File size: 23,894 Bytes
fc5b876 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 |
2023-10-14 00:33:33,347 ----------------------------------------------------------------------------------------------------
2023-10-14 00:33:33,348 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(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): BertEncoder(
(layer): ModuleList(
(0-11): 12 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(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): BertSelfOutput(
(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): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(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)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=768, out_features=768, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=13, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-14 00:33:33,348 ----------------------------------------------------------------------------------------------------
2023-10-14 00:33:33,349 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
2023-10-14 00:33:33,349 ----------------------------------------------------------------------------------------------------
2023-10-14 00:33:33,349 Train: 7936 sentences
2023-10-14 00:33:33,349 (train_with_dev=False, train_with_test=False)
2023-10-14 00:33:33,349 ----------------------------------------------------------------------------------------------------
2023-10-14 00:33:33,349 Training Params:
2023-10-14 00:33:33,349 - learning_rate: "3e-05"
2023-10-14 00:33:33,349 - mini_batch_size: "4"
2023-10-14 00:33:33,349 - max_epochs: "10"
2023-10-14 00:33:33,349 - shuffle: "True"
2023-10-14 00:33:33,349 ----------------------------------------------------------------------------------------------------
2023-10-14 00:33:33,349 Plugins:
2023-10-14 00:33:33,349 - LinearScheduler | warmup_fraction: '0.1'
2023-10-14 00:33:33,349 ----------------------------------------------------------------------------------------------------
2023-10-14 00:33:33,349 Final evaluation on model from best epoch (best-model.pt)
2023-10-14 00:33:33,349 - metric: "('micro avg', 'f1-score')"
2023-10-14 00:33:33,349 ----------------------------------------------------------------------------------------------------
2023-10-14 00:33:33,349 Computation:
2023-10-14 00:33:33,349 - compute on device: cuda:0
2023-10-14 00:33:33,349 - embedding storage: none
2023-10-14 00:33:33,349 ----------------------------------------------------------------------------------------------------
2023-10-14 00:33:33,349 Model training base path: "hmbench-icdar/fr-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5"
2023-10-14 00:33:33,349 ----------------------------------------------------------------------------------------------------
2023-10-14 00:33:33,349 ----------------------------------------------------------------------------------------------------
2023-10-14 00:33:42,420 epoch 1 - iter 198/1984 - loss 1.84648731 - time (sec): 9.07 - samples/sec: 1702.75 - lr: 0.000003 - momentum: 0.000000
2023-10-14 00:33:51,452 epoch 1 - iter 396/1984 - loss 1.08861314 - time (sec): 18.10 - samples/sec: 1741.41 - lr: 0.000006 - momentum: 0.000000
2023-10-14 00:34:00,444 epoch 1 - iter 594/1984 - loss 0.79789327 - time (sec): 27.09 - samples/sec: 1775.55 - lr: 0.000009 - momentum: 0.000000
2023-10-14 00:34:09,335 epoch 1 - iter 792/1984 - loss 0.64340507 - time (sec): 35.98 - samples/sec: 1788.69 - lr: 0.000012 - momentum: 0.000000
2023-10-14 00:34:18,432 epoch 1 - iter 990/1984 - loss 0.55085242 - time (sec): 45.08 - samples/sec: 1792.80 - lr: 0.000015 - momentum: 0.000000
2023-10-14 00:34:27,429 epoch 1 - iter 1188/1984 - loss 0.48020788 - time (sec): 54.08 - samples/sec: 1809.66 - lr: 0.000018 - momentum: 0.000000
2023-10-14 00:34:36,495 epoch 1 - iter 1386/1984 - loss 0.43363784 - time (sec): 63.14 - samples/sec: 1801.24 - lr: 0.000021 - momentum: 0.000000
2023-10-14 00:34:45,534 epoch 1 - iter 1584/1984 - loss 0.39617398 - time (sec): 72.18 - samples/sec: 1801.87 - lr: 0.000024 - momentum: 0.000000
2023-10-14 00:34:54,461 epoch 1 - iter 1782/1984 - loss 0.36780931 - time (sec): 81.11 - samples/sec: 1803.46 - lr: 0.000027 - momentum: 0.000000
2023-10-14 00:35:03,459 epoch 1 - iter 1980/1984 - loss 0.34466223 - time (sec): 90.11 - samples/sec: 1813.44 - lr: 0.000030 - momentum: 0.000000
2023-10-14 00:35:03,679 ----------------------------------------------------------------------------------------------------
2023-10-14 00:35:03,679 EPOCH 1 done: loss 0.3439 - lr: 0.000030
2023-10-14 00:35:06,904 DEV : loss 0.11840364336967468 - f1-score (micro avg) 0.6206
2023-10-14 00:35:06,925 saving best model
2023-10-14 00:35:07,303 ----------------------------------------------------------------------------------------------------
2023-10-14 00:35:16,413 epoch 2 - iter 198/1984 - loss 0.14075901 - time (sec): 9.11 - samples/sec: 1669.48 - lr: 0.000030 - momentum: 0.000000
2023-10-14 00:35:25,436 epoch 2 - iter 396/1984 - loss 0.12400082 - time (sec): 18.13 - samples/sec: 1720.40 - lr: 0.000029 - momentum: 0.000000
2023-10-14 00:35:34,763 epoch 2 - iter 594/1984 - loss 0.12219424 - time (sec): 27.46 - samples/sec: 1716.50 - lr: 0.000029 - momentum: 0.000000
2023-10-14 00:35:43,860 epoch 2 - iter 792/1984 - loss 0.11620520 - time (sec): 36.56 - samples/sec: 1741.22 - lr: 0.000029 - momentum: 0.000000
2023-10-14 00:35:52,817 epoch 2 - iter 990/1984 - loss 0.11433495 - time (sec): 45.51 - samples/sec: 1776.12 - lr: 0.000028 - momentum: 0.000000
2023-10-14 00:36:01,818 epoch 2 - iter 1188/1984 - loss 0.11548844 - time (sec): 54.51 - samples/sec: 1792.84 - lr: 0.000028 - momentum: 0.000000
2023-10-14 00:36:10,792 epoch 2 - iter 1386/1984 - loss 0.11442529 - time (sec): 63.49 - samples/sec: 1799.30 - lr: 0.000028 - momentum: 0.000000
2023-10-14 00:36:19,715 epoch 2 - iter 1584/1984 - loss 0.11261525 - time (sec): 72.41 - samples/sec: 1798.49 - lr: 0.000027 - momentum: 0.000000
2023-10-14 00:36:29,002 epoch 2 - iter 1782/1984 - loss 0.11148968 - time (sec): 81.70 - samples/sec: 1799.89 - lr: 0.000027 - momentum: 0.000000
2023-10-14 00:36:38,044 epoch 2 - iter 1980/1984 - loss 0.11272837 - time (sec): 90.74 - samples/sec: 1801.90 - lr: 0.000027 - momentum: 0.000000
2023-10-14 00:36:38,255 ----------------------------------------------------------------------------------------------------
2023-10-14 00:36:38,256 EPOCH 2 done: loss 0.1127 - lr: 0.000027
2023-10-14 00:36:41,653 DEV : loss 0.12019851058721542 - f1-score (micro avg) 0.7239
2023-10-14 00:36:41,675 saving best model
2023-10-14 00:36:42,206 ----------------------------------------------------------------------------------------------------
2023-10-14 00:36:51,213 epoch 3 - iter 198/1984 - loss 0.06749262 - time (sec): 9.01 - samples/sec: 1676.79 - lr: 0.000026 - momentum: 0.000000
2023-10-14 00:37:00,237 epoch 3 - iter 396/1984 - loss 0.07556813 - time (sec): 18.03 - samples/sec: 1799.64 - lr: 0.000026 - momentum: 0.000000
2023-10-14 00:37:09,114 epoch 3 - iter 594/1984 - loss 0.07935076 - time (sec): 26.91 - samples/sec: 1779.72 - lr: 0.000026 - momentum: 0.000000
2023-10-14 00:37:18,151 epoch 3 - iter 792/1984 - loss 0.08102940 - time (sec): 35.94 - samples/sec: 1773.27 - lr: 0.000025 - momentum: 0.000000
2023-10-14 00:37:27,454 epoch 3 - iter 990/1984 - loss 0.07941764 - time (sec): 45.25 - samples/sec: 1800.13 - lr: 0.000025 - momentum: 0.000000
2023-10-14 00:37:36,997 epoch 3 - iter 1188/1984 - loss 0.08141394 - time (sec): 54.79 - samples/sec: 1785.56 - lr: 0.000025 - momentum: 0.000000
2023-10-14 00:37:46,204 epoch 3 - iter 1386/1984 - loss 0.08073553 - time (sec): 64.00 - samples/sec: 1790.81 - lr: 0.000024 - momentum: 0.000000
2023-10-14 00:37:55,490 epoch 3 - iter 1584/1984 - loss 0.08000645 - time (sec): 73.28 - samples/sec: 1792.85 - lr: 0.000024 - momentum: 0.000000
2023-10-14 00:38:04,457 epoch 3 - iter 1782/1984 - loss 0.08063126 - time (sec): 82.25 - samples/sec: 1786.95 - lr: 0.000024 - momentum: 0.000000
2023-10-14 00:38:13,357 epoch 3 - iter 1980/1984 - loss 0.08177117 - time (sec): 91.15 - samples/sec: 1794.98 - lr: 0.000023 - momentum: 0.000000
2023-10-14 00:38:13,537 ----------------------------------------------------------------------------------------------------
2023-10-14 00:38:13,537 EPOCH 3 done: loss 0.0817 - lr: 0.000023
2023-10-14 00:38:17,007 DEV : loss 0.1288405805826187 - f1-score (micro avg) 0.763
2023-10-14 00:38:17,032 saving best model
2023-10-14 00:38:17,503 ----------------------------------------------------------------------------------------------------
2023-10-14 00:38:26,719 epoch 4 - iter 198/1984 - loss 0.05060713 - time (sec): 9.21 - samples/sec: 1901.80 - lr: 0.000023 - momentum: 0.000000
2023-10-14 00:38:35,936 epoch 4 - iter 396/1984 - loss 0.05243662 - time (sec): 18.43 - samples/sec: 1833.07 - lr: 0.000023 - momentum: 0.000000
2023-10-14 00:38:44,927 epoch 4 - iter 594/1984 - loss 0.05486600 - time (sec): 27.42 - samples/sec: 1826.93 - lr: 0.000022 - momentum: 0.000000
2023-10-14 00:38:54,019 epoch 4 - iter 792/1984 - loss 0.05512558 - time (sec): 36.51 - samples/sec: 1809.34 - lr: 0.000022 - momentum: 0.000000
2023-10-14 00:39:03,272 epoch 4 - iter 990/1984 - loss 0.05454064 - time (sec): 45.76 - samples/sec: 1807.82 - lr: 0.000022 - momentum: 0.000000
2023-10-14 00:39:12,825 epoch 4 - iter 1188/1984 - loss 0.05627708 - time (sec): 55.32 - samples/sec: 1785.28 - lr: 0.000021 - momentum: 0.000000
2023-10-14 00:39:21,891 epoch 4 - iter 1386/1984 - loss 0.05694993 - time (sec): 64.38 - samples/sec: 1775.39 - lr: 0.000021 - momentum: 0.000000
2023-10-14 00:39:30,648 epoch 4 - iter 1584/1984 - loss 0.05757246 - time (sec): 73.14 - samples/sec: 1784.06 - lr: 0.000021 - momentum: 0.000000
2023-10-14 00:39:39,551 epoch 4 - iter 1782/1984 - loss 0.05843888 - time (sec): 82.04 - samples/sec: 1781.71 - lr: 0.000020 - momentum: 0.000000
2023-10-14 00:39:48,827 epoch 4 - iter 1980/1984 - loss 0.06142881 - time (sec): 91.32 - samples/sec: 1792.12 - lr: 0.000020 - momentum: 0.000000
2023-10-14 00:39:49,033 ----------------------------------------------------------------------------------------------------
2023-10-14 00:39:49,033 EPOCH 4 done: loss 0.0614 - lr: 0.000020
2023-10-14 00:39:52,419 DEV : loss 0.14116324484348297 - f1-score (micro avg) 0.7451
2023-10-14 00:39:52,439 ----------------------------------------------------------------------------------------------------
2023-10-14 00:40:01,484 epoch 5 - iter 198/1984 - loss 0.04525493 - time (sec): 9.04 - samples/sec: 1830.03 - lr: 0.000020 - momentum: 0.000000
2023-10-14 00:40:10,539 epoch 5 - iter 396/1984 - loss 0.04542360 - time (sec): 18.10 - samples/sec: 1842.65 - lr: 0.000019 - momentum: 0.000000
2023-10-14 00:40:19,522 epoch 5 - iter 594/1984 - loss 0.04551483 - time (sec): 27.08 - samples/sec: 1820.33 - lr: 0.000019 - momentum: 0.000000
2023-10-14 00:40:28,471 epoch 5 - iter 792/1984 - loss 0.04325209 - time (sec): 36.03 - samples/sec: 1828.91 - lr: 0.000019 - momentum: 0.000000
2023-10-14 00:40:37,434 epoch 5 - iter 990/1984 - loss 0.04250521 - time (sec): 44.99 - samples/sec: 1829.61 - lr: 0.000018 - momentum: 0.000000
2023-10-14 00:40:46,472 epoch 5 - iter 1188/1984 - loss 0.04318675 - time (sec): 54.03 - samples/sec: 1830.16 - lr: 0.000018 - momentum: 0.000000
2023-10-14 00:40:55,443 epoch 5 - iter 1386/1984 - loss 0.04436125 - time (sec): 63.00 - samples/sec: 1812.70 - lr: 0.000018 - momentum: 0.000000
2023-10-14 00:41:04,539 epoch 5 - iter 1584/1984 - loss 0.04521410 - time (sec): 72.10 - samples/sec: 1820.00 - lr: 0.000017 - momentum: 0.000000
2023-10-14 00:41:13,633 epoch 5 - iter 1782/1984 - loss 0.04534807 - time (sec): 81.19 - samples/sec: 1813.67 - lr: 0.000017 - momentum: 0.000000
2023-10-14 00:41:22,673 epoch 5 - iter 1980/1984 - loss 0.04610443 - time (sec): 90.23 - samples/sec: 1814.23 - lr: 0.000017 - momentum: 0.000000
2023-10-14 00:41:22,851 ----------------------------------------------------------------------------------------------------
2023-10-14 00:41:22,851 EPOCH 5 done: loss 0.0461 - lr: 0.000017
2023-10-14 00:41:26,783 DEV : loss 0.15994729101657867 - f1-score (micro avg) 0.7573
2023-10-14 00:41:26,804 ----------------------------------------------------------------------------------------------------
2023-10-14 00:41:36,094 epoch 6 - iter 198/1984 - loss 0.04012582 - time (sec): 9.29 - samples/sec: 1865.57 - lr: 0.000016 - momentum: 0.000000
2023-10-14 00:41:45,056 epoch 6 - iter 396/1984 - loss 0.03880028 - time (sec): 18.25 - samples/sec: 1830.40 - lr: 0.000016 - momentum: 0.000000
2023-10-14 00:41:54,099 epoch 6 - iter 594/1984 - loss 0.03591012 - time (sec): 27.29 - samples/sec: 1803.29 - lr: 0.000016 - momentum: 0.000000
2023-10-14 00:42:03,182 epoch 6 - iter 792/1984 - loss 0.03583443 - time (sec): 36.38 - samples/sec: 1810.54 - lr: 0.000015 - momentum: 0.000000
2023-10-14 00:42:12,153 epoch 6 - iter 990/1984 - loss 0.03618946 - time (sec): 45.35 - samples/sec: 1818.44 - lr: 0.000015 - momentum: 0.000000
2023-10-14 00:42:21,087 epoch 6 - iter 1188/1984 - loss 0.03540426 - time (sec): 54.28 - samples/sec: 1812.97 - lr: 0.000015 - momentum: 0.000000
2023-10-14 00:42:30,099 epoch 6 - iter 1386/1984 - loss 0.03591058 - time (sec): 63.29 - samples/sec: 1812.29 - lr: 0.000014 - momentum: 0.000000
2023-10-14 00:42:39,105 epoch 6 - iter 1584/1984 - loss 0.03527012 - time (sec): 72.30 - samples/sec: 1813.34 - lr: 0.000014 - momentum: 0.000000
2023-10-14 00:42:48,273 epoch 6 - iter 1782/1984 - loss 0.03466812 - time (sec): 81.47 - samples/sec: 1817.65 - lr: 0.000014 - momentum: 0.000000
2023-10-14 00:42:57,276 epoch 6 - iter 1980/1984 - loss 0.03459241 - time (sec): 90.47 - samples/sec: 1809.43 - lr: 0.000013 - momentum: 0.000000
2023-10-14 00:42:57,461 ----------------------------------------------------------------------------------------------------
2023-10-14 00:42:57,461 EPOCH 6 done: loss 0.0345 - lr: 0.000013
2023-10-14 00:43:00,862 DEV : loss 0.1856544464826584 - f1-score (micro avg) 0.7573
2023-10-14 00:43:00,883 ----------------------------------------------------------------------------------------------------
2023-10-14 00:43:09,962 epoch 7 - iter 198/1984 - loss 0.02135452 - time (sec): 9.08 - samples/sec: 1781.01 - lr: 0.000013 - momentum: 0.000000
2023-10-14 00:43:18,579 epoch 7 - iter 396/1984 - loss 0.02314302 - time (sec): 17.70 - samples/sec: 1814.63 - lr: 0.000013 - momentum: 0.000000
2023-10-14 00:43:27,249 epoch 7 - iter 594/1984 - loss 0.02014483 - time (sec): 26.36 - samples/sec: 1861.63 - lr: 0.000012 - momentum: 0.000000
2023-10-14 00:43:35,932 epoch 7 - iter 792/1984 - loss 0.02217636 - time (sec): 35.05 - samples/sec: 1872.99 - lr: 0.000012 - momentum: 0.000000
2023-10-14 00:43:44,530 epoch 7 - iter 990/1984 - loss 0.02248853 - time (sec): 43.65 - samples/sec: 1870.13 - lr: 0.000012 - momentum: 0.000000
2023-10-14 00:43:53,208 epoch 7 - iter 1188/1984 - loss 0.02281695 - time (sec): 52.32 - samples/sec: 1878.85 - lr: 0.000011 - momentum: 0.000000
2023-10-14 00:44:01,962 epoch 7 - iter 1386/1984 - loss 0.02260346 - time (sec): 61.08 - samples/sec: 1883.39 - lr: 0.000011 - momentum: 0.000000
2023-10-14 00:44:10,588 epoch 7 - iter 1584/1984 - loss 0.02346248 - time (sec): 69.70 - samples/sec: 1881.17 - lr: 0.000011 - momentum: 0.000000
2023-10-14 00:44:19,772 epoch 7 - iter 1782/1984 - loss 0.02378911 - time (sec): 78.89 - samples/sec: 1869.24 - lr: 0.000010 - momentum: 0.000000
2023-10-14 00:44:28,729 epoch 7 - iter 1980/1984 - loss 0.02373852 - time (sec): 87.84 - samples/sec: 1862.99 - lr: 0.000010 - momentum: 0.000000
2023-10-14 00:44:28,901 ----------------------------------------------------------------------------------------------------
2023-10-14 00:44:28,901 EPOCH 7 done: loss 0.0238 - lr: 0.000010
2023-10-14 00:44:32,838 DEV : loss 0.19240804016590118 - f1-score (micro avg) 0.7615
2023-10-14 00:44:32,859 ----------------------------------------------------------------------------------------------------
2023-10-14 00:44:41,845 epoch 8 - iter 198/1984 - loss 0.01575059 - time (sec): 8.98 - samples/sec: 1912.11 - lr: 0.000010 - momentum: 0.000000
2023-10-14 00:44:50,888 epoch 8 - iter 396/1984 - loss 0.01422090 - time (sec): 18.03 - samples/sec: 1844.50 - lr: 0.000009 - momentum: 0.000000
2023-10-14 00:44:59,843 epoch 8 - iter 594/1984 - loss 0.01545793 - time (sec): 26.98 - samples/sec: 1809.45 - lr: 0.000009 - momentum: 0.000000
2023-10-14 00:45:09,054 epoch 8 - iter 792/1984 - loss 0.01632885 - time (sec): 36.19 - samples/sec: 1826.74 - lr: 0.000009 - momentum: 0.000000
2023-10-14 00:45:18,131 epoch 8 - iter 990/1984 - loss 0.01657672 - time (sec): 45.27 - samples/sec: 1833.70 - lr: 0.000008 - momentum: 0.000000
2023-10-14 00:45:27,193 epoch 8 - iter 1188/1984 - loss 0.01688354 - time (sec): 54.33 - samples/sec: 1838.93 - lr: 0.000008 - momentum: 0.000000
2023-10-14 00:45:36,076 epoch 8 - iter 1386/1984 - loss 0.01653230 - time (sec): 63.22 - samples/sec: 1837.14 - lr: 0.000008 - momentum: 0.000000
2023-10-14 00:45:45,263 epoch 8 - iter 1584/1984 - loss 0.01631411 - time (sec): 72.40 - samples/sec: 1821.74 - lr: 0.000007 - momentum: 0.000000
2023-10-14 00:45:54,512 epoch 8 - iter 1782/1984 - loss 0.01666752 - time (sec): 81.65 - samples/sec: 1811.34 - lr: 0.000007 - momentum: 0.000000
2023-10-14 00:46:03,561 epoch 8 - iter 1980/1984 - loss 0.01708603 - time (sec): 90.70 - samples/sec: 1805.56 - lr: 0.000007 - momentum: 0.000000
2023-10-14 00:46:03,739 ----------------------------------------------------------------------------------------------------
2023-10-14 00:46:03,739 EPOCH 8 done: loss 0.0171 - lr: 0.000007
2023-10-14 00:46:07,144 DEV : loss 0.20113840699195862 - f1-score (micro avg) 0.7714
2023-10-14 00:46:07,165 saving best model
2023-10-14 00:46:07,687 ----------------------------------------------------------------------------------------------------
2023-10-14 00:46:16,719 epoch 9 - iter 198/1984 - loss 0.01121190 - time (sec): 9.03 - samples/sec: 1791.90 - lr: 0.000006 - momentum: 0.000000
2023-10-14 00:46:25,702 epoch 9 - iter 396/1984 - loss 0.01335710 - time (sec): 18.01 - samples/sec: 1844.29 - lr: 0.000006 - momentum: 0.000000
2023-10-14 00:46:34,718 epoch 9 - iter 594/1984 - loss 0.01317916 - time (sec): 27.03 - samples/sec: 1851.45 - lr: 0.000006 - momentum: 0.000000
2023-10-14 00:46:43,597 epoch 9 - iter 792/1984 - loss 0.01271864 - time (sec): 35.91 - samples/sec: 1828.80 - lr: 0.000005 - momentum: 0.000000
2023-10-14 00:46:52,551 epoch 9 - iter 990/1984 - loss 0.01201250 - time (sec): 44.86 - samples/sec: 1834.11 - lr: 0.000005 - momentum: 0.000000
2023-10-14 00:47:01,618 epoch 9 - iter 1188/1984 - loss 0.01186694 - time (sec): 53.93 - samples/sec: 1830.03 - lr: 0.000005 - momentum: 0.000000
2023-10-14 00:47:10,838 epoch 9 - iter 1386/1984 - loss 0.01295976 - time (sec): 63.15 - samples/sec: 1817.33 - lr: 0.000004 - momentum: 0.000000
2023-10-14 00:47:19,883 epoch 9 - iter 1584/1984 - loss 0.01303216 - time (sec): 72.19 - samples/sec: 1825.07 - lr: 0.000004 - momentum: 0.000000
2023-10-14 00:47:28,749 epoch 9 - iter 1782/1984 - loss 0.01278500 - time (sec): 81.06 - samples/sec: 1821.39 - lr: 0.000004 - momentum: 0.000000
2023-10-14 00:47:37,683 epoch 9 - iter 1980/1984 - loss 0.01286673 - time (sec): 89.99 - samples/sec: 1817.30 - lr: 0.000003 - momentum: 0.000000
2023-10-14 00:47:37,879 ----------------------------------------------------------------------------------------------------
2023-10-14 00:47:37,879 EPOCH 9 done: loss 0.0128 - lr: 0.000003
2023-10-14 00:47:41,336 DEV : loss 0.2166709452867508 - f1-score (micro avg) 0.7711
2023-10-14 00:47:41,357 ----------------------------------------------------------------------------------------------------
2023-10-14 00:47:50,410 epoch 10 - iter 198/1984 - loss 0.00699997 - time (sec): 9.05 - samples/sec: 1941.31 - lr: 0.000003 - momentum: 0.000000
2023-10-14 00:47:59,477 epoch 10 - iter 396/1984 - loss 0.00743539 - time (sec): 18.12 - samples/sec: 1881.47 - lr: 0.000003 - momentum: 0.000000
2023-10-14 00:48:08,429 epoch 10 - iter 594/1984 - loss 0.00701043 - time (sec): 27.07 - samples/sec: 1816.63 - lr: 0.000002 - momentum: 0.000000
2023-10-14 00:48:17,379 epoch 10 - iter 792/1984 - loss 0.00691793 - time (sec): 36.02 - samples/sec: 1826.01 - lr: 0.000002 - momentum: 0.000000
2023-10-14 00:48:26,370 epoch 10 - iter 990/1984 - loss 0.00733336 - time (sec): 45.01 - samples/sec: 1829.49 - lr: 0.000002 - momentum: 0.000000
2023-10-14 00:48:35,376 epoch 10 - iter 1188/1984 - loss 0.00787838 - time (sec): 54.02 - samples/sec: 1821.42 - lr: 0.000001 - momentum: 0.000000
2023-10-14 00:48:44,332 epoch 10 - iter 1386/1984 - loss 0.00801026 - time (sec): 62.97 - samples/sec: 1820.26 - lr: 0.000001 - momentum: 0.000000
2023-10-14 00:48:53,392 epoch 10 - iter 1584/1984 - loss 0.00795308 - time (sec): 72.03 - samples/sec: 1822.69 - lr: 0.000001 - momentum: 0.000000
2023-10-14 00:49:02,320 epoch 10 - iter 1782/1984 - loss 0.00774112 - time (sec): 80.96 - samples/sec: 1826.42 - lr: 0.000000 - momentum: 0.000000
2023-10-14 00:49:11,494 epoch 10 - iter 1980/1984 - loss 0.00777068 - time (sec): 90.14 - samples/sec: 1816.02 - lr: 0.000000 - momentum: 0.000000
2023-10-14 00:49:11,673 ----------------------------------------------------------------------------------------------------
2023-10-14 00:49:11,673 EPOCH 10 done: loss 0.0078 - lr: 0.000000
2023-10-14 00:49:15,510 DEV : loss 0.22600804269313812 - f1-score (micro avg) 0.7689
2023-10-14 00:49:15,955 ----------------------------------------------------------------------------------------------------
2023-10-14 00:49:15,956 Loading model from best epoch ...
2023-10-14 00:49:17,330 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
2023-10-14 00:49:20,598
Results:
- F-score (micro) 0.7769
- F-score (macro) 0.6893
- Accuracy 0.6567
By class:
precision recall f1-score support
LOC 0.8162 0.8473 0.8315 655
PER 0.7083 0.8386 0.7680 223
ORG 0.5474 0.4094 0.4685 127
micro avg 0.7642 0.7900 0.7769 1005
macro avg 0.6906 0.6984 0.6893 1005
weighted avg 0.7583 0.7900 0.7715 1005
2023-10-14 00:49:20,598 ----------------------------------------------------------------------------------------------------
|