File size: 24,065 Bytes
f57246c |
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 |
2023-10-17 08:58:17,407 ----------------------------------------------------------------------------------------------------
2023-10-17 08:58:17,408 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 08:58:17,408 ----------------------------------------------------------------------------------------------------
2023-10-17 08:58:17,408 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 08:58:17,408 ----------------------------------------------------------------------------------------------------
2023-10-17 08:58:17,408 Train: 1100 sentences
2023-10-17 08:58:17,408 (train_with_dev=False, train_with_test=False)
2023-10-17 08:58:17,408 ----------------------------------------------------------------------------------------------------
2023-10-17 08:58:17,408 Training Params:
2023-10-17 08:58:17,408 - learning_rate: "5e-05"
2023-10-17 08:58:17,408 - mini_batch_size: "4"
2023-10-17 08:58:17,408 - max_epochs: "10"
2023-10-17 08:58:17,408 - shuffle: "True"
2023-10-17 08:58:17,408 ----------------------------------------------------------------------------------------------------
2023-10-17 08:58:17,408 Plugins:
2023-10-17 08:58:17,409 - TensorboardLogger
2023-10-17 08:58:17,409 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 08:58:17,409 ----------------------------------------------------------------------------------------------------
2023-10-17 08:58:17,409 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 08:58:17,409 - metric: "('micro avg', 'f1-score')"
2023-10-17 08:58:17,409 ----------------------------------------------------------------------------------------------------
2023-10-17 08:58:17,409 Computation:
2023-10-17 08:58:17,409 - compute on device: cuda:0
2023-10-17 08:58:17,409 - embedding storage: none
2023-10-17 08:58:17,409 ----------------------------------------------------------------------------------------------------
2023-10-17 08:58:17,409 Model training base path: "hmbench-ajmc/de-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-5"
2023-10-17 08:58:17,409 ----------------------------------------------------------------------------------------------------
2023-10-17 08:58:17,409 ----------------------------------------------------------------------------------------------------
2023-10-17 08:58:17,409 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 08:58:18,605 epoch 1 - iter 27/275 - loss 4.07866641 - time (sec): 1.19 - samples/sec: 1976.27 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:58:19,775 epoch 1 - iter 54/275 - loss 3.29122762 - time (sec): 2.37 - samples/sec: 1984.04 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:58:20,926 epoch 1 - iter 81/275 - loss 2.54691304 - time (sec): 3.52 - samples/sec: 1958.01 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:58:22,121 epoch 1 - iter 108/275 - loss 2.12670496 - time (sec): 4.71 - samples/sec: 1912.47 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:58:23,343 epoch 1 - iter 135/275 - loss 1.79077335 - time (sec): 5.93 - samples/sec: 1917.19 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:58:24,549 epoch 1 - iter 162/275 - loss 1.55375745 - time (sec): 7.14 - samples/sec: 1907.76 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:58:25,780 epoch 1 - iter 189/275 - loss 1.37607396 - time (sec): 8.37 - samples/sec: 1919.35 - lr: 0.000034 - momentum: 0.000000
2023-10-17 08:58:27,007 epoch 1 - iter 216/275 - loss 1.24831931 - time (sec): 9.60 - samples/sec: 1917.67 - lr: 0.000039 - momentum: 0.000000
2023-10-17 08:58:28,229 epoch 1 - iter 243/275 - loss 1.15443524 - time (sec): 10.82 - samples/sec: 1879.47 - lr: 0.000044 - momentum: 0.000000
2023-10-17 08:58:29,450 epoch 1 - iter 270/275 - loss 1.07150887 - time (sec): 12.04 - samples/sec: 1855.75 - lr: 0.000049 - momentum: 0.000000
2023-10-17 08:58:29,669 ----------------------------------------------------------------------------------------------------
2023-10-17 08:58:29,669 EPOCH 1 done: loss 1.0540 - lr: 0.000049
2023-10-17 08:58:30,219 DEV : loss 0.21305261552333832 - f1-score (micro avg) 0.689
2023-10-17 08:58:30,225 saving best model
2023-10-17 08:58:30,589 ----------------------------------------------------------------------------------------------------
2023-10-17 08:58:31,803 epoch 2 - iter 27/275 - loss 0.17379043 - time (sec): 1.21 - samples/sec: 1836.45 - lr: 0.000049 - momentum: 0.000000
2023-10-17 08:58:33,023 epoch 2 - iter 54/275 - loss 0.20261401 - time (sec): 2.43 - samples/sec: 1731.08 - lr: 0.000049 - momentum: 0.000000
2023-10-17 08:58:34,261 epoch 2 - iter 81/275 - loss 0.18846942 - time (sec): 3.67 - samples/sec: 1771.69 - lr: 0.000048 - momentum: 0.000000
2023-10-17 08:58:35,484 epoch 2 - iter 108/275 - loss 0.20079845 - time (sec): 4.89 - samples/sec: 1774.39 - lr: 0.000048 - momentum: 0.000000
2023-10-17 08:58:36,710 epoch 2 - iter 135/275 - loss 0.19197758 - time (sec): 6.12 - samples/sec: 1762.60 - lr: 0.000047 - momentum: 0.000000
2023-10-17 08:58:37,914 epoch 2 - iter 162/275 - loss 0.18863668 - time (sec): 7.32 - samples/sec: 1773.15 - lr: 0.000047 - momentum: 0.000000
2023-10-17 08:58:39,138 epoch 2 - iter 189/275 - loss 0.18366584 - time (sec): 8.55 - samples/sec: 1781.81 - lr: 0.000046 - momentum: 0.000000
2023-10-17 08:58:40,371 epoch 2 - iter 216/275 - loss 0.17800713 - time (sec): 9.78 - samples/sec: 1797.98 - lr: 0.000046 - momentum: 0.000000
2023-10-17 08:58:41,611 epoch 2 - iter 243/275 - loss 0.17153577 - time (sec): 11.02 - samples/sec: 1811.35 - lr: 0.000045 - momentum: 0.000000
2023-10-17 08:58:42,818 epoch 2 - iter 270/275 - loss 0.17760855 - time (sec): 12.23 - samples/sec: 1826.07 - lr: 0.000045 - momentum: 0.000000
2023-10-17 08:58:43,040 ----------------------------------------------------------------------------------------------------
2023-10-17 08:58:43,040 EPOCH 2 done: loss 0.1758 - lr: 0.000045
2023-10-17 08:58:43,808 DEV : loss 0.18843930959701538 - f1-score (micro avg) 0.7895
2023-10-17 08:58:43,815 saving best model
2023-10-17 08:58:44,366 ----------------------------------------------------------------------------------------------------
2023-10-17 08:58:45,831 epoch 3 - iter 27/275 - loss 0.11796556 - time (sec): 1.46 - samples/sec: 1664.07 - lr: 0.000044 - momentum: 0.000000
2023-10-17 08:58:47,270 epoch 3 - iter 54/275 - loss 0.11204503 - time (sec): 2.90 - samples/sec: 1569.24 - lr: 0.000043 - momentum: 0.000000
2023-10-17 08:58:48,521 epoch 3 - iter 81/275 - loss 0.10541729 - time (sec): 4.15 - samples/sec: 1678.87 - lr: 0.000043 - momentum: 0.000000
2023-10-17 08:58:49,769 epoch 3 - iter 108/275 - loss 0.09944435 - time (sec): 5.40 - samples/sec: 1682.28 - lr: 0.000042 - momentum: 0.000000
2023-10-17 08:58:51,020 epoch 3 - iter 135/275 - loss 0.08997580 - time (sec): 6.65 - samples/sec: 1716.47 - lr: 0.000042 - momentum: 0.000000
2023-10-17 08:58:52,209 epoch 3 - iter 162/275 - loss 0.09204905 - time (sec): 7.84 - samples/sec: 1716.14 - lr: 0.000041 - momentum: 0.000000
2023-10-17 08:58:53,645 epoch 3 - iter 189/275 - loss 0.09910161 - time (sec): 9.28 - samples/sec: 1684.31 - lr: 0.000041 - momentum: 0.000000
2023-10-17 08:58:54,860 epoch 3 - iter 216/275 - loss 0.10368806 - time (sec): 10.49 - samples/sec: 1711.55 - lr: 0.000040 - momentum: 0.000000
2023-10-17 08:58:56,133 epoch 3 - iter 243/275 - loss 0.10108127 - time (sec): 11.77 - samples/sec: 1727.07 - lr: 0.000040 - momentum: 0.000000
2023-10-17 08:58:57,332 epoch 3 - iter 270/275 - loss 0.10499294 - time (sec): 12.96 - samples/sec: 1727.39 - lr: 0.000039 - momentum: 0.000000
2023-10-17 08:58:57,564 ----------------------------------------------------------------------------------------------------
2023-10-17 08:58:57,564 EPOCH 3 done: loss 0.1036 - lr: 0.000039
2023-10-17 08:58:58,201 DEV : loss 0.17471981048583984 - f1-score (micro avg) 0.8381
2023-10-17 08:58:58,206 saving best model
2023-10-17 08:58:58,662 ----------------------------------------------------------------------------------------------------
2023-10-17 08:58:59,945 epoch 4 - iter 27/275 - loss 0.06961920 - time (sec): 1.28 - samples/sec: 1517.71 - lr: 0.000038 - momentum: 0.000000
2023-10-17 08:59:01,182 epoch 4 - iter 54/275 - loss 0.05928687 - time (sec): 2.52 - samples/sec: 1617.38 - lr: 0.000038 - momentum: 0.000000
2023-10-17 08:59:02,416 epoch 4 - iter 81/275 - loss 0.07798742 - time (sec): 3.75 - samples/sec: 1739.45 - lr: 0.000037 - momentum: 0.000000
2023-10-17 08:59:03,637 epoch 4 - iter 108/275 - loss 0.07703170 - time (sec): 4.97 - samples/sec: 1778.57 - lr: 0.000037 - momentum: 0.000000
2023-10-17 08:59:04,939 epoch 4 - iter 135/275 - loss 0.06995502 - time (sec): 6.27 - samples/sec: 1798.39 - lr: 0.000036 - momentum: 0.000000
2023-10-17 08:59:06,139 epoch 4 - iter 162/275 - loss 0.07538902 - time (sec): 7.47 - samples/sec: 1800.10 - lr: 0.000036 - momentum: 0.000000
2023-10-17 08:59:07,403 epoch 4 - iter 189/275 - loss 0.09041096 - time (sec): 8.74 - samples/sec: 1797.81 - lr: 0.000035 - momentum: 0.000000
2023-10-17 08:59:08,633 epoch 4 - iter 216/275 - loss 0.08712686 - time (sec): 9.97 - samples/sec: 1824.14 - lr: 0.000035 - momentum: 0.000000
2023-10-17 08:59:09,874 epoch 4 - iter 243/275 - loss 0.08979277 - time (sec): 11.21 - samples/sec: 1831.00 - lr: 0.000034 - momentum: 0.000000
2023-10-17 08:59:11,099 epoch 4 - iter 270/275 - loss 0.08447754 - time (sec): 12.43 - samples/sec: 1803.26 - lr: 0.000034 - momentum: 0.000000
2023-10-17 08:59:11,324 ----------------------------------------------------------------------------------------------------
2023-10-17 08:59:11,324 EPOCH 4 done: loss 0.0854 - lr: 0.000034
2023-10-17 08:59:11,991 DEV : loss 0.20809108018875122 - f1-score (micro avg) 0.8396
2023-10-17 08:59:11,998 saving best model
2023-10-17 08:59:12,432 ----------------------------------------------------------------------------------------------------
2023-10-17 08:59:13,700 epoch 5 - iter 27/275 - loss 0.04462372 - time (sec): 1.27 - samples/sec: 1776.94 - lr: 0.000033 - momentum: 0.000000
2023-10-17 08:59:14,965 epoch 5 - iter 54/275 - loss 0.09665540 - time (sec): 2.53 - samples/sec: 1809.57 - lr: 0.000032 - momentum: 0.000000
2023-10-17 08:59:16,195 epoch 5 - iter 81/275 - loss 0.08000189 - time (sec): 3.76 - samples/sec: 1901.41 - lr: 0.000032 - momentum: 0.000000
2023-10-17 08:59:17,441 epoch 5 - iter 108/275 - loss 0.06849516 - time (sec): 5.01 - samples/sec: 1890.98 - lr: 0.000031 - momentum: 0.000000
2023-10-17 08:59:18,647 epoch 5 - iter 135/275 - loss 0.06277789 - time (sec): 6.21 - samples/sec: 1875.02 - lr: 0.000031 - momentum: 0.000000
2023-10-17 08:59:19,862 epoch 5 - iter 162/275 - loss 0.06311559 - time (sec): 7.43 - samples/sec: 1871.34 - lr: 0.000030 - momentum: 0.000000
2023-10-17 08:59:21,067 epoch 5 - iter 189/275 - loss 0.06254311 - time (sec): 8.63 - samples/sec: 1855.50 - lr: 0.000030 - momentum: 0.000000
2023-10-17 08:59:22,325 epoch 5 - iter 216/275 - loss 0.06440827 - time (sec): 9.89 - samples/sec: 1834.23 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:59:23,541 epoch 5 - iter 243/275 - loss 0.06533553 - time (sec): 11.11 - samples/sec: 1812.20 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:59:24,772 epoch 5 - iter 270/275 - loss 0.06177798 - time (sec): 12.34 - samples/sec: 1807.63 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:59:25,003 ----------------------------------------------------------------------------------------------------
2023-10-17 08:59:25,003 EPOCH 5 done: loss 0.0605 - lr: 0.000028
2023-10-17 08:59:25,661 DEV : loss 0.16310758888721466 - f1-score (micro avg) 0.8638
2023-10-17 08:59:25,666 saving best model
2023-10-17 08:59:26,106 ----------------------------------------------------------------------------------------------------
2023-10-17 08:59:27,381 epoch 6 - iter 27/275 - loss 0.05938802 - time (sec): 1.27 - samples/sec: 1761.49 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:59:28,605 epoch 6 - iter 54/275 - loss 0.03521374 - time (sec): 2.49 - samples/sec: 1880.76 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:59:29,907 epoch 6 - iter 81/275 - loss 0.04884466 - time (sec): 3.80 - samples/sec: 1829.45 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:59:31,160 epoch 6 - iter 108/275 - loss 0.05158786 - time (sec): 5.05 - samples/sec: 1783.64 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:59:32,408 epoch 6 - iter 135/275 - loss 0.04511006 - time (sec): 6.30 - samples/sec: 1770.43 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:59:33,652 epoch 6 - iter 162/275 - loss 0.04555781 - time (sec): 7.54 - samples/sec: 1788.09 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:59:34,877 epoch 6 - iter 189/275 - loss 0.04640001 - time (sec): 8.77 - samples/sec: 1811.19 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:59:36,129 epoch 6 - iter 216/275 - loss 0.04597427 - time (sec): 10.02 - samples/sec: 1807.09 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:59:37,371 epoch 6 - iter 243/275 - loss 0.04373227 - time (sec): 11.26 - samples/sec: 1793.75 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:59:38,625 epoch 6 - iter 270/275 - loss 0.04377672 - time (sec): 12.51 - samples/sec: 1790.20 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:59:38,866 ----------------------------------------------------------------------------------------------------
2023-10-17 08:59:38,866 EPOCH 6 done: loss 0.0436 - lr: 0.000022
2023-10-17 08:59:39,503 DEV : loss 0.18482080101966858 - f1-score (micro avg) 0.8667
2023-10-17 08:59:39,507 saving best model
2023-10-17 08:59:39,942 ----------------------------------------------------------------------------------------------------
2023-10-17 08:59:41,185 epoch 7 - iter 27/275 - loss 0.00375937 - time (sec): 1.24 - samples/sec: 1787.06 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:59:42,401 epoch 7 - iter 54/275 - loss 0.03130435 - time (sec): 2.46 - samples/sec: 1792.81 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:59:43,658 epoch 7 - iter 81/275 - loss 0.02778327 - time (sec): 3.71 - samples/sec: 1813.60 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:59:44,936 epoch 7 - iter 108/275 - loss 0.02223742 - time (sec): 4.99 - samples/sec: 1784.16 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:59:46,108 epoch 7 - iter 135/275 - loss 0.02443230 - time (sec): 6.16 - samples/sec: 1804.63 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:59:47,286 epoch 7 - iter 162/275 - loss 0.02403903 - time (sec): 7.34 - samples/sec: 1819.87 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:59:48,489 epoch 7 - iter 189/275 - loss 0.02872301 - time (sec): 8.54 - samples/sec: 1812.79 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:59:49,711 epoch 7 - iter 216/275 - loss 0.02738304 - time (sec): 9.77 - samples/sec: 1827.30 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:59:50,937 epoch 7 - iter 243/275 - loss 0.02863521 - time (sec): 10.99 - samples/sec: 1840.26 - lr: 0.000017 - momentum: 0.000000
2023-10-17 08:59:52,153 epoch 7 - iter 270/275 - loss 0.02704741 - time (sec): 12.21 - samples/sec: 1830.73 - lr: 0.000017 - momentum: 0.000000
2023-10-17 08:59:52,375 ----------------------------------------------------------------------------------------------------
2023-10-17 08:59:52,375 EPOCH 7 done: loss 0.0266 - lr: 0.000017
2023-10-17 08:59:53,010 DEV : loss 0.18936654925346375 - f1-score (micro avg) 0.8661
2023-10-17 08:59:53,015 ----------------------------------------------------------------------------------------------------
2023-10-17 08:59:54,241 epoch 8 - iter 27/275 - loss 0.01591955 - time (sec): 1.22 - samples/sec: 1905.76 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:59:55,473 epoch 8 - iter 54/275 - loss 0.01525013 - time (sec): 2.46 - samples/sec: 1931.06 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:59:56,676 epoch 8 - iter 81/275 - loss 0.01576595 - time (sec): 3.66 - samples/sec: 1865.01 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:59:57,880 epoch 8 - iter 108/275 - loss 0.02936768 - time (sec): 4.86 - samples/sec: 1838.25 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:59:59,121 epoch 8 - iter 135/275 - loss 0.02952848 - time (sec): 6.10 - samples/sec: 1823.65 - lr: 0.000014 - momentum: 0.000000
2023-10-17 09:00:00,326 epoch 8 - iter 162/275 - loss 0.02768193 - time (sec): 7.31 - samples/sec: 1838.08 - lr: 0.000014 - momentum: 0.000000
2023-10-17 09:00:01,548 epoch 8 - iter 189/275 - loss 0.02911434 - time (sec): 8.53 - samples/sec: 1827.23 - lr: 0.000013 - momentum: 0.000000
2023-10-17 09:00:02,765 epoch 8 - iter 216/275 - loss 0.02582105 - time (sec): 9.75 - samples/sec: 1824.74 - lr: 0.000012 - momentum: 0.000000
2023-10-17 09:00:04,011 epoch 8 - iter 243/275 - loss 0.02566323 - time (sec): 11.00 - samples/sec: 1839.93 - lr: 0.000012 - momentum: 0.000000
2023-10-17 09:00:05,181 epoch 8 - iter 270/275 - loss 0.02392802 - time (sec): 12.17 - samples/sec: 1849.66 - lr: 0.000011 - momentum: 0.000000
2023-10-17 09:00:05,394 ----------------------------------------------------------------------------------------------------
2023-10-17 09:00:05,394 EPOCH 8 done: loss 0.0236 - lr: 0.000011
2023-10-17 09:00:06,050 DEV : loss 0.19396452605724335 - f1-score (micro avg) 0.8786
2023-10-17 09:00:06,055 saving best model
2023-10-17 09:00:06,500 ----------------------------------------------------------------------------------------------------
2023-10-17 09:00:07,759 epoch 9 - iter 27/275 - loss 0.02131835 - time (sec): 1.26 - samples/sec: 1831.06 - lr: 0.000011 - momentum: 0.000000
2023-10-17 09:00:08,979 epoch 9 - iter 54/275 - loss 0.01584945 - time (sec): 2.48 - samples/sec: 1792.56 - lr: 0.000010 - momentum: 0.000000
2023-10-17 09:00:10,187 epoch 9 - iter 81/275 - loss 0.01163448 - time (sec): 3.68 - samples/sec: 1788.02 - lr: 0.000010 - momentum: 0.000000
2023-10-17 09:00:11,413 epoch 9 - iter 108/275 - loss 0.01764179 - time (sec): 4.91 - samples/sec: 1802.25 - lr: 0.000009 - momentum: 0.000000
2023-10-17 09:00:12,655 epoch 9 - iter 135/275 - loss 0.01935884 - time (sec): 6.15 - samples/sec: 1791.20 - lr: 0.000009 - momentum: 0.000000
2023-10-17 09:00:13,878 epoch 9 - iter 162/275 - loss 0.01708720 - time (sec): 7.38 - samples/sec: 1786.96 - lr: 0.000008 - momentum: 0.000000
2023-10-17 09:00:15,124 epoch 9 - iter 189/275 - loss 0.01596518 - time (sec): 8.62 - samples/sec: 1813.88 - lr: 0.000007 - momentum: 0.000000
2023-10-17 09:00:16,378 epoch 9 - iter 216/275 - loss 0.01572838 - time (sec): 9.88 - samples/sec: 1836.50 - lr: 0.000007 - momentum: 0.000000
2023-10-17 09:00:17,606 epoch 9 - iter 243/275 - loss 0.01611628 - time (sec): 11.10 - samples/sec: 1842.04 - lr: 0.000006 - momentum: 0.000000
2023-10-17 09:00:18,832 epoch 9 - iter 270/275 - loss 0.01566388 - time (sec): 12.33 - samples/sec: 1816.57 - lr: 0.000006 - momentum: 0.000000
2023-10-17 09:00:19,062 ----------------------------------------------------------------------------------------------------
2023-10-17 09:00:19,062 EPOCH 9 done: loss 0.0154 - lr: 0.000006
2023-10-17 09:00:19,699 DEV : loss 0.19913232326507568 - f1-score (micro avg) 0.8758
2023-10-17 09:00:19,703 ----------------------------------------------------------------------------------------------------
2023-10-17 09:00:20,910 epoch 10 - iter 27/275 - loss 0.02597141 - time (sec): 1.21 - samples/sec: 1665.80 - lr: 0.000005 - momentum: 0.000000
2023-10-17 09:00:22,172 epoch 10 - iter 54/275 - loss 0.01805506 - time (sec): 2.47 - samples/sec: 1648.16 - lr: 0.000005 - momentum: 0.000000
2023-10-17 09:00:23,400 epoch 10 - iter 81/275 - loss 0.01317504 - time (sec): 3.70 - samples/sec: 1721.95 - lr: 0.000004 - momentum: 0.000000
2023-10-17 09:00:24,643 epoch 10 - iter 108/275 - loss 0.01046857 - time (sec): 4.94 - samples/sec: 1772.68 - lr: 0.000004 - momentum: 0.000000
2023-10-17 09:00:25,890 epoch 10 - iter 135/275 - loss 0.00992712 - time (sec): 6.19 - samples/sec: 1788.40 - lr: 0.000003 - momentum: 0.000000
2023-10-17 09:00:27,118 epoch 10 - iter 162/275 - loss 0.01541164 - time (sec): 7.41 - samples/sec: 1821.87 - lr: 0.000002 - momentum: 0.000000
2023-10-17 09:00:28,329 epoch 10 - iter 189/275 - loss 0.01383796 - time (sec): 8.62 - samples/sec: 1826.20 - lr: 0.000002 - momentum: 0.000000
2023-10-17 09:00:29,529 epoch 10 - iter 216/275 - loss 0.01212638 - time (sec): 9.82 - samples/sec: 1830.76 - lr: 0.000001 - momentum: 0.000000
2023-10-17 09:00:30,752 epoch 10 - iter 243/275 - loss 0.01189330 - time (sec): 11.05 - samples/sec: 1833.49 - lr: 0.000001 - momentum: 0.000000
2023-10-17 09:00:31,980 epoch 10 - iter 270/275 - loss 0.01145146 - time (sec): 12.28 - samples/sec: 1830.13 - lr: 0.000000 - momentum: 0.000000
2023-10-17 09:00:32,203 ----------------------------------------------------------------------------------------------------
2023-10-17 09:00:32,203 EPOCH 10 done: loss 0.0113 - lr: 0.000000
2023-10-17 09:00:32,836 DEV : loss 0.2009844183921814 - f1-score (micro avg) 0.8785
2023-10-17 09:00:33,192 ----------------------------------------------------------------------------------------------------
2023-10-17 09:00:33,193 Loading model from best epoch ...
2023-10-17 09:00:34,548 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 09:00:35,319
Results:
- F-score (micro) 0.9048
- F-score (macro) 0.8717
- Accuracy 0.8403
By class:
precision recall f1-score support
scope 0.8920 0.8920 0.8920 176
pers 0.9837 0.9453 0.9641 128
work 0.8472 0.8243 0.8356 74
object 1.0000 1.0000 1.0000 2
loc 1.0000 0.5000 0.6667 2
micro avg 0.9144 0.8953 0.9048 382
macro avg 0.9446 0.8323 0.8717 382
weighted avg 0.9152 0.8953 0.9047 382
2023-10-17 09:00:35,319 ----------------------------------------------------------------------------------------------------
|