 - momentum: 0.000000 2023-10-11 03:44:51,064 epoch 6 - iter 58/292 - loss 0.11175773 - time (sec): 18.01 - samples/sec: 432.68 - lr: 0.000085 - momentum: 0.000000 2023-10-11 03:45:00,824 epoch 6 - iter 87/292 - loss 0.10165883 - time (sec): 27.77 - samples/sec: 447.56 - lr: 0.000084 - momentum: 0.000000 2023-10-11 03:45:10,403 epoch 6 - iter 116/292 - loss 0.09678991 - time (sec): 37.35 - samples/sec: 445.89 - lr: 0.000082 - momentum: 0.000000 2023-10-11 03:45:20,476 epoch 6 - iter 145/292 - loss 0.09397512 - time (sec): 47.42 - samples/sec: 455.48 - lr: 0.000080 - momentum: 0.000000 2023-10-11 03:45:30,725 epoch 6 - iter 174/292 - loss 0.08697011 - time (sec): 57.67 - samples/sec: 461.02 - lr: 0.000078 - momentum: 0.000000 2023-10-11 03:45:39,853 epoch 6 - iter 203/292 - loss 0.08810664 - time (sec): 66.80 - samples/sec: 455.49 - lr: 0.000077 - momentum: 0.000000 2023-10-11 03:45:48,933 epoch 6 - iter 232/292 - loss 0.08610263 - time (sec): 75.88 - samples/sec: 448.75 - lr: 0.000075 - momentum: 0.000000 2023-10-11 03:46:00,055 epoch 6 - iter 261/292 - loss 0.08625545 - time (sec): 87.00 - samples/sec: 453.30 - lr: 0.000073 - momentum: 0.000000 2023-10-11 03:46:09,886 epoch 6 - iter 290/292 - loss 0.08521640 - time (sec): 96.83 - samples/sec: 455.25 - lr: 0.000071 - momentum: 0.000000 2023-10-11 03:46:10,525 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:46:10,525 EPOCH 6 done: loss 0.0846 - lr: 0.000071 2023-10-11 03:46:16,096 DEV : loss 0.13130022585391998 - f1-score (micro avg) 0.7613 2023-10-11 03:46:16,105 saving best model 2023-10-11 03:46:18,785 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:46:27,822 epoch 7 - iter 29/292 - loss 0.07101005 - time (sec): 9.03 - samples/sec: 422.44 - lr: 0.000069 - momentum: 0.000000 2023-10-11 03:46:36,891 epoch 7 - iter 58/292 - loss 0.06386797 - time (sec): 18.10 - samples/sec: 437.68 - lr: 0.000068 - momentum: 0.000000 2023-10-11 03:46:45,457 epoch 7 - iter 87/292 - loss 0.06618248 - time (sec): 26.67 - samples/sec: 427.44 - lr: 0.000066 - momentum: 0.000000 2023-10-11 03:46:56,357 epoch 7 - iter 116/292 - loss 0.06714686 - time (sec): 37.57 - samples/sec: 453.34 - lr: 0.000064 - momentum: 0.000000 2023-10-11 03:47:06,080 epoch 7 - iter 145/292 - loss 0.06385049 - time (sec): 47.29 - samples/sec: 446.11 - lr: 0.000062 - momentum: 0.000000 2023-10-11 03:47:15,959 epoch 7 - iter 174/292 - loss 0.06762135 - time (sec): 57.17 - samples/sec: 442.09 - lr: 0.000061 - momentum: 0.000000 2023-10-11 03:47:25,666 epoch 7 - iter 203/292 - loss 0.06778033 - time (sec): 66.88 - samples/sec: 443.83 - lr: 0.000059 - momentum: 0.000000 2023-10-11 03:47:36,174 epoch 7 - iter 232/292 - loss 0.06382795 - time (sec): 77.38 - samples/sec: 445.91 - lr: 0.000057 - momentum: 0.000000 2023-10-11 03:47:46,819 epoch 7 - iter 261/292 - loss 0.06495314 - time (sec): 88.03 - samples/sec: 446.36 - lr: 0.000055 - momentum: 0.000000 2023-10-11 03:47:57,307 epoch 7 - iter 290/292 - loss 0.06399569 - time (sec): 98.52 - samples/sec: 449.10 - lr: 0.000054 - momentum: 0.000000 2023-10-11 03:47:57,797 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:47:57,798 EPOCH 7 done: loss 0.0640 - lr: 0.000054 2023-10-11 03:48:03,890 DEV : loss 0.11788583546876907 - f1-score (micro avg) 0.7922 2023-10-11 03:48:03,899 saving best model 2023-10-11 03:48:06,447 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:48:15,755 epoch 8 - iter 29/292 - loss 0.06045212 - time (sec): 9.30 - samples/sec: 440.49 - lr: 0.000052 - momentum: 0.000000 2023-10-11 03:48:25,290 epoch 8 - iter 58/292 - loss 0.05668848 - time (sec): 18.84 - samples/sec: 448.24 - lr: 0.000050 - momentum: 0.000000 2023-10-11 03:48:34,932 epoch 8 - iter 87/292 - loss 0.06064471 - time (sec): 28.48 - samples/sec: 448.98 - lr: 0.000048 - momentum: 0.000000 2023-10-11 03:48:45,261 epoch 8 - iter 116/292 - loss 0.06304311 - time (sec): 38.81 - samples/sec: 458.96 - lr: 0.000046 - momentum: 0.000000 2023-10-11 03:48:54,956 epoch 8 - iter 145/292 - loss 0.06183104 - time (sec): 48.50 - samples/sec: 456.93 - lr: 0.000045 - momentum: 0.000000 2023-10-11 03:49:03,980 epoch 8 - iter 174/292 - loss 0.05807061 - time (sec): 57.53 - samples/sec: 450.09 - lr: 0.000043 - momentum: 0.000000 2023-10-11 03:49:14,058 epoch 8 - iter 203/292 - loss 0.05673632 - time (sec): 67.61 - samples/sec: 454.78 - lr: 0.000041 - momentum: 0.000000 2023-10-11 03:49:24,114 epoch 8 - iter 232/292 - loss 0.05801900 - time (sec): 77.66 - samples/sec: 454.79 - lr: 0.000039 - momentum: 0.000000 2023-10-11 03:49:33,999 epoch 8 - iter 261/292 - loss 0.05350185 - time (sec): 87.55 - samples/sec: 447.75 - lr: 0.000038 - momentum: 0.000000 2023-10-11 03:49:44,408 epoch 8 - iter 290/292 - loss 0.05081514 - time (sec): 97.96 - samples/sec: 451.22 - lr: 0.000036 - momentum: 0.000000 2023-10-11 03:49:44,954 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:49:44,955 EPOCH 8 done: loss 0.0507 - lr: 0.000036 2023-10-11 03:49:50,893 DEV : loss 0.1216011717915535 - f1-score (micro avg) 0.7983 2023-10-11 03:49:50,903 saving best model 2023-10-11 03:49:53,505 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:50:03,543 epoch 9 - iter 29/292 - loss 0.03943985 - time (sec): 10.03 - samples/sec: 458.19 - lr: 0.000034 - momentum: 0.000000 2023-10-11 03:50:12,469 epoch 9 - iter 58/292 - loss 0.04844555 - time (sec): 18.96 - samples/sec: 434.67 - lr: 0.000032 - momentum: 0.000000 2023-10-11 03:50:21,870 epoch 9 - iter 87/292 - loss 0.04471746 - time (sec): 28.36 - samples/sec: 437.51 - lr: 0.000030 - momentum: 0.000000 2023-10-11 03:50:31,201 epoch 9 - iter 116/292 - loss 0.04485485 - time (sec): 37.69 - samples/sec: 436.29 - lr: 0.000029 - momentum: 0.000000 2023-10-11 03:50:41,656 epoch 9 - iter 145/292 - loss 0.04694636 - time (sec): 48.15 - samples/sec: 446.77 - lr: 0.000027 - momentum: 0.000000 2023-10-11 03:50:51,729 epoch 9 - iter 174/292 - loss 0.04399638 - time (sec): 58.22 - samples/sec: 449.01 - lr: 0.000025 - momentum: 0.000000 2023-10-11 03:51:02,209 epoch 9 - iter 203/292 - loss 0.04168589 - time (sec): 68.70 - samples/sec: 447.28 - lr: 0.000023 - momentum: 0.000000 2023-10-11 03:51:12,835 epoch 9 - iter 232/292 - loss 0.04315952 - time (sec): 79.33 - samples/sec: 451.79 - lr: 0.000022 - momentum: 0.000000 2023-10-11 03:51:22,130 epoch 9 - iter 261/292 - loss 0.04201939 - time (sec): 88.62 - samples/sec: 451.08 - lr: 0.000020 - momentum: 0.000000 2023-10-11 03:51:31,943 epoch 9 - iter 290/292 - loss 0.04206582 - time (sec): 98.43 - samples/sec: 449.63 - lr: 0.000018 - momentum: 0.000000 2023-10-11 03:51:32,417 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:51:32,417 EPOCH 9 done: loss 0.0420 - lr: 0.000018 2023-10-11 03:51:38,147 DEV : loss 0.12305182963609695 - f1-score (micro avg) 0.794 2023-10-11 03:51:38,156 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:51:47,466 epoch 10 - iter 29/292 - loss 0.03708349 - time (sec): 9.31 - samples/sec: 452.73 - lr: 0.000016 - momentum: 0.000000 2023-10-11 03:51:56,648 epoch 10 - iter 58/292 - loss 0.04525890 - time (sec): 18.49 - samples/sec: 451.53 - lr: 0.000014 - momentum: 0.000000 2023-10-11 03:52:06,950 epoch 10 - iter 87/292 - loss 0.03777979 - time (sec): 28.79 - samples/sec: 458.84 - lr: 0.000013 - momentum: 0.000000 2023-10-11 03:52:16,910 epoch 10 - iter 116/292 - loss 0.03392100 - time (sec): 38.75 - samples/sec: 455.20 - lr: 0.000011 - momentum: 0.000000 2023-10-11 03:52:26,359 epoch 10 - iter 145/292 - loss 0.03634942 - time (sec): 48.20 - samples/sec: 457.11 - lr: 0.000009 - momentum: 0.000000 2023-10-11 03:52:35,578 epoch 10 - iter 174/292 - loss 0.03671887 - time (sec): 57.42 - samples/sec: 449.89 - lr: 0.000007 - momentum: 0.000000 2023-10-11 03:52:45,924 epoch 10 - iter 203/292 - loss 0.03799343 - time (sec): 67.77 - samples/sec: 457.28 - lr: 0.000006 - momentum: 0.000000 2023-10-11 03:52:55,496 epoch 10 - iter 232/292 - loss 0.03741065 - time (sec): 77.34 - samples/sec: 455.74 - lr: 0.000004 - momentum: 0.000000 2023-10-11 03:53:05,434 epoch 10 - iter 261/292 - loss 0.03813911 - time (sec): 87.28 - samples/sec: 456.28 - lr: 0.000002 - momentum: 0.000000 2023-10-11 03:53:15,261 epoch 10 - iter 290/292 - loss 0.03854453 - time (sec): 97.10 - samples/sec: 453.32 - lr: 0.000000 - momentum: 0.000000 2023-10-11 03:53:15,951 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:53:15,951 EPOCH 10 done: loss 0.0383 - lr: 0.000000 2023-10-11 03:53:21,696 DEV : loss 0.12359649688005447 - f1-score (micro avg) 0.7863 2023-10-11 03:53:22,642 ---------------------------------------------------------------------------------------------------- 2023-10-11 03:53:22,644 Loading model from best epoch ... 2023-10-11 03:53:27,154 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 03:53:40,165 Results: - F-score (micro) 0.7459 - F-score (macro) 0.7006 - Accuracy 0.6127 By class: precision recall f1-score support PER 0.8209 0.8563 0.8383 348 LOC 0.5972 0.8123 0.6883 261 ORG 0.4000 0.3846 0.3922 52 HumanProd 0.9048 0.8636 0.8837 22 micro avg 0.6958 0.8038 0.7459 683 macro avg 0.6807 0.7292 0.7006 683 weighted avg 0.7061 0.8038 0.7485 683 2023-10-11 03:53:40,165 ----------------------------------------------------------------------------------------------------