File size: 23,786 Bytes
780b0f3 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 |
2023-10-17 14:10:33,346 ----------------------------------------------------------------------------------------------------
2023-10-17 14:10:33,347 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): ElectraModel(
(embeddings): ElectraEmbeddings(
(word_embeddings): Embedding(32001, 768)
(position_embeddings): Embedding(512, 768)
(token_type_embeddings): Embedding(2, 768)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): ElectraEncoder(
(layer): ModuleList(
(0-11): 12 x ElectraLayer(
(attention): ElectraAttention(
(self): ElectraSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): ElectraSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): ElectraIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): ElectraOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=13, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 14:10:33,347 ----------------------------------------------------------------------------------------------------
2023-10-17 14:10:33,347 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
2023-10-17 14:10:33,347 ----------------------------------------------------------------------------------------------------
2023-10-17 14:10:33,347 Train: 7936 sentences
2023-10-17 14:10:33,347 (train_with_dev=False, train_with_test=False)
2023-10-17 14:10:33,347 ----------------------------------------------------------------------------------------------------
2023-10-17 14:10:33,347 Training Params:
2023-10-17 14:10:33,347 - learning_rate: "5e-05"
2023-10-17 14:10:33,347 - mini_batch_size: "8"
2023-10-17 14:10:33,347 - max_epochs: "10"
2023-10-17 14:10:33,348 - shuffle: "True"
2023-10-17 14:10:33,348 ----------------------------------------------------------------------------------------------------
2023-10-17 14:10:33,348 Plugins:
2023-10-17 14:10:33,348 - TensorboardLogger
2023-10-17 14:10:33,348 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 14:10:33,348 ----------------------------------------------------------------------------------------------------
2023-10-17 14:10:33,348 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 14:10:33,348 - metric: "('micro avg', 'f1-score')"
2023-10-17 14:10:33,348 ----------------------------------------------------------------------------------------------------
2023-10-17 14:10:33,348 Computation:
2023-10-17 14:10:33,348 - compute on device: cuda:0
2023-10-17 14:10:33,348 - embedding storage: none
2023-10-17 14:10:33,348 ----------------------------------------------------------------------------------------------------
2023-10-17 14:10:33,348 Model training base path: "hmbench-icdar/fr-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-17 14:10:33,348 ----------------------------------------------------------------------------------------------------
2023-10-17 14:10:33,348 ----------------------------------------------------------------------------------------------------
2023-10-17 14:10:33,348 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 14:10:39,222 epoch 1 - iter 99/992 - loss 2.56583698 - time (sec): 5.87 - samples/sec: 2821.57 - lr: 0.000005 - momentum: 0.000000
2023-10-17 14:10:45,260 epoch 1 - iter 198/992 - loss 1.44067154 - time (sec): 11.91 - samples/sec: 2815.77 - lr: 0.000010 - momentum: 0.000000
2023-10-17 14:10:50,904 epoch 1 - iter 297/992 - loss 1.06311481 - time (sec): 17.56 - samples/sec: 2793.50 - lr: 0.000015 - momentum: 0.000000
2023-10-17 14:10:56,606 epoch 1 - iter 396/992 - loss 0.85235507 - time (sec): 23.26 - samples/sec: 2791.92 - lr: 0.000020 - momentum: 0.000000
2023-10-17 14:11:02,780 epoch 1 - iter 495/992 - loss 0.70397019 - time (sec): 29.43 - samples/sec: 2809.19 - lr: 0.000025 - momentum: 0.000000
2023-10-17 14:11:08,842 epoch 1 - iter 594/992 - loss 0.61754102 - time (sec): 35.49 - samples/sec: 2781.20 - lr: 0.000030 - momentum: 0.000000
2023-10-17 14:11:14,877 epoch 1 - iter 693/992 - loss 0.55425408 - time (sec): 41.53 - samples/sec: 2767.88 - lr: 0.000035 - momentum: 0.000000
2023-10-17 14:11:21,011 epoch 1 - iter 792/992 - loss 0.50304327 - time (sec): 47.66 - samples/sec: 2761.17 - lr: 0.000040 - momentum: 0.000000
2023-10-17 14:11:26,898 epoch 1 - iter 891/992 - loss 0.46403461 - time (sec): 53.55 - samples/sec: 2766.86 - lr: 0.000045 - momentum: 0.000000
2023-10-17 14:11:33,013 epoch 1 - iter 990/992 - loss 0.43242766 - time (sec): 59.66 - samples/sec: 2743.63 - lr: 0.000050 - momentum: 0.000000
2023-10-17 14:11:33,117 ----------------------------------------------------------------------------------------------------
2023-10-17 14:11:33,117 EPOCH 1 done: loss 0.4321 - lr: 0.000050
2023-10-17 14:11:36,203 DEV : loss 0.08887125551700592 - f1-score (micro avg) 0.7233
2023-10-17 14:11:36,224 saving best model
2023-10-17 14:11:36,696 ----------------------------------------------------------------------------------------------------
2023-10-17 14:11:42,469 epoch 2 - iter 99/992 - loss 0.11598058 - time (sec): 5.77 - samples/sec: 2830.44 - lr: 0.000049 - momentum: 0.000000
2023-10-17 14:11:48,383 epoch 2 - iter 198/992 - loss 0.11266582 - time (sec): 11.69 - samples/sec: 2820.53 - lr: 0.000049 - momentum: 0.000000
2023-10-17 14:11:54,000 epoch 2 - iter 297/992 - loss 0.11132229 - time (sec): 17.30 - samples/sec: 2803.21 - lr: 0.000048 - momentum: 0.000000
2023-10-17 14:11:59,687 epoch 2 - iter 396/992 - loss 0.11223753 - time (sec): 22.99 - samples/sec: 2812.50 - lr: 0.000048 - momentum: 0.000000
2023-10-17 14:12:05,527 epoch 2 - iter 495/992 - loss 0.10847458 - time (sec): 28.83 - samples/sec: 2810.26 - lr: 0.000047 - momentum: 0.000000
2023-10-17 14:12:11,595 epoch 2 - iter 594/992 - loss 0.10790183 - time (sec): 34.90 - samples/sec: 2794.95 - lr: 0.000047 - momentum: 0.000000
2023-10-17 14:12:17,710 epoch 2 - iter 693/992 - loss 0.10908683 - time (sec): 41.01 - samples/sec: 2796.29 - lr: 0.000046 - momentum: 0.000000
2023-10-17 14:12:23,919 epoch 2 - iter 792/992 - loss 0.10920125 - time (sec): 47.22 - samples/sec: 2773.55 - lr: 0.000046 - momentum: 0.000000
2023-10-17 14:12:29,935 epoch 2 - iter 891/992 - loss 0.10838687 - time (sec): 53.24 - samples/sec: 2760.47 - lr: 0.000045 - momentum: 0.000000
2023-10-17 14:12:35,919 epoch 2 - iter 990/992 - loss 0.10734045 - time (sec): 59.22 - samples/sec: 2766.07 - lr: 0.000044 - momentum: 0.000000
2023-10-17 14:12:36,025 ----------------------------------------------------------------------------------------------------
2023-10-17 14:12:36,025 EPOCH 2 done: loss 0.1074 - lr: 0.000044
2023-10-17 14:12:39,465 DEV : loss 0.08363784104585648 - f1-score (micro avg) 0.7596
2023-10-17 14:12:39,487 saving best model
2023-10-17 14:12:40,074 ----------------------------------------------------------------------------------------------------
2023-10-17 14:12:46,174 epoch 3 - iter 99/992 - loss 0.07378378 - time (sec): 6.10 - samples/sec: 2782.50 - lr: 0.000044 - momentum: 0.000000
2023-10-17 14:12:52,120 epoch 3 - iter 198/992 - loss 0.07492355 - time (sec): 12.04 - samples/sec: 2770.67 - lr: 0.000043 - momentum: 0.000000
2023-10-17 14:12:57,854 epoch 3 - iter 297/992 - loss 0.07489005 - time (sec): 17.78 - samples/sec: 2780.55 - lr: 0.000043 - momentum: 0.000000
2023-10-17 14:13:03,942 epoch 3 - iter 396/992 - loss 0.07422590 - time (sec): 23.87 - samples/sec: 2761.11 - lr: 0.000042 - momentum: 0.000000
2023-10-17 14:13:09,975 epoch 3 - iter 495/992 - loss 0.07723264 - time (sec): 29.90 - samples/sec: 2764.56 - lr: 0.000042 - momentum: 0.000000
2023-10-17 14:13:16,155 epoch 3 - iter 594/992 - loss 0.07724225 - time (sec): 36.08 - samples/sec: 2761.92 - lr: 0.000041 - momentum: 0.000000
2023-10-17 14:13:22,176 epoch 3 - iter 693/992 - loss 0.07604513 - time (sec): 42.10 - samples/sec: 2774.23 - lr: 0.000041 - momentum: 0.000000
2023-10-17 14:13:28,294 epoch 3 - iter 792/992 - loss 0.07536649 - time (sec): 48.22 - samples/sec: 2761.96 - lr: 0.000040 - momentum: 0.000000
2023-10-17 14:13:34,071 epoch 3 - iter 891/992 - loss 0.07551830 - time (sec): 53.99 - samples/sec: 2754.35 - lr: 0.000039 - momentum: 0.000000
2023-10-17 14:13:39,833 epoch 3 - iter 990/992 - loss 0.07517252 - time (sec): 59.76 - samples/sec: 2739.02 - lr: 0.000039 - momentum: 0.000000
2023-10-17 14:13:39,973 ----------------------------------------------------------------------------------------------------
2023-10-17 14:13:39,974 EPOCH 3 done: loss 0.0751 - lr: 0.000039
2023-10-17 14:13:43,451 DEV : loss 0.1065104603767395 - f1-score (micro avg) 0.748
2023-10-17 14:13:43,475 ----------------------------------------------------------------------------------------------------
2023-10-17 14:13:49,475 epoch 4 - iter 99/992 - loss 0.04524979 - time (sec): 6.00 - samples/sec: 2803.07 - lr: 0.000038 - momentum: 0.000000
2023-10-17 14:13:55,680 epoch 4 - iter 198/992 - loss 0.05478362 - time (sec): 12.20 - samples/sec: 2690.30 - lr: 0.000038 - momentum: 0.000000
2023-10-17 14:14:01,531 epoch 4 - iter 297/992 - loss 0.05412289 - time (sec): 18.05 - samples/sec: 2698.58 - lr: 0.000037 - momentum: 0.000000
2023-10-17 14:14:07,545 epoch 4 - iter 396/992 - loss 0.05506235 - time (sec): 24.07 - samples/sec: 2730.99 - lr: 0.000037 - momentum: 0.000000
2023-10-17 14:14:13,343 epoch 4 - iter 495/992 - loss 0.05562356 - time (sec): 29.87 - samples/sec: 2748.67 - lr: 0.000036 - momentum: 0.000000
2023-10-17 14:14:19,477 epoch 4 - iter 594/992 - loss 0.05609226 - time (sec): 36.00 - samples/sec: 2751.59 - lr: 0.000036 - momentum: 0.000000
2023-10-17 14:14:25,172 epoch 4 - iter 693/992 - loss 0.05560238 - time (sec): 41.70 - samples/sec: 2758.09 - lr: 0.000035 - momentum: 0.000000
2023-10-17 14:14:31,622 epoch 4 - iter 792/992 - loss 0.05713362 - time (sec): 48.15 - samples/sec: 2727.19 - lr: 0.000034 - momentum: 0.000000
2023-10-17 14:14:37,562 epoch 4 - iter 891/992 - loss 0.05527867 - time (sec): 54.09 - samples/sec: 2723.46 - lr: 0.000034 - momentum: 0.000000
2023-10-17 14:14:43,601 epoch 4 - iter 990/992 - loss 0.05470377 - time (sec): 60.12 - samples/sec: 2723.71 - lr: 0.000033 - momentum: 0.000000
2023-10-17 14:14:43,697 ----------------------------------------------------------------------------------------------------
2023-10-17 14:14:43,697 EPOCH 4 done: loss 0.0547 - lr: 0.000033
2023-10-17 14:14:47,132 DEV : loss 0.13179177045822144 - f1-score (micro avg) 0.7556
2023-10-17 14:14:47,153 ----------------------------------------------------------------------------------------------------
2023-10-17 14:14:52,850 epoch 5 - iter 99/992 - loss 0.05176063 - time (sec): 5.70 - samples/sec: 2801.03 - lr: 0.000033 - momentum: 0.000000
2023-10-17 14:14:58,673 epoch 5 - iter 198/992 - loss 0.04599398 - time (sec): 11.52 - samples/sec: 2830.16 - lr: 0.000032 - momentum: 0.000000
2023-10-17 14:15:04,479 epoch 5 - iter 297/992 - loss 0.04300166 - time (sec): 17.32 - samples/sec: 2790.18 - lr: 0.000032 - momentum: 0.000000
2023-10-17 14:15:10,429 epoch 5 - iter 396/992 - loss 0.04689564 - time (sec): 23.27 - samples/sec: 2771.19 - lr: 0.000031 - momentum: 0.000000
2023-10-17 14:15:16,294 epoch 5 - iter 495/992 - loss 0.04522045 - time (sec): 29.14 - samples/sec: 2782.82 - lr: 0.000031 - momentum: 0.000000
2023-10-17 14:15:22,519 epoch 5 - iter 594/992 - loss 0.04434071 - time (sec): 35.36 - samples/sec: 2778.12 - lr: 0.000030 - momentum: 0.000000
2023-10-17 14:15:28,400 epoch 5 - iter 693/992 - loss 0.04509966 - time (sec): 41.25 - samples/sec: 2783.70 - lr: 0.000029 - momentum: 0.000000
2023-10-17 14:15:34,527 epoch 5 - iter 792/992 - loss 0.04488398 - time (sec): 47.37 - samples/sec: 2777.19 - lr: 0.000029 - momentum: 0.000000
2023-10-17 14:15:40,289 epoch 5 - iter 891/992 - loss 0.04506764 - time (sec): 53.13 - samples/sec: 2779.31 - lr: 0.000028 - momentum: 0.000000
2023-10-17 14:15:46,075 epoch 5 - iter 990/992 - loss 0.04560631 - time (sec): 58.92 - samples/sec: 2776.63 - lr: 0.000028 - momentum: 0.000000
2023-10-17 14:15:46,209 ----------------------------------------------------------------------------------------------------
2023-10-17 14:15:46,209 EPOCH 5 done: loss 0.0455 - lr: 0.000028
2023-10-17 14:15:49,653 DEV : loss 0.16570182144641876 - f1-score (micro avg) 0.7598
2023-10-17 14:15:49,677 saving best model
2023-10-17 14:15:50,266 ----------------------------------------------------------------------------------------------------
2023-10-17 14:15:56,488 epoch 6 - iter 99/992 - loss 0.03548701 - time (sec): 6.22 - samples/sec: 2691.46 - lr: 0.000027 - momentum: 0.000000
2023-10-17 14:16:02,509 epoch 6 - iter 198/992 - loss 0.03164832 - time (sec): 12.24 - samples/sec: 2707.48 - lr: 0.000027 - momentum: 0.000000
2023-10-17 14:16:08,671 epoch 6 - iter 297/992 - loss 0.03207736 - time (sec): 18.40 - samples/sec: 2692.61 - lr: 0.000026 - momentum: 0.000000
2023-10-17 14:16:14,599 epoch 6 - iter 396/992 - loss 0.03177863 - time (sec): 24.33 - samples/sec: 2718.45 - lr: 0.000026 - momentum: 0.000000
2023-10-17 14:16:20,727 epoch 6 - iter 495/992 - loss 0.03190660 - time (sec): 30.46 - samples/sec: 2725.37 - lr: 0.000025 - momentum: 0.000000
2023-10-17 14:16:26,730 epoch 6 - iter 594/992 - loss 0.03263403 - time (sec): 36.46 - samples/sec: 2715.91 - lr: 0.000024 - momentum: 0.000000
2023-10-17 14:16:32,715 epoch 6 - iter 693/992 - loss 0.03190880 - time (sec): 42.45 - samples/sec: 2743.76 - lr: 0.000024 - momentum: 0.000000
2023-10-17 14:16:38,628 epoch 6 - iter 792/992 - loss 0.03117610 - time (sec): 48.36 - samples/sec: 2738.62 - lr: 0.000023 - momentum: 0.000000
2023-10-17 14:16:44,371 epoch 6 - iter 891/992 - loss 0.03097983 - time (sec): 54.10 - samples/sec: 2738.45 - lr: 0.000023 - momentum: 0.000000
2023-10-17 14:16:50,118 epoch 6 - iter 990/992 - loss 0.03122063 - time (sec): 59.85 - samples/sec: 2734.90 - lr: 0.000022 - momentum: 0.000000
2023-10-17 14:16:50,219 ----------------------------------------------------------------------------------------------------
2023-10-17 14:16:50,219 EPOCH 6 done: loss 0.0312 - lr: 0.000022
2023-10-17 14:16:53,684 DEV : loss 0.18113002181053162 - f1-score (micro avg) 0.767
2023-10-17 14:16:53,705 saving best model
2023-10-17 14:16:54,157 ----------------------------------------------------------------------------------------------------
2023-10-17 14:16:59,988 epoch 7 - iter 99/992 - loss 0.02597015 - time (sec): 5.83 - samples/sec: 2684.60 - lr: 0.000022 - momentum: 0.000000
2023-10-17 14:17:06,231 epoch 7 - iter 198/992 - loss 0.02169306 - time (sec): 12.07 - samples/sec: 2712.41 - lr: 0.000021 - momentum: 0.000000
2023-10-17 14:17:12,255 epoch 7 - iter 297/992 - loss 0.02197468 - time (sec): 18.10 - samples/sec: 2712.38 - lr: 0.000021 - momentum: 0.000000
2023-10-17 14:17:18,365 epoch 7 - iter 396/992 - loss 0.02241929 - time (sec): 24.21 - samples/sec: 2694.37 - lr: 0.000020 - momentum: 0.000000
2023-10-17 14:17:24,253 epoch 7 - iter 495/992 - loss 0.02267823 - time (sec): 30.10 - samples/sec: 2714.40 - lr: 0.000019 - momentum: 0.000000
2023-10-17 14:17:30,023 epoch 7 - iter 594/992 - loss 0.02246580 - time (sec): 35.87 - samples/sec: 2715.96 - lr: 0.000019 - momentum: 0.000000
2023-10-17 14:17:35,851 epoch 7 - iter 693/992 - loss 0.02190549 - time (sec): 41.69 - samples/sec: 2716.69 - lr: 0.000018 - momentum: 0.000000
2023-10-17 14:17:41,839 epoch 7 - iter 792/992 - loss 0.02218982 - time (sec): 47.68 - samples/sec: 2718.42 - lr: 0.000018 - momentum: 0.000000
2023-10-17 14:17:47,671 epoch 7 - iter 891/992 - loss 0.02196221 - time (sec): 53.51 - samples/sec: 2725.79 - lr: 0.000017 - momentum: 0.000000
2023-10-17 14:17:54,071 epoch 7 - iter 990/992 - loss 0.02230131 - time (sec): 59.91 - samples/sec: 2731.72 - lr: 0.000017 - momentum: 0.000000
2023-10-17 14:17:54,194 ----------------------------------------------------------------------------------------------------
2023-10-17 14:17:54,194 EPOCH 7 done: loss 0.0223 - lr: 0.000017
2023-10-17 14:17:58,301 DEV : loss 0.19615165889263153 - f1-score (micro avg) 0.7561
2023-10-17 14:17:58,323 ----------------------------------------------------------------------------------------------------
2023-10-17 14:18:04,361 epoch 8 - iter 99/992 - loss 0.01117805 - time (sec): 6.04 - samples/sec: 2648.12 - lr: 0.000016 - momentum: 0.000000
2023-10-17 14:18:10,390 epoch 8 - iter 198/992 - loss 0.01283703 - time (sec): 12.07 - samples/sec: 2680.35 - lr: 0.000016 - momentum: 0.000000
2023-10-17 14:18:16,340 epoch 8 - iter 297/992 - loss 0.01697295 - time (sec): 18.02 - samples/sec: 2662.18 - lr: 0.000015 - momentum: 0.000000
2023-10-17 14:18:22,180 epoch 8 - iter 396/992 - loss 0.01691545 - time (sec): 23.86 - samples/sec: 2686.17 - lr: 0.000014 - momentum: 0.000000
2023-10-17 14:18:28,167 epoch 8 - iter 495/992 - loss 0.01590682 - time (sec): 29.84 - samples/sec: 2703.03 - lr: 0.000014 - momentum: 0.000000
2023-10-17 14:18:34,030 epoch 8 - iter 594/992 - loss 0.01525603 - time (sec): 35.71 - samples/sec: 2727.45 - lr: 0.000013 - momentum: 0.000000
2023-10-17 14:18:39,887 epoch 8 - iter 693/992 - loss 0.01579587 - time (sec): 41.56 - samples/sec: 2718.25 - lr: 0.000013 - momentum: 0.000000
2023-10-17 14:18:46,164 epoch 8 - iter 792/992 - loss 0.01527372 - time (sec): 47.84 - samples/sec: 2719.87 - lr: 0.000012 - momentum: 0.000000
2023-10-17 14:18:52,001 epoch 8 - iter 891/992 - loss 0.01527021 - time (sec): 53.68 - samples/sec: 2733.81 - lr: 0.000012 - momentum: 0.000000
2023-10-17 14:18:58,203 epoch 8 - iter 990/992 - loss 0.01606507 - time (sec): 59.88 - samples/sec: 2733.60 - lr: 0.000011 - momentum: 0.000000
2023-10-17 14:18:58,303 ----------------------------------------------------------------------------------------------------
2023-10-17 14:18:58,303 EPOCH 8 done: loss 0.0160 - lr: 0.000011
2023-10-17 14:19:01,691 DEV : loss 0.22577238082885742 - f1-score (micro avg) 0.7635
2023-10-17 14:19:01,712 ----------------------------------------------------------------------------------------------------
2023-10-17 14:19:07,524 epoch 9 - iter 99/992 - loss 0.00989044 - time (sec): 5.81 - samples/sec: 2697.63 - lr: 0.000011 - momentum: 0.000000
2023-10-17 14:19:13,330 epoch 9 - iter 198/992 - loss 0.01505293 - time (sec): 11.62 - samples/sec: 2783.58 - lr: 0.000010 - momentum: 0.000000
2023-10-17 14:19:19,081 epoch 9 - iter 297/992 - loss 0.01350984 - time (sec): 17.37 - samples/sec: 2745.86 - lr: 0.000009 - momentum: 0.000000
2023-10-17 14:19:25,043 epoch 9 - iter 396/992 - loss 0.01226420 - time (sec): 23.33 - samples/sec: 2731.47 - lr: 0.000009 - momentum: 0.000000
2023-10-17 14:19:30,941 epoch 9 - iter 495/992 - loss 0.01239703 - time (sec): 29.23 - samples/sec: 2743.58 - lr: 0.000008 - momentum: 0.000000
2023-10-17 14:19:37,003 epoch 9 - iter 594/992 - loss 0.01210630 - time (sec): 35.29 - samples/sec: 2739.17 - lr: 0.000008 - momentum: 0.000000
2023-10-17 14:19:43,129 epoch 9 - iter 693/992 - loss 0.01202169 - time (sec): 41.42 - samples/sec: 2746.32 - lr: 0.000007 - momentum: 0.000000
2023-10-17 14:19:49,259 epoch 9 - iter 792/992 - loss 0.01182813 - time (sec): 47.55 - samples/sec: 2749.42 - lr: 0.000007 - momentum: 0.000000
2023-10-17 14:19:55,176 epoch 9 - iter 891/992 - loss 0.01183421 - time (sec): 53.46 - samples/sec: 2746.32 - lr: 0.000006 - momentum: 0.000000
2023-10-17 14:20:01,241 epoch 9 - iter 990/992 - loss 0.01157965 - time (sec): 59.53 - samples/sec: 2748.33 - lr: 0.000006 - momentum: 0.000000
2023-10-17 14:20:01,395 ----------------------------------------------------------------------------------------------------
2023-10-17 14:20:01,395 EPOCH 9 done: loss 0.0116 - lr: 0.000006
2023-10-17 14:20:04,942 DEV : loss 0.21687279641628265 - f1-score (micro avg) 0.7509
2023-10-17 14:20:04,963 ----------------------------------------------------------------------------------------------------
2023-10-17 14:20:10,822 epoch 10 - iter 99/992 - loss 0.00708023 - time (sec): 5.86 - samples/sec: 2779.19 - lr: 0.000005 - momentum: 0.000000
2023-10-17 14:20:16,971 epoch 10 - iter 198/992 - loss 0.00762459 - time (sec): 12.01 - samples/sec: 2789.93 - lr: 0.000004 - momentum: 0.000000
2023-10-17 14:20:22,928 epoch 10 - iter 297/992 - loss 0.00709401 - time (sec): 17.96 - samples/sec: 2760.24 - lr: 0.000004 - momentum: 0.000000
2023-10-17 14:20:28,815 epoch 10 - iter 396/992 - loss 0.00681691 - time (sec): 23.85 - samples/sec: 2739.23 - lr: 0.000003 - momentum: 0.000000
2023-10-17 14:20:34,724 epoch 10 - iter 495/992 - loss 0.00708234 - time (sec): 29.76 - samples/sec: 2750.97 - lr: 0.000003 - momentum: 0.000000
2023-10-17 14:20:40,813 epoch 10 - iter 594/992 - loss 0.00757164 - time (sec): 35.85 - samples/sec: 2743.70 - lr: 0.000002 - momentum: 0.000000
2023-10-17 14:20:46,718 epoch 10 - iter 693/992 - loss 0.00802286 - time (sec): 41.75 - samples/sec: 2750.56 - lr: 0.000002 - momentum: 0.000000
2023-10-17 14:20:52,723 epoch 10 - iter 792/992 - loss 0.00785135 - time (sec): 47.76 - samples/sec: 2749.46 - lr: 0.000001 - momentum: 0.000000
2023-10-17 14:20:58,677 epoch 10 - iter 891/992 - loss 0.00789678 - time (sec): 53.71 - samples/sec: 2743.95 - lr: 0.000001 - momentum: 0.000000
2023-10-17 14:21:04,642 epoch 10 - iter 990/992 - loss 0.00784035 - time (sec): 59.68 - samples/sec: 2743.80 - lr: 0.000000 - momentum: 0.000000
2023-10-17 14:21:04,763 ----------------------------------------------------------------------------------------------------
2023-10-17 14:21:04,763 EPOCH 10 done: loss 0.0078 - lr: 0.000000
2023-10-17 14:21:08,189 DEV : loss 0.23605531454086304 - f1-score (micro avg) 0.7495
2023-10-17 14:21:08,645 ----------------------------------------------------------------------------------------------------
2023-10-17 14:21:08,646 Loading model from best epoch ...
2023-10-17 14:21:10,060 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
2023-10-17 14:21:13,417
Results:
- F-score (micro) 0.7731
- F-score (macro) 0.6853
- Accuracy 0.6466
By class:
precision recall f1-score support
LOC 0.8043 0.8534 0.8281 655
PER 0.7066 0.8206 0.7593 223
ORG 0.5474 0.4094 0.4685 127
micro avg 0.7569 0.7900 0.7731 1005
macro avg 0.6861 0.6945 0.6853 1005
weighted avg 0.7502 0.7900 0.7674 1005
2023-10-17 14:21:13,417 ----------------------------------------------------------------------------------------------------
|