File size: 23,904 Bytes
f85bc34 |
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 |
2023-10-17 11:50:59,878 ----------------------------------------------------------------------------------------------------
2023-10-17 11:50:59,879 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 11:50:59,879 ----------------------------------------------------------------------------------------------------
2023-10-17 11:50:59,879 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 11:50:59,879 ----------------------------------------------------------------------------------------------------
2023-10-17 11:50:59,879 Train: 7936 sentences
2023-10-17 11:50:59,879 (train_with_dev=False, train_with_test=False)
2023-10-17 11:50:59,879 ----------------------------------------------------------------------------------------------------
2023-10-17 11:50:59,879 Training Params:
2023-10-17 11:50:59,879 - learning_rate: "5e-05"
2023-10-17 11:50:59,879 - mini_batch_size: "4"
2023-10-17 11:50:59,879 - max_epochs: "10"
2023-10-17 11:50:59,879 - shuffle: "True"
2023-10-17 11:50:59,879 ----------------------------------------------------------------------------------------------------
2023-10-17 11:50:59,879 Plugins:
2023-10-17 11:50:59,879 - TensorboardLogger
2023-10-17 11:50:59,879 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 11:50:59,879 ----------------------------------------------------------------------------------------------------
2023-10-17 11:50:59,879 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 11:50:59,879 - metric: "('micro avg', 'f1-score')"
2023-10-17 11:50:59,879 ----------------------------------------------------------------------------------------------------
2023-10-17 11:50:59,879 Computation:
2023-10-17 11:50:59,880 - compute on device: cuda:0
2023-10-17 11:50:59,880 - embedding storage: none
2023-10-17 11:50:59,880 ----------------------------------------------------------------------------------------------------
2023-10-17 11:50:59,880 Model training base path: "hmbench-icdar/fr-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-17 11:50:59,880 ----------------------------------------------------------------------------------------------------
2023-10-17 11:50:59,880 ----------------------------------------------------------------------------------------------------
2023-10-17 11:50:59,880 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 11:51:08,865 epoch 1 - iter 198/1984 - loss 1.99658672 - time (sec): 8.98 - samples/sec: 1773.62 - lr: 0.000005 - momentum: 0.000000
2023-10-17 11:51:17,977 epoch 1 - iter 396/1984 - loss 1.11484466 - time (sec): 18.10 - samples/sec: 1803.46 - lr: 0.000010 - momentum: 0.000000
2023-10-17 11:51:26,677 epoch 1 - iter 594/1984 - loss 0.82031286 - time (sec): 26.80 - samples/sec: 1822.19 - lr: 0.000015 - momentum: 0.000000
2023-10-17 11:51:35,306 epoch 1 - iter 792/1984 - loss 0.65663170 - time (sec): 35.42 - samples/sec: 1846.14 - lr: 0.000020 - momentum: 0.000000
2023-10-17 11:51:44,319 epoch 1 - iter 990/1984 - loss 0.55179417 - time (sec): 44.44 - samples/sec: 1861.84 - lr: 0.000025 - momentum: 0.000000
2023-10-17 11:51:53,348 epoch 1 - iter 1188/1984 - loss 0.49119338 - time (sec): 53.47 - samples/sec: 1852.30 - lr: 0.000030 - momentum: 0.000000
2023-10-17 11:52:02,343 epoch 1 - iter 1386/1984 - loss 0.44256195 - time (sec): 62.46 - samples/sec: 1858.68 - lr: 0.000035 - momentum: 0.000000
2023-10-17 11:52:11,442 epoch 1 - iter 1584/1984 - loss 0.40740944 - time (sec): 71.56 - samples/sec: 1839.94 - lr: 0.000040 - momentum: 0.000000
2023-10-17 11:52:20,490 epoch 1 - iter 1782/1984 - loss 0.37924328 - time (sec): 80.61 - samples/sec: 1831.13 - lr: 0.000045 - momentum: 0.000000
2023-10-17 11:52:29,468 epoch 1 - iter 1980/1984 - loss 0.35390636 - time (sec): 89.59 - samples/sec: 1827.85 - lr: 0.000050 - momentum: 0.000000
2023-10-17 11:52:29,645 ----------------------------------------------------------------------------------------------------
2023-10-17 11:52:29,645 EPOCH 1 done: loss 0.3535 - lr: 0.000050
2023-10-17 11:52:32,795 DEV : loss 0.11604664474725723 - f1-score (micro avg) 0.7365
2023-10-17 11:52:32,816 saving best model
2023-10-17 11:52:33,192 ----------------------------------------------------------------------------------------------------
2023-10-17 11:52:42,328 epoch 2 - iter 198/1984 - loss 0.14192747 - time (sec): 9.13 - samples/sec: 1705.31 - lr: 0.000049 - momentum: 0.000000
2023-10-17 11:52:51,310 epoch 2 - iter 396/1984 - loss 0.12714871 - time (sec): 18.12 - samples/sec: 1747.11 - lr: 0.000049 - momentum: 0.000000
2023-10-17 11:53:00,111 epoch 2 - iter 594/1984 - loss 0.12855381 - time (sec): 26.92 - samples/sec: 1792.02 - lr: 0.000048 - momentum: 0.000000
2023-10-17 11:53:08,839 epoch 2 - iter 792/1984 - loss 0.12867843 - time (sec): 35.65 - samples/sec: 1811.84 - lr: 0.000048 - momentum: 0.000000
2023-10-17 11:53:17,959 epoch 2 - iter 990/1984 - loss 0.12838419 - time (sec): 44.77 - samples/sec: 1825.07 - lr: 0.000047 - momentum: 0.000000
2023-10-17 11:53:26,874 epoch 2 - iter 1188/1984 - loss 0.12740182 - time (sec): 53.68 - samples/sec: 1822.71 - lr: 0.000047 - momentum: 0.000000
2023-10-17 11:53:35,927 epoch 2 - iter 1386/1984 - loss 0.12351792 - time (sec): 62.73 - samples/sec: 1822.16 - lr: 0.000046 - momentum: 0.000000
2023-10-17 11:53:44,983 epoch 2 - iter 1584/1984 - loss 0.12218031 - time (sec): 71.79 - samples/sec: 1830.55 - lr: 0.000046 - momentum: 0.000000
2023-10-17 11:53:54,028 epoch 2 - iter 1782/1984 - loss 0.12157226 - time (sec): 80.83 - samples/sec: 1829.60 - lr: 0.000045 - momentum: 0.000000
2023-10-17 11:54:03,013 epoch 2 - iter 1980/1984 - loss 0.12534668 - time (sec): 89.82 - samples/sec: 1822.84 - lr: 0.000044 - momentum: 0.000000
2023-10-17 11:54:03,190 ----------------------------------------------------------------------------------------------------
2023-10-17 11:54:03,190 EPOCH 2 done: loss 0.1253 - lr: 0.000044
2023-10-17 11:54:06,991 DEV : loss 0.09573990851640701 - f1-score (micro avg) 0.7572
2023-10-17 11:54:07,012 saving best model
2023-10-17 11:54:07,504 ----------------------------------------------------------------------------------------------------
2023-10-17 11:54:16,693 epoch 3 - iter 198/1984 - loss 0.09494312 - time (sec): 9.19 - samples/sec: 1831.31 - lr: 0.000044 - momentum: 0.000000
2023-10-17 11:54:25,610 epoch 3 - iter 396/1984 - loss 0.09329297 - time (sec): 18.10 - samples/sec: 1821.18 - lr: 0.000043 - momentum: 0.000000
2023-10-17 11:54:34,690 epoch 3 - iter 594/1984 - loss 0.09450932 - time (sec): 27.18 - samples/sec: 1832.48 - lr: 0.000043 - momentum: 0.000000
2023-10-17 11:54:43,567 epoch 3 - iter 792/1984 - loss 0.09880612 - time (sec): 36.06 - samples/sec: 1796.09 - lr: 0.000042 - momentum: 0.000000
2023-10-17 11:54:52,286 epoch 3 - iter 990/1984 - loss 0.09577827 - time (sec): 44.78 - samples/sec: 1815.05 - lr: 0.000042 - momentum: 0.000000
2023-10-17 11:55:00,816 epoch 3 - iter 1188/1984 - loss 0.09551043 - time (sec): 53.31 - samples/sec: 1829.07 - lr: 0.000041 - momentum: 0.000000
2023-10-17 11:55:09,757 epoch 3 - iter 1386/1984 - loss 0.09500722 - time (sec): 62.25 - samples/sec: 1852.97 - lr: 0.000041 - momentum: 0.000000
2023-10-17 11:55:18,875 epoch 3 - iter 1584/1984 - loss 0.09605364 - time (sec): 71.37 - samples/sec: 1848.06 - lr: 0.000040 - momentum: 0.000000
2023-10-17 11:55:27,832 epoch 3 - iter 1782/1984 - loss 0.09581236 - time (sec): 80.32 - samples/sec: 1834.01 - lr: 0.000039 - momentum: 0.000000
2023-10-17 11:55:36,807 epoch 3 - iter 1980/1984 - loss 0.09610021 - time (sec): 89.30 - samples/sec: 1832.73 - lr: 0.000039 - momentum: 0.000000
2023-10-17 11:55:36,985 ----------------------------------------------------------------------------------------------------
2023-10-17 11:55:36,986 EPOCH 3 done: loss 0.0960 - lr: 0.000039
2023-10-17 11:55:40,348 DEV : loss 0.12434609979391098 - f1-score (micro avg) 0.7354
2023-10-17 11:55:40,368 ----------------------------------------------------------------------------------------------------
2023-10-17 11:55:49,368 epoch 4 - iter 198/1984 - loss 0.05981882 - time (sec): 9.00 - samples/sec: 1857.83 - lr: 0.000038 - momentum: 0.000000
2023-10-17 11:55:58,236 epoch 4 - iter 396/1984 - loss 0.06629030 - time (sec): 17.87 - samples/sec: 1808.21 - lr: 0.000038 - momentum: 0.000000
2023-10-17 11:56:07,217 epoch 4 - iter 594/1984 - loss 0.07164876 - time (sec): 26.85 - samples/sec: 1770.52 - lr: 0.000037 - momentum: 0.000000
2023-10-17 11:56:16,630 epoch 4 - iter 792/1984 - loss 0.07351127 - time (sec): 36.26 - samples/sec: 1781.58 - lr: 0.000037 - momentum: 0.000000
2023-10-17 11:56:25,747 epoch 4 - iter 990/1984 - loss 0.07151819 - time (sec): 45.38 - samples/sec: 1786.64 - lr: 0.000036 - momentum: 0.000000
2023-10-17 11:56:34,907 epoch 4 - iter 1188/1984 - loss 0.07095217 - time (sec): 54.54 - samples/sec: 1778.05 - lr: 0.000036 - momentum: 0.000000
2023-10-17 11:56:43,989 epoch 4 - iter 1386/1984 - loss 0.07050035 - time (sec): 63.62 - samples/sec: 1786.16 - lr: 0.000035 - momentum: 0.000000
2023-10-17 11:56:53,161 epoch 4 - iter 1584/1984 - loss 0.07691480 - time (sec): 72.79 - samples/sec: 1792.55 - lr: 0.000034 - momentum: 0.000000
2023-10-17 11:57:02,381 epoch 4 - iter 1782/1984 - loss 0.07542490 - time (sec): 82.01 - samples/sec: 1799.04 - lr: 0.000034 - momentum: 0.000000
2023-10-17 11:57:11,654 epoch 4 - iter 1980/1984 - loss 0.07641851 - time (sec): 91.28 - samples/sec: 1793.27 - lr: 0.000033 - momentum: 0.000000
2023-10-17 11:57:11,837 ----------------------------------------------------------------------------------------------------
2023-10-17 11:57:11,837 EPOCH 4 done: loss 0.0763 - lr: 0.000033
2023-10-17 11:57:15,244 DEV : loss 0.18466949462890625 - f1-score (micro avg) 0.7408
2023-10-17 11:57:15,266 ----------------------------------------------------------------------------------------------------
2023-10-17 11:57:24,362 epoch 5 - iter 198/1984 - loss 0.04859297 - time (sec): 9.10 - samples/sec: 1863.76 - lr: 0.000033 - momentum: 0.000000
2023-10-17 11:57:33,455 epoch 5 - iter 396/1984 - loss 0.04635554 - time (sec): 18.19 - samples/sec: 1835.22 - lr: 0.000032 - momentum: 0.000000
2023-10-17 11:57:42,513 epoch 5 - iter 594/1984 - loss 0.04871548 - time (sec): 27.25 - samples/sec: 1821.40 - lr: 0.000032 - momentum: 0.000000
2023-10-17 11:57:51,623 epoch 5 - iter 792/1984 - loss 0.05252835 - time (sec): 36.36 - samples/sec: 1805.94 - lr: 0.000031 - momentum: 0.000000
2023-10-17 11:58:00,844 epoch 5 - iter 990/1984 - loss 0.05388850 - time (sec): 45.58 - samples/sec: 1792.29 - lr: 0.000031 - momentum: 0.000000
2023-10-17 11:58:09,855 epoch 5 - iter 1188/1984 - loss 0.05515150 - time (sec): 54.59 - samples/sec: 1778.02 - lr: 0.000030 - momentum: 0.000000
2023-10-17 11:58:19,340 epoch 5 - iter 1386/1984 - loss 0.05623579 - time (sec): 64.07 - samples/sec: 1779.81 - lr: 0.000029 - momentum: 0.000000
2023-10-17 11:58:28,807 epoch 5 - iter 1584/1984 - loss 0.05445857 - time (sec): 73.54 - samples/sec: 1779.22 - lr: 0.000029 - momentum: 0.000000
2023-10-17 11:58:37,858 epoch 5 - iter 1782/1984 - loss 0.05549020 - time (sec): 82.59 - samples/sec: 1778.92 - lr: 0.000028 - momentum: 0.000000
2023-10-17 11:58:46,900 epoch 5 - iter 1980/1984 - loss 0.05573079 - time (sec): 91.63 - samples/sec: 1786.04 - lr: 0.000028 - momentum: 0.000000
2023-10-17 11:58:47,077 ----------------------------------------------------------------------------------------------------
2023-10-17 11:58:47,077 EPOCH 5 done: loss 0.0556 - lr: 0.000028
2023-10-17 11:58:50,444 DEV : loss 0.1826779842376709 - f1-score (micro avg) 0.7561
2023-10-17 11:58:50,465 ----------------------------------------------------------------------------------------------------
2023-10-17 11:58:59,462 epoch 6 - iter 198/1984 - loss 0.03393163 - time (sec): 9.00 - samples/sec: 1795.87 - lr: 0.000027 - momentum: 0.000000
2023-10-17 11:59:08,616 epoch 6 - iter 396/1984 - loss 0.03547336 - time (sec): 18.15 - samples/sec: 1805.34 - lr: 0.000027 - momentum: 0.000000
2023-10-17 11:59:17,653 epoch 6 - iter 594/1984 - loss 0.03602657 - time (sec): 27.19 - samples/sec: 1787.83 - lr: 0.000026 - momentum: 0.000000
2023-10-17 11:59:26,822 epoch 6 - iter 792/1984 - loss 0.03952462 - time (sec): 36.35 - samples/sec: 1804.95 - lr: 0.000026 - momentum: 0.000000
2023-10-17 11:59:36,151 epoch 6 - iter 990/1984 - loss 0.03912622 - time (sec): 45.68 - samples/sec: 1810.09 - lr: 0.000025 - momentum: 0.000000
2023-10-17 11:59:45,232 epoch 6 - iter 1188/1984 - loss 0.03870866 - time (sec): 54.77 - samples/sec: 1813.32 - lr: 0.000024 - momentum: 0.000000
2023-10-17 11:59:53,643 epoch 6 - iter 1386/1984 - loss 0.04046966 - time (sec): 63.18 - samples/sec: 1815.61 - lr: 0.000024 - momentum: 0.000000
2023-10-17 12:00:02,231 epoch 6 - iter 1584/1984 - loss 0.03983182 - time (sec): 71.76 - samples/sec: 1827.05 - lr: 0.000023 - momentum: 0.000000
2023-10-17 12:00:10,739 epoch 6 - iter 1782/1984 - loss 0.03988919 - time (sec): 80.27 - samples/sec: 1834.23 - lr: 0.000023 - momentum: 0.000000
2023-10-17 12:00:19,695 epoch 6 - iter 1980/1984 - loss 0.04066461 - time (sec): 89.23 - samples/sec: 1834.70 - lr: 0.000022 - momentum: 0.000000
2023-10-17 12:00:19,885 ----------------------------------------------------------------------------------------------------
2023-10-17 12:00:19,885 EPOCH 6 done: loss 0.0406 - lr: 0.000022
2023-10-17 12:00:23,280 DEV : loss 0.1922488510608673 - f1-score (micro avg) 0.7417
2023-10-17 12:00:23,303 ----------------------------------------------------------------------------------------------------
2023-10-17 12:00:32,599 epoch 7 - iter 198/1984 - loss 0.03507544 - time (sec): 9.29 - samples/sec: 1779.65 - lr: 0.000022 - momentum: 0.000000
2023-10-17 12:00:41,907 epoch 7 - iter 396/1984 - loss 0.03033014 - time (sec): 18.60 - samples/sec: 1789.70 - lr: 0.000021 - momentum: 0.000000
2023-10-17 12:00:51,166 epoch 7 - iter 594/1984 - loss 0.02968150 - time (sec): 27.86 - samples/sec: 1802.14 - lr: 0.000021 - momentum: 0.000000
2023-10-17 12:01:00,212 epoch 7 - iter 792/1984 - loss 0.02789631 - time (sec): 36.91 - samples/sec: 1810.72 - lr: 0.000020 - momentum: 0.000000
2023-10-17 12:01:08,802 epoch 7 - iter 990/1984 - loss 0.02776484 - time (sec): 45.50 - samples/sec: 1838.23 - lr: 0.000019 - momentum: 0.000000
2023-10-17 12:01:17,842 epoch 7 - iter 1188/1984 - loss 0.02661241 - time (sec): 54.54 - samples/sec: 1822.87 - lr: 0.000019 - momentum: 0.000000
2023-10-17 12:01:26,856 epoch 7 - iter 1386/1984 - loss 0.02640200 - time (sec): 63.55 - samples/sec: 1809.99 - lr: 0.000018 - momentum: 0.000000
2023-10-17 12:01:35,967 epoch 7 - iter 1584/1984 - loss 0.02663353 - time (sec): 72.66 - samples/sec: 1806.05 - lr: 0.000018 - momentum: 0.000000
2023-10-17 12:01:44,950 epoch 7 - iter 1782/1984 - loss 0.02669130 - time (sec): 81.64 - samples/sec: 1798.68 - lr: 0.000017 - momentum: 0.000000
2023-10-17 12:01:54,310 epoch 7 - iter 1980/1984 - loss 0.02730554 - time (sec): 91.00 - samples/sec: 1798.73 - lr: 0.000017 - momentum: 0.000000
2023-10-17 12:01:54,485 ----------------------------------------------------------------------------------------------------
2023-10-17 12:01:54,485 EPOCH 7 done: loss 0.0273 - lr: 0.000017
2023-10-17 12:01:58,292 DEV : loss 0.18583810329437256 - f1-score (micro avg) 0.756
2023-10-17 12:01:58,313 ----------------------------------------------------------------------------------------------------
2023-10-17 12:02:07,377 epoch 8 - iter 198/1984 - loss 0.01962946 - time (sec): 9.06 - samples/sec: 1832.57 - lr: 0.000016 - momentum: 0.000000
2023-10-17 12:02:16,417 epoch 8 - iter 396/1984 - loss 0.02070604 - time (sec): 18.10 - samples/sec: 1860.49 - lr: 0.000016 - momentum: 0.000000
2023-10-17 12:02:25,338 epoch 8 - iter 594/1984 - loss 0.02069916 - time (sec): 27.02 - samples/sec: 1830.33 - lr: 0.000015 - momentum: 0.000000
2023-10-17 12:02:34,277 epoch 8 - iter 792/1984 - loss 0.02086583 - time (sec): 35.96 - samples/sec: 1822.60 - lr: 0.000014 - momentum: 0.000000
2023-10-17 12:02:43,312 epoch 8 - iter 990/1984 - loss 0.01924450 - time (sec): 45.00 - samples/sec: 1831.26 - lr: 0.000014 - momentum: 0.000000
2023-10-17 12:02:52,365 epoch 8 - iter 1188/1984 - loss 0.01949509 - time (sec): 54.05 - samples/sec: 1819.84 - lr: 0.000013 - momentum: 0.000000
2023-10-17 12:03:01,315 epoch 8 - iter 1386/1984 - loss 0.01960191 - time (sec): 63.00 - samples/sec: 1804.63 - lr: 0.000013 - momentum: 0.000000
2023-10-17 12:03:10,558 epoch 8 - iter 1584/1984 - loss 0.02035400 - time (sec): 72.24 - samples/sec: 1811.71 - lr: 0.000012 - momentum: 0.000000
2023-10-17 12:03:19,519 epoch 8 - iter 1782/1984 - loss 0.01981080 - time (sec): 81.20 - samples/sec: 1809.61 - lr: 0.000012 - momentum: 0.000000
2023-10-17 12:03:28,832 epoch 8 - iter 1980/1984 - loss 0.01959358 - time (sec): 90.52 - samples/sec: 1808.27 - lr: 0.000011 - momentum: 0.000000
2023-10-17 12:03:29,026 ----------------------------------------------------------------------------------------------------
2023-10-17 12:03:29,027 EPOCH 8 done: loss 0.0196 - lr: 0.000011
2023-10-17 12:03:32,455 DEV : loss 0.23994652926921844 - f1-score (micro avg) 0.7582
2023-10-17 12:03:32,476 saving best model
2023-10-17 12:03:32,872 ----------------------------------------------------------------------------------------------------
2023-10-17 12:03:42,075 epoch 9 - iter 198/1984 - loss 0.01369371 - time (sec): 9.20 - samples/sec: 1698.60 - lr: 0.000011 - momentum: 0.000000
2023-10-17 12:03:51,142 epoch 9 - iter 396/1984 - loss 0.01542831 - time (sec): 18.27 - samples/sec: 1744.84 - lr: 0.000010 - momentum: 0.000000
2023-10-17 12:04:00,413 epoch 9 - iter 594/1984 - loss 0.01701451 - time (sec): 27.54 - samples/sec: 1752.27 - lr: 0.000009 - momentum: 0.000000
2023-10-17 12:04:09,646 epoch 9 - iter 792/1984 - loss 0.01617942 - time (sec): 36.77 - samples/sec: 1769.71 - lr: 0.000009 - momentum: 0.000000
2023-10-17 12:04:18,868 epoch 9 - iter 990/1984 - loss 0.01513742 - time (sec): 45.99 - samples/sec: 1764.51 - lr: 0.000008 - momentum: 0.000000
2023-10-17 12:04:28,200 epoch 9 - iter 1188/1984 - loss 0.01474000 - time (sec): 55.33 - samples/sec: 1778.49 - lr: 0.000008 - momentum: 0.000000
2023-10-17 12:04:37,255 epoch 9 - iter 1386/1984 - loss 0.01453604 - time (sec): 64.38 - samples/sec: 1788.32 - lr: 0.000007 - momentum: 0.000000
2023-10-17 12:04:46,295 epoch 9 - iter 1584/1984 - loss 0.01380448 - time (sec): 73.42 - samples/sec: 1786.61 - lr: 0.000007 - momentum: 0.000000
2023-10-17 12:04:55,527 epoch 9 - iter 1782/1984 - loss 0.01369184 - time (sec): 82.65 - samples/sec: 1786.61 - lr: 0.000006 - momentum: 0.000000
2023-10-17 12:05:04,642 epoch 9 - iter 1980/1984 - loss 0.01333927 - time (sec): 91.77 - samples/sec: 1784.22 - lr: 0.000006 - momentum: 0.000000
2023-10-17 12:05:04,808 ----------------------------------------------------------------------------------------------------
2023-10-17 12:05:04,808 EPOCH 9 done: loss 0.0133 - lr: 0.000006
2023-10-17 12:05:08,300 DEV : loss 0.25141069293022156 - f1-score (micro avg) 0.7495
2023-10-17 12:05:08,331 ----------------------------------------------------------------------------------------------------
2023-10-17 12:05:19,095 epoch 10 - iter 198/1984 - loss 0.00791381 - time (sec): 10.76 - samples/sec: 1555.82 - lr: 0.000005 - momentum: 0.000000
2023-10-17 12:05:28,038 epoch 10 - iter 396/1984 - loss 0.00907342 - time (sec): 19.70 - samples/sec: 1675.12 - lr: 0.000004 - momentum: 0.000000
2023-10-17 12:05:37,094 epoch 10 - iter 594/1984 - loss 0.00841106 - time (sec): 28.76 - samples/sec: 1736.20 - lr: 0.000004 - momentum: 0.000000
2023-10-17 12:05:46,297 epoch 10 - iter 792/1984 - loss 0.00778224 - time (sec): 37.96 - samples/sec: 1732.33 - lr: 0.000003 - momentum: 0.000000
2023-10-17 12:05:55,632 epoch 10 - iter 990/1984 - loss 0.00790979 - time (sec): 47.30 - samples/sec: 1754.39 - lr: 0.000003 - momentum: 0.000000
2023-10-17 12:06:06,086 epoch 10 - iter 1188/1984 - loss 0.00810074 - time (sec): 57.75 - samples/sec: 1728.90 - lr: 0.000002 - momentum: 0.000000
2023-10-17 12:06:16,433 epoch 10 - iter 1386/1984 - loss 0.00812282 - time (sec): 68.10 - samples/sec: 1708.90 - lr: 0.000002 - momentum: 0.000000
2023-10-17 12:06:26,896 epoch 10 - iter 1584/1984 - loss 0.00827390 - time (sec): 78.56 - samples/sec: 1683.18 - lr: 0.000001 - momentum: 0.000000
2023-10-17 12:06:37,153 epoch 10 - iter 1782/1984 - loss 0.00865749 - time (sec): 88.82 - samples/sec: 1664.10 - lr: 0.000001 - momentum: 0.000000
2023-10-17 12:06:47,233 epoch 10 - iter 1980/1984 - loss 0.00828889 - time (sec): 98.90 - samples/sec: 1655.75 - lr: 0.000000 - momentum: 0.000000
2023-10-17 12:06:47,441 ----------------------------------------------------------------------------------------------------
2023-10-17 12:06:47,441 EPOCH 10 done: loss 0.0083 - lr: 0.000000
2023-10-17 12:06:51,115 DEV : loss 0.2530412971973419 - f1-score (micro avg) 0.7575
2023-10-17 12:06:51,546 ----------------------------------------------------------------------------------------------------
2023-10-17 12:06:51,547 Loading model from best epoch ...
2023-10-17 12:06:52,995 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 12:06:56,574
Results:
- F-score (micro) 0.7556
- F-score (macro) 0.6677
- Accuracy 0.6401
By class:
precision recall f1-score support
LOC 0.8259 0.8183 0.8221 655
PER 0.6522 0.8072 0.7214 223
ORG 0.5000 0.4252 0.4596 127
micro avg 0.7454 0.7662 0.7556 1005
macro avg 0.6594 0.6836 0.6677 1005
weighted avg 0.7462 0.7662 0.7539 1005
2023-10-17 12:06:56,574 ----------------------------------------------------------------------------------------------------
|