File size: 24,153 Bytes
9df28b6 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 |
2023-10-17 08:18:18,987 ----------------------------------------------------------------------------------------------------
2023-10-17 08:18:18,988 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=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 08:18:18,988 ----------------------------------------------------------------------------------------------------
2023-10-17 08:18:18,989 MultiCorpus: 1100 train + 206 dev + 240 test sentences
- NER_HIPE_2022 Corpus: 1100 train + 206 dev + 240 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/de/with_doc_seperator
2023-10-17 08:18:18,989 ----------------------------------------------------------------------------------------------------
2023-10-17 08:18:18,989 Train: 1100 sentences
2023-10-17 08:18:18,989 (train_with_dev=False, train_with_test=False)
2023-10-17 08:18:18,989 ----------------------------------------------------------------------------------------------------
2023-10-17 08:18:18,989 Training Params:
2023-10-17 08:18:18,989 - learning_rate: "3e-05"
2023-10-17 08:18:18,989 - mini_batch_size: "4"
2023-10-17 08:18:18,989 - max_epochs: "10"
2023-10-17 08:18:18,989 - shuffle: "True"
2023-10-17 08:18:18,989 ----------------------------------------------------------------------------------------------------
2023-10-17 08:18:18,990 Plugins:
2023-10-17 08:18:18,990 - TensorboardLogger
2023-10-17 08:18:18,990 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 08:18:18,990 ----------------------------------------------------------------------------------------------------
2023-10-17 08:18:18,990 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 08:18:18,990 - metric: "('micro avg', 'f1-score')"
2023-10-17 08:18:18,990 ----------------------------------------------------------------------------------------------------
2023-10-17 08:18:18,990 Computation:
2023-10-17 08:18:18,990 - compute on device: cuda:0
2023-10-17 08:18:18,990 - embedding storage: none
2023-10-17 08:18:18,990 ----------------------------------------------------------------------------------------------------
2023-10-17 08:18:18,990 Model training base path: "hmbench-ajmc/de-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-17 08:18:18,990 ----------------------------------------------------------------------------------------------------
2023-10-17 08:18:18,990 ----------------------------------------------------------------------------------------------------
2023-10-17 08:18:18,990 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 08:18:21,395 epoch 1 - iter 27/275 - loss 3.52317630 - time (sec): 2.40 - samples/sec: 889.99 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:18:22,640 epoch 1 - iter 54/275 - loss 2.97222654 - time (sec): 3.65 - samples/sec: 1166.58 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:18:23,847 epoch 1 - iter 81/275 - loss 2.46172200 - time (sec): 4.86 - samples/sec: 1322.50 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:18:25,088 epoch 1 - iter 108/275 - loss 1.97206923 - time (sec): 6.10 - samples/sec: 1432.99 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:18:26,313 epoch 1 - iter 135/275 - loss 1.66054933 - time (sec): 7.32 - samples/sec: 1525.26 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:18:27,520 epoch 1 - iter 162/275 - loss 1.46914393 - time (sec): 8.53 - samples/sec: 1586.70 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:18:28,712 epoch 1 - iter 189/275 - loss 1.31842235 - time (sec): 9.72 - samples/sec: 1612.89 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:18:29,926 epoch 1 - iter 216/275 - loss 1.20141186 - time (sec): 10.93 - samples/sec: 1639.21 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:18:31,173 epoch 1 - iter 243/275 - loss 1.10689649 - time (sec): 12.18 - samples/sec: 1648.32 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:18:32,501 epoch 1 - iter 270/275 - loss 1.01950699 - time (sec): 13.51 - samples/sec: 1661.37 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:18:32,713 ----------------------------------------------------------------------------------------------------
2023-10-17 08:18:32,713 EPOCH 1 done: loss 1.0079 - lr: 0.000029
2023-10-17 08:18:33,248 DEV : loss 0.20805643498897552 - f1-score (micro avg) 0.6826
2023-10-17 08:18:33,253 saving best model
2023-10-17 08:18:33,620 ----------------------------------------------------------------------------------------------------
2023-10-17 08:18:34,872 epoch 2 - iter 27/275 - loss 0.25196567 - time (sec): 1.25 - samples/sec: 1920.87 - lr: 0.000030 - momentum: 0.000000
2023-10-17 08:18:36,149 epoch 2 - iter 54/275 - loss 0.23255918 - time (sec): 2.53 - samples/sec: 1806.59 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:18:37,438 epoch 2 - iter 81/275 - loss 0.23733841 - time (sec): 3.82 - samples/sec: 1792.91 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:18:38,697 epoch 2 - iter 108/275 - loss 0.23047487 - time (sec): 5.08 - samples/sec: 1789.25 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:18:39,965 epoch 2 - iter 135/275 - loss 0.20976341 - time (sec): 6.34 - samples/sec: 1772.08 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:18:41,271 epoch 2 - iter 162/275 - loss 0.20039640 - time (sec): 7.65 - samples/sec: 1762.10 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:18:42,533 epoch 2 - iter 189/275 - loss 0.19365150 - time (sec): 8.91 - samples/sec: 1751.79 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:18:43,812 epoch 2 - iter 216/275 - loss 0.18866455 - time (sec): 10.19 - samples/sec: 1736.70 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:18:45,068 epoch 2 - iter 243/275 - loss 0.18395908 - time (sec): 11.45 - samples/sec: 1748.72 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:18:46,313 epoch 2 - iter 270/275 - loss 0.17611786 - time (sec): 12.69 - samples/sec: 1763.41 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:18:46,559 ----------------------------------------------------------------------------------------------------
2023-10-17 08:18:46,560 EPOCH 2 done: loss 0.1740 - lr: 0.000027
2023-10-17 08:18:47,242 DEV : loss 0.16901393234729767 - f1-score (micro avg) 0.7929
2023-10-17 08:18:47,246 saving best model
2023-10-17 08:18:47,722 ----------------------------------------------------------------------------------------------------
2023-10-17 08:18:48,941 epoch 3 - iter 27/275 - loss 0.10060364 - time (sec): 1.22 - samples/sec: 1741.81 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:18:50,135 epoch 3 - iter 54/275 - loss 0.10019335 - time (sec): 2.41 - samples/sec: 1896.16 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:18:51,322 epoch 3 - iter 81/275 - loss 0.08798651 - time (sec): 3.60 - samples/sec: 1849.18 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:18:52,704 epoch 3 - iter 108/275 - loss 0.09566087 - time (sec): 4.98 - samples/sec: 1745.02 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:18:53,897 epoch 3 - iter 135/275 - loss 0.10177219 - time (sec): 6.17 - samples/sec: 1804.29 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:18:55,069 epoch 3 - iter 162/275 - loss 0.10278867 - time (sec): 7.35 - samples/sec: 1808.06 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:18:56,239 epoch 3 - iter 189/275 - loss 0.10878498 - time (sec): 8.52 - samples/sec: 1855.28 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:18:57,406 epoch 3 - iter 216/275 - loss 0.10765900 - time (sec): 9.68 - samples/sec: 1843.84 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:18:58,582 epoch 3 - iter 243/275 - loss 0.10656534 - time (sec): 10.86 - samples/sec: 1842.94 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:18:59,761 epoch 3 - iter 270/275 - loss 0.11369212 - time (sec): 12.04 - samples/sec: 1858.79 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:18:59,977 ----------------------------------------------------------------------------------------------------
2023-10-17 08:18:59,977 EPOCH 3 done: loss 0.1126 - lr: 0.000023
2023-10-17 08:19:00,681 DEV : loss 0.15869425237178802 - f1-score (micro avg) 0.8309
2023-10-17 08:19:00,686 saving best model
2023-10-17 08:19:01,142 ----------------------------------------------------------------------------------------------------
2023-10-17 08:19:02,386 epoch 4 - iter 27/275 - loss 0.06874097 - time (sec): 1.24 - samples/sec: 1949.41 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:19:03,631 epoch 4 - iter 54/275 - loss 0.08195825 - time (sec): 2.49 - samples/sec: 1826.88 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:19:04,887 epoch 4 - iter 81/275 - loss 0.08181971 - time (sec): 3.74 - samples/sec: 1795.67 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:19:06,127 epoch 4 - iter 108/275 - loss 0.07748029 - time (sec): 4.98 - samples/sec: 1780.79 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:19:07,424 epoch 4 - iter 135/275 - loss 0.08497099 - time (sec): 6.28 - samples/sec: 1797.29 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:19:08,724 epoch 4 - iter 162/275 - loss 0.09504136 - time (sec): 7.58 - samples/sec: 1766.63 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:19:09,956 epoch 4 - iter 189/275 - loss 0.09094486 - time (sec): 8.81 - samples/sec: 1772.94 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:19:11,184 epoch 4 - iter 216/275 - loss 0.08800892 - time (sec): 10.04 - samples/sec: 1784.02 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:19:12,403 epoch 4 - iter 243/275 - loss 0.08680239 - time (sec): 11.26 - samples/sec: 1792.19 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:19:13,608 epoch 4 - iter 270/275 - loss 0.08449332 - time (sec): 12.46 - samples/sec: 1788.89 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:19:13,845 ----------------------------------------------------------------------------------------------------
2023-10-17 08:19:13,846 EPOCH 4 done: loss 0.0831 - lr: 0.000020
2023-10-17 08:19:14,565 DEV : loss 0.16908997297286987 - f1-score (micro avg) 0.8429
2023-10-17 08:19:14,573 saving best model
2023-10-17 08:19:15,204 ----------------------------------------------------------------------------------------------------
2023-10-17 08:19:16,662 epoch 5 - iter 27/275 - loss 0.08528267 - time (sec): 1.45 - samples/sec: 1566.38 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:19:18,127 epoch 5 - iter 54/275 - loss 0.07615872 - time (sec): 2.92 - samples/sec: 1573.43 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:19:19,459 epoch 5 - iter 81/275 - loss 0.07198106 - time (sec): 4.25 - samples/sec: 1618.80 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:19:20,723 epoch 5 - iter 108/275 - loss 0.07243756 - time (sec): 5.52 - samples/sec: 1637.53 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:19:21,943 epoch 5 - iter 135/275 - loss 0.06687607 - time (sec): 6.74 - samples/sec: 1638.71 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:19:23,173 epoch 5 - iter 162/275 - loss 0.06194550 - time (sec): 7.97 - samples/sec: 1646.70 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:19:24,396 epoch 5 - iter 189/275 - loss 0.06101984 - time (sec): 9.19 - samples/sec: 1675.13 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:19:25,682 epoch 5 - iter 216/275 - loss 0.07343089 - time (sec): 10.47 - samples/sec: 1709.15 - lr: 0.000017 - momentum: 0.000000
2023-10-17 08:19:26,912 epoch 5 - iter 243/275 - loss 0.07128583 - time (sec): 11.70 - samples/sec: 1727.91 - lr: 0.000017 - momentum: 0.000000
2023-10-17 08:19:28,168 epoch 5 - iter 270/275 - loss 0.06883251 - time (sec): 12.96 - samples/sec: 1723.92 - lr: 0.000017 - momentum: 0.000000
2023-10-17 08:19:28,404 ----------------------------------------------------------------------------------------------------
2023-10-17 08:19:28,405 EPOCH 5 done: loss 0.0675 - lr: 0.000017
2023-10-17 08:19:29,050 DEV : loss 0.20913541316986084 - f1-score (micro avg) 0.8408
2023-10-17 08:19:29,055 ----------------------------------------------------------------------------------------------------
2023-10-17 08:19:30,327 epoch 6 - iter 27/275 - loss 0.02455276 - time (sec): 1.27 - samples/sec: 1684.88 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:19:31,573 epoch 6 - iter 54/275 - loss 0.04978974 - time (sec): 2.52 - samples/sec: 1936.57 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:19:32,812 epoch 6 - iter 81/275 - loss 0.05084963 - time (sec): 3.76 - samples/sec: 1861.04 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:19:34,085 epoch 6 - iter 108/275 - loss 0.04648955 - time (sec): 5.03 - samples/sec: 1811.81 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:19:35,338 epoch 6 - iter 135/275 - loss 0.04047721 - time (sec): 6.28 - samples/sec: 1811.90 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:19:36,590 epoch 6 - iter 162/275 - loss 0.04107054 - time (sec): 7.53 - samples/sec: 1820.12 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:19:37,811 epoch 6 - iter 189/275 - loss 0.04430490 - time (sec): 8.75 - samples/sec: 1798.07 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:19:39,078 epoch 6 - iter 216/275 - loss 0.04737745 - time (sec): 10.02 - samples/sec: 1787.07 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:19:40,376 epoch 6 - iter 243/275 - loss 0.04440348 - time (sec): 11.32 - samples/sec: 1771.37 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:19:41,646 epoch 6 - iter 270/275 - loss 0.04893048 - time (sec): 12.59 - samples/sec: 1771.91 - lr: 0.000013 - momentum: 0.000000
2023-10-17 08:19:41,878 ----------------------------------------------------------------------------------------------------
2023-10-17 08:19:41,878 EPOCH 6 done: loss 0.0531 - lr: 0.000013
2023-10-17 08:19:42,537 DEV : loss 0.1726471483707428 - f1-score (micro avg) 0.8558
2023-10-17 08:19:42,542 saving best model
2023-10-17 08:19:43,038 ----------------------------------------------------------------------------------------------------
2023-10-17 08:19:44,248 epoch 7 - iter 27/275 - loss 0.01218684 - time (sec): 1.20 - samples/sec: 1675.75 - lr: 0.000013 - momentum: 0.000000
2023-10-17 08:19:45,460 epoch 7 - iter 54/275 - loss 0.02183532 - time (sec): 2.42 - samples/sec: 1748.01 - lr: 0.000013 - momentum: 0.000000
2023-10-17 08:19:46,706 epoch 7 - iter 81/275 - loss 0.04935737 - time (sec): 3.66 - samples/sec: 1757.34 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:19:47,936 epoch 7 - iter 108/275 - loss 0.04925165 - time (sec): 4.89 - samples/sec: 1770.13 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:19:49,176 epoch 7 - iter 135/275 - loss 0.04722796 - time (sec): 6.13 - samples/sec: 1800.64 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:19:50,411 epoch 7 - iter 162/275 - loss 0.04194004 - time (sec): 7.37 - samples/sec: 1786.62 - lr: 0.000011 - momentum: 0.000000
2023-10-17 08:19:51,673 epoch 7 - iter 189/275 - loss 0.03932751 - time (sec): 8.63 - samples/sec: 1776.38 - lr: 0.000011 - momentum: 0.000000
2023-10-17 08:19:52,925 epoch 7 - iter 216/275 - loss 0.04019275 - time (sec): 9.88 - samples/sec: 1786.05 - lr: 0.000011 - momentum: 0.000000
2023-10-17 08:19:54,200 epoch 7 - iter 243/275 - loss 0.03985253 - time (sec): 11.16 - samples/sec: 1797.40 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:19:55,409 epoch 7 - iter 270/275 - loss 0.03718124 - time (sec): 12.36 - samples/sec: 1803.75 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:19:55,645 ----------------------------------------------------------------------------------------------------
2023-10-17 08:19:55,645 EPOCH 7 done: loss 0.0364 - lr: 0.000010
2023-10-17 08:19:56,329 DEV : loss 0.18324865400791168 - f1-score (micro avg) 0.8651
2023-10-17 08:19:56,335 saving best model
2023-10-17 08:19:56,874 ----------------------------------------------------------------------------------------------------
2023-10-17 08:19:58,102 epoch 8 - iter 27/275 - loss 0.03846923 - time (sec): 1.23 - samples/sec: 1788.12 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:19:59,403 epoch 8 - iter 54/275 - loss 0.04835504 - time (sec): 2.53 - samples/sec: 1810.51 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:20:00,655 epoch 8 - iter 81/275 - loss 0.05253058 - time (sec): 3.78 - samples/sec: 1792.79 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:20:01,914 epoch 8 - iter 108/275 - loss 0.04887803 - time (sec): 5.04 - samples/sec: 1802.31 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:20:03,235 epoch 8 - iter 135/275 - loss 0.03991749 - time (sec): 6.36 - samples/sec: 1794.32 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:20:04,504 epoch 8 - iter 162/275 - loss 0.03528465 - time (sec): 7.63 - samples/sec: 1774.58 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:20:05,743 epoch 8 - iter 189/275 - loss 0.03234484 - time (sec): 8.87 - samples/sec: 1776.78 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:20:06,988 epoch 8 - iter 216/275 - loss 0.03274941 - time (sec): 10.11 - samples/sec: 1769.21 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:20:08,253 epoch 8 - iter 243/275 - loss 0.03518058 - time (sec): 11.38 - samples/sec: 1784.29 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:20:09,546 epoch 8 - iter 270/275 - loss 0.03348453 - time (sec): 12.67 - samples/sec: 1764.84 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:20:09,775 ----------------------------------------------------------------------------------------------------
2023-10-17 08:20:09,775 EPOCH 8 done: loss 0.0329 - lr: 0.000007
2023-10-17 08:20:10,406 DEV : loss 0.18320710957050323 - f1-score (micro avg) 0.866
2023-10-17 08:20:10,411 saving best model
2023-10-17 08:20:10,971 ----------------------------------------------------------------------------------------------------
2023-10-17 08:20:12,277 epoch 9 - iter 27/275 - loss 0.03749169 - time (sec): 1.30 - samples/sec: 1642.16 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:20:13,497 epoch 9 - iter 54/275 - loss 0.04785379 - time (sec): 2.52 - samples/sec: 1651.04 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:20:14,733 epoch 9 - iter 81/275 - loss 0.03406399 - time (sec): 3.76 - samples/sec: 1711.57 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:20:16,057 epoch 9 - iter 108/275 - loss 0.02763835 - time (sec): 5.08 - samples/sec: 1726.18 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:20:17,376 epoch 9 - iter 135/275 - loss 0.02907429 - time (sec): 6.40 - samples/sec: 1717.44 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:20:18,595 epoch 9 - iter 162/275 - loss 0.02739866 - time (sec): 7.62 - samples/sec: 1768.24 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:20:19,821 epoch 9 - iter 189/275 - loss 0.02560578 - time (sec): 8.85 - samples/sec: 1769.08 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:20:21,040 epoch 9 - iter 216/275 - loss 0.02617679 - time (sec): 10.07 - samples/sec: 1753.28 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:20:22,265 epoch 9 - iter 243/275 - loss 0.02508920 - time (sec): 11.29 - samples/sec: 1749.91 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:20:23,500 epoch 9 - iter 270/275 - loss 0.02536099 - time (sec): 12.53 - samples/sec: 1776.21 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:20:23,730 ----------------------------------------------------------------------------------------------------
2023-10-17 08:20:23,730 EPOCH 9 done: loss 0.0259 - lr: 0.000003
2023-10-17 08:20:24,360 DEV : loss 0.18943718075752258 - f1-score (micro avg) 0.8685
2023-10-17 08:20:24,365 saving best model
2023-10-17 08:20:24,836 ----------------------------------------------------------------------------------------------------
2023-10-17 08:20:25,991 epoch 10 - iter 27/275 - loss 0.03037133 - time (sec): 1.15 - samples/sec: 1892.87 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:20:27,258 epoch 10 - iter 54/275 - loss 0.01674465 - time (sec): 2.42 - samples/sec: 1742.46 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:20:28,480 epoch 10 - iter 81/275 - loss 0.01931327 - time (sec): 3.64 - samples/sec: 1772.42 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:20:29,804 epoch 10 - iter 108/275 - loss 0.01642879 - time (sec): 4.96 - samples/sec: 1718.68 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:20:31,134 epoch 10 - iter 135/275 - loss 0.01363607 - time (sec): 6.29 - samples/sec: 1725.93 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:20:32,428 epoch 10 - iter 162/275 - loss 0.01853421 - time (sec): 7.59 - samples/sec: 1707.95 - lr: 0.000001 - momentum: 0.000000
2023-10-17 08:20:33,669 epoch 10 - iter 189/275 - loss 0.02043455 - time (sec): 8.83 - samples/sec: 1753.41 - lr: 0.000001 - momentum: 0.000000
2023-10-17 08:20:34,907 epoch 10 - iter 216/275 - loss 0.01890215 - time (sec): 10.07 - samples/sec: 1775.84 - lr: 0.000001 - momentum: 0.000000
2023-10-17 08:20:36,129 epoch 10 - iter 243/275 - loss 0.02207666 - time (sec): 11.29 - samples/sec: 1787.67 - lr: 0.000000 - momentum: 0.000000
2023-10-17 08:20:37,341 epoch 10 - iter 270/275 - loss 0.02111457 - time (sec): 12.50 - samples/sec: 1789.88 - lr: 0.000000 - momentum: 0.000000
2023-10-17 08:20:37,561 ----------------------------------------------------------------------------------------------------
2023-10-17 08:20:37,562 EPOCH 10 done: loss 0.0209 - lr: 0.000000
2023-10-17 08:20:38,287 DEV : loss 0.1890801340341568 - f1-score (micro avg) 0.8798
2023-10-17 08:20:38,294 saving best model
2023-10-17 08:20:39,125 ----------------------------------------------------------------------------------------------------
2023-10-17 08:20:39,126 Loading model from best epoch ...
2023-10-17 08:20:40,778 SequenceTagger predicts: Dictionary with 25 tags: O, S-scope, B-scope, E-scope, I-scope, S-pers, B-pers, E-pers, I-pers, S-work, B-work, E-work, I-work, S-loc, B-loc, E-loc, I-loc, S-object, B-object, E-object, I-object, S-date, B-date, E-date, I-date
2023-10-17 08:20:41,567
Results:
- F-score (micro) 0.9237
- F-score (macro) 0.7548
- Accuracy 0.8667
By class:
precision recall f1-score support
scope 0.9143 0.9091 0.9117 176
pers 0.9606 0.9531 0.9569 128
work 0.9054 0.9054 0.9054 74
loc 1.0000 1.0000 1.0000 2
object 0.0000 0.0000 0.0000 2
micro avg 0.9286 0.9188 0.9237 382
macro avg 0.7561 0.7535 0.7548 382
weighted avg 0.9238 0.9188 0.9213 382
2023-10-17 08:20:41,567 ----------------------------------------------------------------------------------------------------
|