File size: 24,147 Bytes
418af5b |
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 08:48:55,885 ----------------------------------------------------------------------------------------------------
2023-10-17 08:48:55,886 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 08:48:55,886 ----------------------------------------------------------------------------------------------------
2023-10-17 08:48:55,886 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 08:48:55,886 ----------------------------------------------------------------------------------------------------
2023-10-17 08:48:55,886 Train: 1100 sentences
2023-10-17 08:48:55,886 (train_with_dev=False, train_with_test=False)
2023-10-17 08:48:55,886 ----------------------------------------------------------------------------------------------------
2023-10-17 08:48:55,886 Training Params:
2023-10-17 08:48:55,886 - learning_rate: "5e-05"
2023-10-17 08:48:55,886 - mini_batch_size: "4"
2023-10-17 08:48:55,886 - max_epochs: "10"
2023-10-17 08:48:55,886 - shuffle: "True"
2023-10-17 08:48:55,886 ----------------------------------------------------------------------------------------------------
2023-10-17 08:48:55,887 Plugins:
2023-10-17 08:48:55,887 - TensorboardLogger
2023-10-17 08:48:55,887 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 08:48:55,887 ----------------------------------------------------------------------------------------------------
2023-10-17 08:48:55,887 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 08:48:55,887 - metric: "('micro avg', 'f1-score')"
2023-10-17 08:48:55,887 ----------------------------------------------------------------------------------------------------
2023-10-17 08:48:55,887 Computation:
2023-10-17 08:48:55,887 - compute on device: cuda:0
2023-10-17 08:48:55,887 - embedding storage: none
2023-10-17 08:48:55,887 ----------------------------------------------------------------------------------------------------
2023-10-17 08:48:55,887 Model training base path: "hmbench-ajmc/de-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-17 08:48:55,887 ----------------------------------------------------------------------------------------------------
2023-10-17 08:48:55,887 ----------------------------------------------------------------------------------------------------
2023-10-17 08:48:55,887 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 08:48:57,189 epoch 1 - iter 27/275 - loss 3.82248822 - time (sec): 1.30 - samples/sec: 1812.72 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:48:58,499 epoch 1 - iter 54/275 - loss 3.00028749 - time (sec): 2.61 - samples/sec: 1646.40 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:48:59,783 epoch 1 - iter 81/275 - loss 2.32531297 - time (sec): 3.90 - samples/sec: 1712.74 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:49:01,016 epoch 1 - iter 108/275 - loss 1.90971994 - time (sec): 5.13 - samples/sec: 1705.45 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:49:02,232 epoch 1 - iter 135/275 - loss 1.61165235 - time (sec): 6.34 - samples/sec: 1723.79 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:49:03,455 epoch 1 - iter 162/275 - loss 1.39866068 - time (sec): 7.57 - samples/sec: 1746.50 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:49:04,648 epoch 1 - iter 189/275 - loss 1.23507602 - time (sec): 8.76 - samples/sec: 1795.24 - lr: 0.000034 - momentum: 0.000000
2023-10-17 08:49:05,818 epoch 1 - iter 216/275 - loss 1.10155245 - time (sec): 9.93 - samples/sec: 1837.18 - lr: 0.000039 - momentum: 0.000000
2023-10-17 08:49:06,970 epoch 1 - iter 243/275 - loss 1.02083629 - time (sec): 11.08 - samples/sec: 1819.41 - lr: 0.000044 - momentum: 0.000000
2023-10-17 08:49:08,216 epoch 1 - iter 270/275 - loss 0.94422994 - time (sec): 12.33 - samples/sec: 1813.98 - lr: 0.000049 - momentum: 0.000000
2023-10-17 08:49:08,435 ----------------------------------------------------------------------------------------------------
2023-10-17 08:49:08,435 EPOCH 1 done: loss 0.9301 - lr: 0.000049
2023-10-17 08:49:09,128 DEV : loss 0.18998880684375763 - f1-score (micro avg) 0.7718
2023-10-17 08:49:09,133 saving best model
2023-10-17 08:49:09,453 ----------------------------------------------------------------------------------------------------
2023-10-17 08:49:10,646 epoch 2 - iter 27/275 - loss 0.19771864 - time (sec): 1.19 - samples/sec: 1897.13 - lr: 0.000049 - momentum: 0.000000
2023-10-17 08:49:11,838 epoch 2 - iter 54/275 - loss 0.18078864 - time (sec): 2.38 - samples/sec: 1879.34 - lr: 0.000049 - momentum: 0.000000
2023-10-17 08:49:13,069 epoch 2 - iter 81/275 - loss 0.16888107 - time (sec): 3.61 - samples/sec: 1844.34 - lr: 0.000048 - momentum: 0.000000
2023-10-17 08:49:14,296 epoch 2 - iter 108/275 - loss 0.17030036 - time (sec): 4.84 - samples/sec: 1868.12 - lr: 0.000048 - momentum: 0.000000
2023-10-17 08:49:15,500 epoch 2 - iter 135/275 - loss 0.16720867 - time (sec): 6.05 - samples/sec: 1855.67 - lr: 0.000047 - momentum: 0.000000
2023-10-17 08:49:16,715 epoch 2 - iter 162/275 - loss 0.16947469 - time (sec): 7.26 - samples/sec: 1863.99 - lr: 0.000047 - momentum: 0.000000
2023-10-17 08:49:17,929 epoch 2 - iter 189/275 - loss 0.17047367 - time (sec): 8.47 - samples/sec: 1859.20 - lr: 0.000046 - momentum: 0.000000
2023-10-17 08:49:19,167 epoch 2 - iter 216/275 - loss 0.17531921 - time (sec): 9.71 - samples/sec: 1888.06 - lr: 0.000046 - momentum: 0.000000
2023-10-17 08:49:20,375 epoch 2 - iter 243/275 - loss 0.17237634 - time (sec): 10.92 - samples/sec: 1877.41 - lr: 0.000045 - momentum: 0.000000
2023-10-17 08:49:21,602 epoch 2 - iter 270/275 - loss 0.16957074 - time (sec): 12.15 - samples/sec: 1848.01 - lr: 0.000045 - momentum: 0.000000
2023-10-17 08:49:21,825 ----------------------------------------------------------------------------------------------------
2023-10-17 08:49:21,825 EPOCH 2 done: loss 0.1709 - lr: 0.000045
2023-10-17 08:49:22,458 DEV : loss 0.18550686538219452 - f1-score (micro avg) 0.824
2023-10-17 08:49:22,462 saving best model
2023-10-17 08:49:22,891 ----------------------------------------------------------------------------------------------------
2023-10-17 08:49:24,207 epoch 3 - iter 27/275 - loss 0.12270099 - time (sec): 1.31 - samples/sec: 1606.52 - lr: 0.000044 - momentum: 0.000000
2023-10-17 08:49:25,506 epoch 3 - iter 54/275 - loss 0.09803233 - time (sec): 2.61 - samples/sec: 1509.06 - lr: 0.000043 - momentum: 0.000000
2023-10-17 08:49:26,764 epoch 3 - iter 81/275 - loss 0.09418740 - time (sec): 3.87 - samples/sec: 1601.17 - lr: 0.000043 - momentum: 0.000000
2023-10-17 08:49:27,994 epoch 3 - iter 108/275 - loss 0.10543827 - time (sec): 5.10 - samples/sec: 1699.99 - lr: 0.000042 - momentum: 0.000000
2023-10-17 08:49:29,225 epoch 3 - iter 135/275 - loss 0.10796688 - time (sec): 6.33 - samples/sec: 1699.71 - lr: 0.000042 - momentum: 0.000000
2023-10-17 08:49:30,466 epoch 3 - iter 162/275 - loss 0.10881893 - time (sec): 7.57 - samples/sec: 1728.72 - lr: 0.000041 - momentum: 0.000000
2023-10-17 08:49:31,684 epoch 3 - iter 189/275 - loss 0.11070236 - time (sec): 8.79 - samples/sec: 1734.97 - lr: 0.000041 - momentum: 0.000000
2023-10-17 08:49:32,894 epoch 3 - iter 216/275 - loss 0.10809719 - time (sec): 10.00 - samples/sec: 1755.49 - lr: 0.000040 - momentum: 0.000000
2023-10-17 08:49:34,147 epoch 3 - iter 243/275 - loss 0.10810593 - time (sec): 11.25 - samples/sec: 1771.20 - lr: 0.000040 - momentum: 0.000000
2023-10-17 08:49:35,385 epoch 3 - iter 270/275 - loss 0.11303403 - time (sec): 12.49 - samples/sec: 1784.32 - lr: 0.000039 - momentum: 0.000000
2023-10-17 08:49:35,619 ----------------------------------------------------------------------------------------------------
2023-10-17 08:49:35,619 EPOCH 3 done: loss 0.1122 - lr: 0.000039
2023-10-17 08:49:36,325 DEV : loss 0.16965429484844208 - f1-score (micro avg) 0.8448
2023-10-17 08:49:36,329 saving best model
2023-10-17 08:49:36,790 ----------------------------------------------------------------------------------------------------
2023-10-17 08:49:38,021 epoch 4 - iter 27/275 - loss 0.06300598 - time (sec): 1.23 - samples/sec: 1727.72 - lr: 0.000038 - momentum: 0.000000
2023-10-17 08:49:39,256 epoch 4 - iter 54/275 - loss 0.05146376 - time (sec): 2.46 - samples/sec: 1758.25 - lr: 0.000038 - momentum: 0.000000
2023-10-17 08:49:40,479 epoch 4 - iter 81/275 - loss 0.09438839 - time (sec): 3.69 - samples/sec: 1757.01 - lr: 0.000037 - momentum: 0.000000
2023-10-17 08:49:41,724 epoch 4 - iter 108/275 - loss 0.09296037 - time (sec): 4.93 - samples/sec: 1769.01 - lr: 0.000037 - momentum: 0.000000
2023-10-17 08:49:42,952 epoch 4 - iter 135/275 - loss 0.09238656 - time (sec): 6.16 - samples/sec: 1761.09 - lr: 0.000036 - momentum: 0.000000
2023-10-17 08:49:44,188 epoch 4 - iter 162/275 - loss 0.09545995 - time (sec): 7.40 - samples/sec: 1796.30 - lr: 0.000036 - momentum: 0.000000
2023-10-17 08:49:45,411 epoch 4 - iter 189/275 - loss 0.09007918 - time (sec): 8.62 - samples/sec: 1787.86 - lr: 0.000035 - momentum: 0.000000
2023-10-17 08:49:46,643 epoch 4 - iter 216/275 - loss 0.08952483 - time (sec): 9.85 - samples/sec: 1815.73 - lr: 0.000035 - momentum: 0.000000
2023-10-17 08:49:47,879 epoch 4 - iter 243/275 - loss 0.08857733 - time (sec): 11.09 - samples/sec: 1813.51 - lr: 0.000034 - momentum: 0.000000
2023-10-17 08:49:49,136 epoch 4 - iter 270/275 - loss 0.09220419 - time (sec): 12.34 - samples/sec: 1804.50 - lr: 0.000034 - momentum: 0.000000
2023-10-17 08:49:49,356 ----------------------------------------------------------------------------------------------------
2023-10-17 08:49:49,356 EPOCH 4 done: loss 0.0907 - lr: 0.000034
2023-10-17 08:49:49,995 DEV : loss 0.1864178627729416 - f1-score (micro avg) 0.8524
2023-10-17 08:49:50,000 saving best model
2023-10-17 08:49:50,426 ----------------------------------------------------------------------------------------------------
2023-10-17 08:49:51,685 epoch 5 - iter 27/275 - loss 0.12790019 - time (sec): 1.26 - samples/sec: 1934.72 - lr: 0.000033 - momentum: 0.000000
2023-10-17 08:49:52,890 epoch 5 - iter 54/275 - loss 0.09482667 - time (sec): 2.46 - samples/sec: 1865.38 - lr: 0.000032 - momentum: 0.000000
2023-10-17 08:49:54,116 epoch 5 - iter 81/275 - loss 0.08039190 - time (sec): 3.69 - samples/sec: 1794.41 - lr: 0.000032 - momentum: 0.000000
2023-10-17 08:49:55,316 epoch 5 - iter 108/275 - loss 0.08969888 - time (sec): 4.89 - samples/sec: 1755.60 - lr: 0.000031 - momentum: 0.000000
2023-10-17 08:49:56,531 epoch 5 - iter 135/275 - loss 0.08769977 - time (sec): 6.10 - samples/sec: 1784.72 - lr: 0.000031 - momentum: 0.000000
2023-10-17 08:49:57,760 epoch 5 - iter 162/275 - loss 0.07990395 - time (sec): 7.33 - samples/sec: 1791.48 - lr: 0.000030 - momentum: 0.000000
2023-10-17 08:49:58,991 epoch 5 - iter 189/275 - loss 0.07643210 - time (sec): 8.56 - samples/sec: 1804.80 - lr: 0.000030 - momentum: 0.000000
2023-10-17 08:50:00,220 epoch 5 - iter 216/275 - loss 0.07245098 - time (sec): 9.79 - samples/sec: 1830.89 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:50:01,431 epoch 5 - iter 243/275 - loss 0.06950953 - time (sec): 11.00 - samples/sec: 1830.96 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:50:02,649 epoch 5 - iter 270/275 - loss 0.06779922 - time (sec): 12.22 - samples/sec: 1832.61 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:50:02,877 ----------------------------------------------------------------------------------------------------
2023-10-17 08:50:02,878 EPOCH 5 done: loss 0.0669 - lr: 0.000028
2023-10-17 08:50:03,511 DEV : loss 0.16903042793273926 - f1-score (micro avg) 0.8663
2023-10-17 08:50:03,516 saving best model
2023-10-17 08:50:03,970 ----------------------------------------------------------------------------------------------------
2023-10-17 08:50:05,172 epoch 6 - iter 27/275 - loss 0.05910721 - time (sec): 1.20 - samples/sec: 1666.81 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:50:06,397 epoch 6 - iter 54/275 - loss 0.06915115 - time (sec): 2.42 - samples/sec: 1731.26 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:50:07,625 epoch 6 - iter 81/275 - loss 0.06797154 - time (sec): 3.65 - samples/sec: 1780.83 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:50:08,844 epoch 6 - iter 108/275 - loss 0.06531264 - time (sec): 4.87 - samples/sec: 1774.06 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:50:10,056 epoch 6 - iter 135/275 - loss 0.05964926 - time (sec): 6.08 - samples/sec: 1797.51 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:50:11,287 epoch 6 - iter 162/275 - loss 0.05173399 - time (sec): 7.31 - samples/sec: 1794.59 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:50:12,522 epoch 6 - iter 189/275 - loss 0.05150710 - time (sec): 8.55 - samples/sec: 1794.17 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:50:13,760 epoch 6 - iter 216/275 - loss 0.05236797 - time (sec): 9.79 - samples/sec: 1799.74 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:50:15,008 epoch 6 - iter 243/275 - loss 0.05636757 - time (sec): 11.04 - samples/sec: 1819.67 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:50:16,240 epoch 6 - iter 270/275 - loss 0.05370161 - time (sec): 12.27 - samples/sec: 1816.80 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:50:16,473 ----------------------------------------------------------------------------------------------------
2023-10-17 08:50:16,473 EPOCH 6 done: loss 0.0527 - lr: 0.000022
2023-10-17 08:50:17,163 DEV : loss 0.19920983910560608 - f1-score (micro avg) 0.8738
2023-10-17 08:50:17,168 saving best model
2023-10-17 08:50:17,614 ----------------------------------------------------------------------------------------------------
2023-10-17 08:50:18,885 epoch 7 - iter 27/275 - loss 0.06705101 - time (sec): 1.27 - samples/sec: 1625.34 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:50:20,102 epoch 7 - iter 54/275 - loss 0.05309652 - time (sec): 2.48 - samples/sec: 1775.45 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:50:21,360 epoch 7 - iter 81/275 - loss 0.06157296 - time (sec): 3.74 - samples/sec: 1789.79 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:50:22,608 epoch 7 - iter 108/275 - loss 0.05165244 - time (sec): 4.99 - samples/sec: 1789.51 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:50:23,836 epoch 7 - iter 135/275 - loss 0.04265718 - time (sec): 6.22 - samples/sec: 1776.49 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:50:25,066 epoch 7 - iter 162/275 - loss 0.03858431 - time (sec): 7.45 - samples/sec: 1798.34 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:50:26,303 epoch 7 - iter 189/275 - loss 0.03652655 - time (sec): 8.69 - samples/sec: 1826.50 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:50:27,502 epoch 7 - iter 216/275 - loss 0.03525941 - time (sec): 9.89 - samples/sec: 1818.61 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:50:28,704 epoch 7 - iter 243/275 - loss 0.03571872 - time (sec): 11.09 - samples/sec: 1807.42 - lr: 0.000017 - momentum: 0.000000
2023-10-17 08:50:29,908 epoch 7 - iter 270/275 - loss 0.03739856 - time (sec): 12.29 - samples/sec: 1817.32 - lr: 0.000017 - momentum: 0.000000
2023-10-17 08:50:30,140 ----------------------------------------------------------------------------------------------------
2023-10-17 08:50:30,140 EPOCH 7 done: loss 0.0367 - lr: 0.000017
2023-10-17 08:50:30,838 DEV : loss 0.18746249377727509 - f1-score (micro avg) 0.8761
2023-10-17 08:50:30,843 saving best model
2023-10-17 08:50:31,282 ----------------------------------------------------------------------------------------------------
2023-10-17 08:50:32,564 epoch 8 - iter 27/275 - loss 0.00694663 - time (sec): 1.28 - samples/sec: 1887.76 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:50:33,784 epoch 8 - iter 54/275 - loss 0.01940040 - time (sec): 2.50 - samples/sec: 1857.57 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:50:34,995 epoch 8 - iter 81/275 - loss 0.01675556 - time (sec): 3.71 - samples/sec: 1850.34 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:50:36,228 epoch 8 - iter 108/275 - loss 0.01499351 - time (sec): 4.94 - samples/sec: 1821.51 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:50:37,467 epoch 8 - iter 135/275 - loss 0.01379214 - time (sec): 6.18 - samples/sec: 1870.49 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:50:38,686 epoch 8 - iter 162/275 - loss 0.01536753 - time (sec): 7.40 - samples/sec: 1842.69 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:50:39,932 epoch 8 - iter 189/275 - loss 0.01547651 - time (sec): 8.65 - samples/sec: 1840.68 - lr: 0.000013 - momentum: 0.000000
2023-10-17 08:50:41,149 epoch 8 - iter 216/275 - loss 0.02440394 - time (sec): 9.86 - samples/sec: 1835.57 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:50:42,403 epoch 8 - iter 243/275 - loss 0.02452509 - time (sec): 11.12 - samples/sec: 1827.30 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:50:43,610 epoch 8 - iter 270/275 - loss 0.02315233 - time (sec): 12.33 - samples/sec: 1819.39 - lr: 0.000011 - momentum: 0.000000
2023-10-17 08:50:43,845 ----------------------------------------------------------------------------------------------------
2023-10-17 08:50:43,845 EPOCH 8 done: loss 0.0236 - lr: 0.000011
2023-10-17 08:50:44,503 DEV : loss 0.18050651252269745 - f1-score (micro avg) 0.8824
2023-10-17 08:50:44,507 saving best model
2023-10-17 08:50:44,952 ----------------------------------------------------------------------------------------------------
2023-10-17 08:50:46,212 epoch 9 - iter 27/275 - loss 0.02741986 - time (sec): 1.26 - samples/sec: 1900.36 - lr: 0.000011 - momentum: 0.000000
2023-10-17 08:50:47,442 epoch 9 - iter 54/275 - loss 0.02099426 - time (sec): 2.49 - samples/sec: 1934.93 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:50:48,640 epoch 9 - iter 81/275 - loss 0.01543202 - time (sec): 3.68 - samples/sec: 1866.88 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:50:49,860 epoch 9 - iter 108/275 - loss 0.01638367 - time (sec): 4.90 - samples/sec: 1823.06 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:50:51,188 epoch 9 - iter 135/275 - loss 0.01630896 - time (sec): 6.23 - samples/sec: 1762.81 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:50:52,422 epoch 9 - iter 162/275 - loss 0.02265358 - time (sec): 7.47 - samples/sec: 1777.55 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:50:53,675 epoch 9 - iter 189/275 - loss 0.02334752 - time (sec): 8.72 - samples/sec: 1778.80 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:50:54,914 epoch 9 - iter 216/275 - loss 0.02140354 - time (sec): 9.96 - samples/sec: 1769.99 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:50:56,137 epoch 9 - iter 243/275 - loss 0.02042862 - time (sec): 11.18 - samples/sec: 1781.86 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:50:57,400 epoch 9 - iter 270/275 - loss 0.02027750 - time (sec): 12.44 - samples/sec: 1790.44 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:50:57,651 ----------------------------------------------------------------------------------------------------
2023-10-17 08:50:57,651 EPOCH 9 done: loss 0.0200 - lr: 0.000006
2023-10-17 08:50:58,324 DEV : loss 0.18982398509979248 - f1-score (micro avg) 0.8806
2023-10-17 08:50:58,329 ----------------------------------------------------------------------------------------------------
2023-10-17 08:50:59,559 epoch 10 - iter 27/275 - loss 0.00426859 - time (sec): 1.23 - samples/sec: 1738.20 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:51:00,828 epoch 10 - iter 54/275 - loss 0.00929609 - time (sec): 2.50 - samples/sec: 1823.60 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:51:02,068 epoch 10 - iter 81/275 - loss 0.01188040 - time (sec): 3.74 - samples/sec: 1806.78 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:51:03,343 epoch 10 - iter 108/275 - loss 0.00961700 - time (sec): 5.01 - samples/sec: 1725.88 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:51:04,569 epoch 10 - iter 135/275 - loss 0.00945368 - time (sec): 6.24 - samples/sec: 1738.12 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:51:05,792 epoch 10 - iter 162/275 - loss 0.01019720 - time (sec): 7.46 - samples/sec: 1749.50 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:51:07,080 epoch 10 - iter 189/275 - loss 0.00913857 - time (sec): 8.75 - samples/sec: 1773.33 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:51:08,377 epoch 10 - iter 216/275 - loss 0.01442529 - time (sec): 10.05 - samples/sec: 1774.17 - lr: 0.000001 - momentum: 0.000000
2023-10-17 08:51:09,613 epoch 10 - iter 243/275 - loss 0.01322089 - time (sec): 11.28 - samples/sec: 1784.49 - lr: 0.000001 - momentum: 0.000000
2023-10-17 08:51:10,831 epoch 10 - iter 270/275 - loss 0.01241629 - time (sec): 12.50 - samples/sec: 1787.20 - lr: 0.000000 - momentum: 0.000000
2023-10-17 08:51:11,066 ----------------------------------------------------------------------------------------------------
2023-10-17 08:51:11,066 EPOCH 10 done: loss 0.0122 - lr: 0.000000
2023-10-17 08:51:11,702 DEV : loss 0.19630618393421173 - f1-score (micro avg) 0.8854
2023-10-17 08:51:11,707 saving best model
2023-10-17 08:51:12,484 ----------------------------------------------------------------------------------------------------
2023-10-17 08:51:12,485 Loading model from best epoch ...
2023-10-17 08:51:13,852 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 08:51:14,649
Results:
- F-score (micro) 0.9048
- F-score (macro) 0.74
- Accuracy 0.8341
By class:
precision recall f1-score support
scope 0.8947 0.8693 0.8818 176
pers 0.9843 0.9766 0.9804 128
work 0.8378 0.8378 0.8378 74
object 1.0000 1.0000 1.0000 2
loc 0.0000 0.0000 0.0000 2
micro avg 0.9144 0.8953 0.9048 382
macro avg 0.7434 0.7367 0.7400 382
weighted avg 0.9096 0.8953 0.9023 382
2023-10-17 08:51:14,649 ----------------------------------------------------------------------------------------------------
|