2023-10-13 12:01:40,581 ---------------------------------------------------------------------------------------------------- 2023-10-13 12:01:40,583 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=13, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-13 12:01:40,583 ---------------------------------------------------------------------------------------------------- 2023-10-13 12:01:40,583 MultiCorpus: 7936 train + 992 dev + 992 test sentences - NER_ICDAR_EUROPEANA Corpus: 7936 train + 992 dev + 992 test sentences - /root/.flair/datasets/ner_icdar_europeana/fr 2023-10-13 12:01:40,583 ---------------------------------------------------------------------------------------------------- 2023-10-13 12:01:40,583 Train: 7936 sentences 2023-10-13 12:01:40,584 (train_with_dev=False, train_with_test=False) 2023-10-13 12:01:40,584 ---------------------------------------------------------------------------------------------------- 2023-10-13 12:01:40,584 Training Params: 2023-10-13 12:01:40,584 - learning_rate: "0.00015" 2023-10-13 12:01:40,584 - mini_batch_size: "4" 2023-10-13 12:01:40,584 - max_epochs: "10" 2023-10-13 12:01:40,584 - shuffle: "True" 2023-10-13 12:01:40,584 ---------------------------------------------------------------------------------------------------- 2023-10-13 12:01:40,584 Plugins: 2023-10-13 12:01:40,584 - TensorboardLogger 2023-10-13 12:01:40,584 - LinearScheduler | warmup_fraction: '0.1' 2023-10-13 12:01:40,584 ---------------------------------------------------------------------------------------------------- 2023-10-13 12:01:40,584 Final evaluation on model from best epoch (best-model.pt) 2023-10-13 12:01:40,584 - metric: "('micro avg', 'f1-score')" 2023-10-13 12:01:40,584 ---------------------------------------------------------------------------------------------------- 2023-10-13 12:01:40,585 Computation: 2023-10-13 12:01:40,585 - compute on device: cuda:0 2023-10-13 12:01:40,585 - embedding storage: none 2023-10-13 12:01:40,585 ---------------------------------------------------------------------------------------------------- 2023-10-13 12:01:40,585 Model training base path: "hmbench-icdar/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-5" 2023-10-13 12:01:40,585 ---------------------------------------------------------------------------------------------------- 2023-10-13 12:01:40,585 ---------------------------------------------------------------------------------------------------- 2023-10-13 12:01:40,585 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-13 12:02:36,986 epoch 1 - iter 198/1984 - loss 2.53587256 - time (sec): 56.40 - samples/sec: 313.46 - lr: 0.000015 - momentum: 0.000000 2023-10-13 12:03:30,634 epoch 1 - iter 396/1984 - loss 2.35663481 - time (sec): 110.05 - samples/sec: 303.24 - lr: 0.000030 - momentum: 0.000000 2023-10-13 12:04:25,848 epoch 1 - iter 594/1984 - loss 2.03833143 - time (sec): 165.26 - samples/sec: 306.16 - lr: 0.000045 - momentum: 0.000000 2023-10-13 12:05:19,291 epoch 1 - iter 792/1984 - loss 1.75219619 - time (sec): 218.70 - samples/sec: 299.99 - lr: 0.000060 - momentum: 0.000000 2023-10-13 12:06:17,514 epoch 1 - iter 990/1984 - loss 1.50365684 - time (sec): 276.93 - samples/sec: 294.62 - lr: 0.000075 - momentum: 0.000000 2023-10-13 12:07:11,217 epoch 1 - iter 1188/1984 - loss 1.30904393 - time (sec): 330.63 - samples/sec: 295.02 - lr: 0.000090 - momentum: 0.000000 2023-10-13 12:08:05,806 epoch 1 - iter 1386/1984 - loss 1.15357563 - time (sec): 385.22 - samples/sec: 296.78 - lr: 0.000105 - momentum: 0.000000 2023-10-13 12:09:02,915 epoch 1 - iter 1584/1984 - loss 1.03648294 - time (sec): 442.33 - samples/sec: 294.58 - lr: 0.000120 - momentum: 0.000000 2023-10-13 12:10:00,529 epoch 1 - iter 1782/1984 - loss 0.93126337 - time (sec): 499.94 - samples/sec: 296.04 - lr: 0.000135 - momentum: 0.000000 2023-10-13 12:11:00,429 epoch 1 - iter 1980/1984 - loss 0.85719194 - time (sec): 559.84 - samples/sec: 292.50 - lr: 0.000150 - momentum: 0.000000 2023-10-13 12:11:01,586 ---------------------------------------------------------------------------------------------------- 2023-10-13 12:11:01,587 EPOCH 1 done: loss 0.8562 - lr: 0.000150 2023-10-13 12:11:27,268 DEV : loss 0.12921574711799622 - f1-score (micro avg) 0.6608 2023-10-13 12:11:27,311 saving best model 2023-10-13 12:11:28,286 ---------------------------------------------------------------------------------------------------- 2023-10-13 12:12:25,572 epoch 2 - iter 198/1984 - loss 0.15246598 - time (sec): 57.28 - samples/sec: 288.91 - lr: 0.000148 - momentum: 0.000000 2023-10-13 12:13:23,204 epoch 2 - iter 396/1984 - loss 0.13947723 - time (sec): 114.92 - samples/sec: 289.19 - lr: 0.000147 - momentum: 0.000000 2023-10-13 12:14:19,334 epoch 2 - iter 594/1984 - loss 0.13236118 - time (sec): 171.05 - samples/sec: 293.49 - lr: 0.000145 - momentum: 0.000000 2023-10-13 12:15:15,682 epoch 2 - iter 792/1984 - loss 0.13139937 - time (sec): 227.39 - samples/sec: 289.49 - lr: 0.000143 - momentum: 0.000000 2023-10-13 12:16:13,148 epoch 2 - iter 990/1984 - loss 0.12747303 - time (sec): 284.86 - samples/sec: 288.91 - lr: 0.000142 - momentum: 0.000000 2023-10-13 12:17:07,128 epoch 2 - iter 1188/1984 - loss 0.12581150 - time (sec): 338.84 - samples/sec: 290.91 - lr: 0.000140 - momentum: 0.000000 2023-10-13 12:18:01,605 epoch 2 - iter 1386/1984 - loss 0.12386868 - time (sec): 393.32 - samples/sec: 292.02 - lr: 0.000138 - momentum: 0.000000 2023-10-13 12:18:57,699 epoch 2 - iter 1584/1984 - loss 0.12091584 - time (sec): 449.41 - samples/sec: 291.07 - lr: 0.000137 - momentum: 0.000000 2023-10-13 12:19:53,333 epoch 2 - iter 1782/1984 - loss 0.11952771 - time (sec): 505.04 - samples/sec: 289.53 - lr: 0.000135 - momentum: 0.000000 2023-10-13 12:20:51,124 epoch 2 - iter 1980/1984 - loss 0.11686668 - time (sec): 562.84 - samples/sec: 290.88 - lr: 0.000133 - momentum: 0.000000 2023-10-13 12:20:52,222 ---------------------------------------------------------------------------------------------------- 2023-10-13 12:20:52,223 EPOCH 2 done: loss 0.1168 - lr: 0.000133 2023-10-13 12:21:22,532 DEV : loss 0.08594389259815216 - f1-score (micro avg) 0.7326 2023-10-13 12:21:22,586 saving best model 2023-10-13 12:21:25,326 ---------------------------------------------------------------------------------------------------- 2023-10-13 12:22:22,376 epoch 3 - iter 198/1984 - loss 0.06835268 - time (sec): 57.05 - samples/sec: 282.30 - lr: 0.000132 - momentum: 0.000000 2023-10-13 12:23:21,479 epoch 3 - iter 396/1984 - loss 0.07762823 - time (sec): 116.15 - samples/sec: 279.06 - lr: 0.000130 - momentum: 0.000000 2023-10-13 12:24:16,710 epoch 3 - iter 594/1984 - loss 0.07827809 - time (sec): 171.38 - samples/sec: 283.05 - lr: 0.000128 - momentum: 0.000000 2023-10-13 12:25:12,056 epoch 3 - iter 792/1984 - loss 0.07869207 - time (sec): 226.73 - samples/sec: 286.28 - lr: 0.000127 - momentum: 0.000000 2023-10-13 12:26:07,407 epoch 3 - iter 990/1984 - loss 0.07826522 - time (sec): 282.08 - samples/sec: 287.01 - lr: 0.000125 - momentum: 0.000000 2023-10-13 12:27:04,007 epoch 3 - iter 1188/1984 - loss 0.07782816 - time (sec): 338.68 - samples/sec: 288.60 - lr: 0.000123 - momentum: 0.000000 2023-10-13 12:28:00,792 epoch 3 - iter 1386/1984 - loss 0.07814075 - time (sec): 395.46 - samples/sec: 288.89 - lr: 0.000122 - momentum: 0.000000 2023-10-13 12:28:58,604 epoch 3 - iter 1584/1984 - loss 0.07681156 - time (sec): 453.27 - samples/sec: 288.09 - lr: 0.000120 - momentum: 0.000000 2023-10-13 12:29:54,823 epoch 3 - iter 1782/1984 - loss 0.07566942 - time (sec): 509.49 - samples/sec: 288.85 - lr: 0.000118 - momentum: 0.000000 2023-10-13 12:30:54,387 epoch 3 - iter 1980/1984 - loss 0.07589565 - time (sec): 569.06 - samples/sec: 287.58 - lr: 0.000117 - momentum: 0.000000 2023-10-13 12:30:55,447 ---------------------------------------------------------------------------------------------------- 2023-10-13 12:30:55,448 EPOCH 3 done: loss 0.0758 - lr: 0.000117 2023-10-13 12:31:22,850 DEV : loss 0.09770967811346054 - f1-score (micro avg) 0.7546 2023-10-13 12:31:22,896 saving best model 2023-10-13 12:31:25,680 ---------------------------------------------------------------------------------------------------- 2023-10-13 12:32:22,939 epoch 4 - iter 198/1984 - loss 0.05842859 - time (sec): 57.25 - samples/sec: 287.82 - lr: 0.000115 - momentum: 0.000000 2023-10-13 12:33:20,230 epoch 4 - iter 396/1984 - loss 0.05467250 - time (sec): 114.55 - samples/sec: 285.40 - lr: 0.000113 - momentum: 0.000000 2023-10-13 12:34:20,580 epoch 4 - iter 594/1984 - loss 0.05230249 - time (sec): 174.90 - samples/sec: 279.68 - lr: 0.000112 - momentum: 0.000000 2023-10-13 12:35:21,055 epoch 4 - iter 792/1984 - loss 0.05292552 - time (sec): 235.37 - samples/sec: 277.36 - lr: 0.000110 - momentum: 0.000000 2023-10-13 12:36:20,546 epoch 4 - iter 990/1984 - loss 0.05358516 - time (sec): 294.86 - samples/sec: 279.46 - lr: 0.000108 - momentum: 0.000000 2023-10-13 12:37:21,108 epoch 4 - iter 1188/1984 - loss 0.05228562 - time (sec): 355.42 - samples/sec: 279.04 - lr: 0.000107 - momentum: 0.000000 2023-10-13 12:38:21,766 epoch 4 - iter 1386/1984 - loss 0.05219360 - time (sec): 416.08 - samples/sec: 276.24 - lr: 0.000105 - momentum: 0.000000 2023-10-13 12:39:19,041 epoch 4 - iter 1584/1984 - loss 0.05282172 - time (sec): 473.36 - samples/sec: 277.03 - lr: 0.000103 - momentum: 0.000000 2023-10-13 12:40:16,271 epoch 4 - iter 1782/1984 - loss 0.05363652 - time (sec): 530.59 - samples/sec: 277.86 - lr: 0.000102 - momentum: 0.000000 2023-10-13 12:41:14,283 epoch 4 - iter 1980/1984 - loss 0.05409466 - time (sec): 588.60 - samples/sec: 278.11 - lr: 0.000100 - momentum: 0.000000 2023-10-13 12:41:15,420 ---------------------------------------------------------------------------------------------------- 2023-10-13 12:41:15,421 EPOCH 4 done: loss 0.0540 - lr: 0.000100 2023-10-13 12:41:43,934 DEV : loss 0.13024039566516876 - f1-score (micro avg) 0.7696 2023-10-13 12:41:43,978 saving best model 2023-10-13 12:41:49,138 ---------------------------------------------------------------------------------------------------- 2023-10-13 12:42:46,154 epoch 5 - iter 198/1984 - loss 0.03679399 - time (sec): 57.01 - samples/sec: 294.51 - lr: 0.000098 - momentum: 0.000000 2023-10-13 12:43:43,295 epoch 5 - iter 396/1984 - loss 0.03905502 - time (sec): 114.15 - samples/sec: 292.02 - lr: 0.000097 - momentum: 0.000000 2023-10-13 12:44:40,686 epoch 5 - iter 594/1984 - loss 0.04296223 - time (sec): 171.54 - samples/sec: 291.74 - lr: 0.000095 - momentum: 0.000000 2023-10-13 12:45:35,672 epoch 5 - iter 792/1984 - loss 0.04216427 - time (sec): 226.53 - samples/sec: 293.31 - lr: 0.000093 - momentum: 0.000000 2023-10-13 12:46:31,117 epoch 5 - iter 990/1984 - loss 0.04149098 - time (sec): 281.97 - samples/sec: 292.56 - lr: 0.000092 - momentum: 0.000000 2023-10-13 12:47:30,500 epoch 5 - iter 1188/1984 - loss 0.04224225 - time (sec): 341.36 - samples/sec: 287.90 - lr: 0.000090 - momentum: 0.000000 2023-10-13 12:48:25,231 epoch 5 - iter 1386/1984 - loss 0.04138722 - time (sec): 396.09 - samples/sec: 288.37 - lr: 0.000088 - momentum: 0.000000 2023-10-13 12:49:21,590 epoch 5 - iter 1584/1984 - loss 0.04151736 - time (sec): 452.45 - samples/sec: 290.10 - lr: 0.000087 - momentum: 0.000000 2023-10-13 12:50:20,385 epoch 5 - iter 1782/1984 - loss 0.04213374 - time (sec): 511.24 - samples/sec: 288.80 - lr: 0.000085 - momentum: 0.000000 2023-10-13 12:51:16,349 epoch 5 - iter 1980/1984 - loss 0.04118791 - time (sec): 567.21 - samples/sec: 288.75 - lr: 0.000083 - momentum: 0.000000 2023-10-13 12:51:17,450 ---------------------------------------------------------------------------------------------------- 2023-10-13 12:51:17,451 EPOCH 5 done: loss 0.0411 - lr: 0.000083 2023-10-13 12:51:45,096 DEV : loss 0.1498626172542572 - f1-score (micro avg) 0.7658 2023-10-13 12:51:45,140 ---------------------------------------------------------------------------------------------------- 2023-10-13 12:52:42,428 epoch 6 - iter 198/1984 - loss 0.02535367 - time (sec): 57.29 - samples/sec: 298.24 - lr: 0.000082 - momentum: 0.000000 2023-10-13 12:53:43,797 epoch 6 - iter 396/1984 - loss 0.02460873 - time (sec): 118.65 - samples/sec: 282.58 - lr: 0.000080 - momentum: 0.000000 2023-10-13 12:54:39,904 epoch 6 - iter 594/1984 - loss 0.02590999 - time (sec): 174.76 - samples/sec: 282.27 - lr: 0.000078 - momentum: 0.000000 2023-10-13 12:55:35,769 epoch 6 - iter 792/1984 - loss 0.02897049 - time (sec): 230.63 - samples/sec: 285.87 - lr: 0.000077 - momentum: 0.000000 2023-10-13 12:56:29,696 epoch 6 - iter 990/1984 - loss 0.02985477 - time (sec): 284.55 - samples/sec: 289.08 - lr: 0.000075 - momentum: 0.000000 2023-10-13 12:57:23,689 epoch 6 - iter 1188/1984 - loss 0.02976773 - time (sec): 338.55 - samples/sec: 291.14 - lr: 0.000073 - momentum: 0.000000 2023-10-13 12:58:17,578 epoch 6 - iter 1386/1984 - loss 0.02995438 - time (sec): 392.44 - samples/sec: 292.99 - lr: 0.000072 - momentum: 0.000000 2023-10-13 12:59:11,837 epoch 6 - iter 1584/1984 - loss 0.03040644 - time (sec): 446.70 - samples/sec: 293.10 - lr: 0.000070 - momentum: 0.000000 2023-10-13 13:00:05,389 epoch 6 - iter 1782/1984 - loss 0.03000902 - time (sec): 500.25 - samples/sec: 294.19 - lr: 0.000068 - momentum: 0.000000 2023-10-13 13:00:58,136 epoch 6 - iter 1980/1984 - loss 0.03131165 - time (sec): 552.99 - samples/sec: 296.04 - lr: 0.000067 - momentum: 0.000000 2023-10-13 13:00:59,167 ---------------------------------------------------------------------------------------------------- 2023-10-13 13:00:59,167 EPOCH 6 done: loss 0.0314 - lr: 0.000067 2023-10-13 13:01:25,953 DEV : loss 0.1546768993139267 - f1-score (micro avg) 0.7604 2023-10-13 13:01:25,994 ---------------------------------------------------------------------------------------------------- 2023-10-13 13:02:20,487 epoch 7 - iter 198/1984 - loss 0.02164230 - time (sec): 54.49 - samples/sec: 302.56 - lr: 0.000065 - momentum: 0.000000 2023-10-13 13:03:15,597 epoch 7 - iter 396/1984 - loss 0.02187396 - time (sec): 109.60 - samples/sec: 294.20 - lr: 0.000063 - momentum: 0.000000 2023-10-13 13:04:11,813 epoch 7 - iter 594/1984 - loss 0.02064500 - time (sec): 165.82 - samples/sec: 296.65 - lr: 0.000062 - momentum: 0.000000 2023-10-13 13:05:08,654 epoch 7 - iter 792/1984 - loss 0.02239191 - time (sec): 222.66 - samples/sec: 292.47 - lr: 0.000060 - momentum: 0.000000 2023-10-13 13:06:06,207 epoch 7 - iter 990/1984 - loss 0.02117540 - time (sec): 280.21 - samples/sec: 290.85 - lr: 0.000058 - momentum: 0.000000 2023-10-13 13:07:00,612 epoch 7 - iter 1188/1984 - loss 0.02163675 - time (sec): 334.62 - samples/sec: 292.24 - lr: 0.000057 - momentum: 0.000000 2023-10-13 13:07:58,537 epoch 7 - iter 1386/1984 - loss 0.02215893 - time (sec): 392.54 - samples/sec: 290.62 - lr: 0.000055 - momentum: 0.000000 2023-10-13 13:08:52,925 epoch 7 - iter 1584/1984 - loss 0.02198376 - time (sec): 446.93 - samples/sec: 290.89 - lr: 0.000053 - momentum: 0.000000 2023-10-13 13:09:48,604 epoch 7 - iter 1782/1984 - loss 0.02278283 - time (sec): 502.61 - samples/sec: 292.61 - lr: 0.000052 - momentum: 0.000000 2023-10-13 13:10:47,760 epoch 7 - iter 1980/1984 - loss 0.02389101 - time (sec): 561.76 - samples/sec: 291.52 - lr: 0.000050 - momentum: 0.000000 2023-10-13 13:10:48,845 ---------------------------------------------------------------------------------------------------- 2023-10-13 13:10:48,845 EPOCH 7 done: loss 0.0239 - lr: 0.000050 2023-10-13 13:11:15,612 DEV : loss 0.183299720287323 - f1-score (micro avg) 0.7642 2023-10-13 13:11:15,659 ---------------------------------------------------------------------------------------------------- 2023-10-13 13:12:11,816 epoch 8 - iter 198/1984 - loss 0.01535571 - time (sec): 56.16 - samples/sec: 293.44 - lr: 0.000048 - momentum: 0.000000 2023-10-13 13:13:07,794 epoch 8 - iter 396/1984 - loss 0.01489584 - time (sec): 112.13 - samples/sec: 296.02 - lr: 0.000047 - momentum: 0.000000 2023-10-13 13:14:01,252 epoch 8 - iter 594/1984 - loss 0.01439117 - time (sec): 165.59 - samples/sec: 297.78 - lr: 0.000045 - momentum: 0.000000 2023-10-13 13:14:54,994 epoch 8 - iter 792/1984 - loss 0.01550966 - time (sec): 219.33 - samples/sec: 300.88 - lr: 0.000043 - momentum: 0.000000 2023-10-13 13:15:53,380 epoch 8 - iter 990/1984 - loss 0.01519291 - time (sec): 277.72 - samples/sec: 296.15 - lr: 0.000042 - momentum: 0.000000 2023-10-13 13:16:45,061 epoch 8 - iter 1188/1984 - loss 0.01508252 - time (sec): 329.40 - samples/sec: 299.29 - lr: 0.000040 - momentum: 0.000000 2023-10-13 13:17:41,489 epoch 8 - iter 1386/1984 - loss 0.01547066 - time (sec): 385.83 - samples/sec: 297.82 - lr: 0.000038 - momentum: 0.000000 2023-10-13 13:18:34,335 epoch 8 - iter 1584/1984 - loss 0.01553468 - time (sec): 438.67 - samples/sec: 297.60 - lr: 0.000037 - momentum: 0.000000 2023-10-13 13:19:27,559 epoch 8 - iter 1782/1984 - loss 0.01564675 - time (sec): 491.90 - samples/sec: 298.83 - lr: 0.000035 - momentum: 0.000000 2023-10-13 13:20:20,410 epoch 8 - iter 1980/1984 - loss 0.01495060 - time (sec): 544.75 - samples/sec: 300.60 - lr: 0.000033 - momentum: 0.000000 2023-10-13 13:20:21,482 ---------------------------------------------------------------------------------------------------- 2023-10-13 13:20:21,483 EPOCH 8 done: loss 0.0151 - lr: 0.000033 2023-10-13 13:20:48,505 DEV : loss 0.20292401313781738 - f1-score (micro avg) 0.7682 2023-10-13 13:20:48,555 ---------------------------------------------------------------------------------------------------- 2023-10-13 13:21:39,886 epoch 9 - iter 198/1984 - loss 0.01309539 - time (sec): 51.33 - samples/sec: 307.26 - lr: 0.000032 - momentum: 0.000000 2023-10-13 13:22:31,621 epoch 9 - iter 396/1984 - loss 0.01271783 - time (sec): 103.06 - samples/sec: 308.62 - lr: 0.000030 - momentum: 0.000000 2023-10-13 13:23:25,218 epoch 9 - iter 594/1984 - loss 0.01154190 - time (sec): 156.66 - samples/sec: 308.95 - lr: 0.000028 - momentum: 0.000000 2023-10-13 13:24:19,703 epoch 9 - iter 792/1984 - loss 0.01142730 - time (sec): 211.15 - samples/sec: 308.12 - lr: 0.000027 - momentum: 0.000000 2023-10-13 13:25:13,668 epoch 9 - iter 990/1984 - loss 0.01281801 - time (sec): 265.11 - samples/sec: 306.46 - lr: 0.000025 - momentum: 0.000000 2023-10-13 13:26:07,047 epoch 9 - iter 1188/1984 - loss 0.01251343 - time (sec): 318.49 - samples/sec: 301.02 - lr: 0.000023 - momentum: 0.000000 2023-10-13 13:27:02,940 epoch 9 - iter 1386/1984 - loss 0.01204291 - time (sec): 374.38 - samples/sec: 302.67 - lr: 0.000022 - momentum: 0.000000 2023-10-13 13:27:56,110 epoch 9 - iter 1584/1984 - loss 0.01200449 - time (sec): 427.55 - samples/sec: 303.59 - lr: 0.000020 - momentum: 0.000000 2023-10-13 13:28:51,579 epoch 9 - iter 1782/1984 - loss 0.01219217 - time (sec): 483.02 - samples/sec: 304.57 - lr: 0.000018 - momentum: 0.000000 2023-10-13 13:29:45,819 epoch 9 - iter 1980/1984 - loss 0.01242827 - time (sec): 537.26 - samples/sec: 304.48 - lr: 0.000017 - momentum: 0.000000 2023-10-13 13:29:47,047 ---------------------------------------------------------------------------------------------------- 2023-10-13 13:29:47,048 EPOCH 9 done: loss 0.0124 - lr: 0.000017 2023-10-13 13:30:13,711 DEV : loss 0.21697697043418884 - f1-score (micro avg) 0.7617 2023-10-13 13:30:13,764 ---------------------------------------------------------------------------------------------------- 2023-10-13 13:31:08,460 epoch 10 - iter 198/1984 - loss 0.00658872 - time (sec): 54.69 - samples/sec: 309.49 - lr: 0.000015 - momentum: 0.000000 2023-10-13 13:32:01,825 epoch 10 - iter 396/1984 - loss 0.00947681 - time (sec): 108.06 - samples/sec: 301.78 - lr: 0.000013 - momentum: 0.000000 2023-10-13 13:32:55,149 epoch 10 - iter 594/1984 - loss 0.00879204 - time (sec): 161.38 - samples/sec: 300.65 - lr: 0.000012 - momentum: 0.000000 2023-10-13 13:33:51,084 epoch 10 - iter 792/1984 - loss 0.00808015 - time (sec): 217.32 - samples/sec: 296.34 - lr: 0.000010 - momentum: 0.000000 2023-10-13 13:34:48,261 epoch 10 - iter 990/1984 - loss 0.00746892 - time (sec): 274.49 - samples/sec: 295.76 - lr: 0.000008 - momentum: 0.000000 2023-10-13 13:35:43,142 epoch 10 - iter 1188/1984 - loss 0.00746131 - time (sec): 329.38 - samples/sec: 297.49 - lr: 0.000007 - momentum: 0.000000 2023-10-13 13:36:37,996 epoch 10 - iter 1386/1984 - loss 0.00742811 - time (sec): 384.23 - samples/sec: 298.45 - lr: 0.000005 - momentum: 0.000000 2023-10-13 13:37:32,826 epoch 10 - iter 1584/1984 - loss 0.00777138 - time (sec): 439.06 - samples/sec: 299.37 - lr: 0.000003 - momentum: 0.000000 2023-10-13 13:38:29,578 epoch 10 - iter 1782/1984 - loss 0.00774151 - time (sec): 495.81 - samples/sec: 298.21 - lr: 0.000002 - momentum: 0.000000 2023-10-13 13:39:24,305 epoch 10 - iter 1980/1984 - loss 0.00806446 - time (sec): 550.54 - samples/sec: 297.17 - lr: 0.000000 - momentum: 0.000000 2023-10-13 13:39:25,573 ---------------------------------------------------------------------------------------------------- 2023-10-13 13:39:25,574 EPOCH 10 done: loss 0.0080 - lr: 0.000000 2023-10-13 13:39:51,477 DEV : loss 0.22803443670272827 - f1-score (micro avg) 0.7611 2023-10-13 13:39:52,471 ---------------------------------------------------------------------------------------------------- 2023-10-13 13:39:52,473 Loading model from best epoch ... 2023-10-13 13:39:57,370 SequenceTagger predicts: Dictionary with 13 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 2023-10-13 13:40:22,958 Results: - F-score (micro) 0.7832 - F-score (macro) 0.6902 - Accuracy 0.6628 By class: precision recall f1-score support LOC 0.8351 0.8656 0.8501 655 PER 0.7284 0.7937 0.7597 223 ORG 0.4828 0.4409 0.4609 127 micro avg 0.7707 0.7960 0.7832 1005 macro avg 0.6821 0.7001 0.6902 1005 weighted avg 0.7669 0.7960 0.7808 1005 2023-10-13 13:40:22,958 ----------------------------------------------------------------------------------------------------