File size: 24,022 Bytes
bae6249 |
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 |
2023-10-17 08:36:58,474 ----------------------------------------------------------------------------------------------------
2023-10-17 08:36:58,475 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 08:36:58,475 ----------------------------------------------------------------------------------------------------
2023-10-17 08:36:58,475 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 08:36:58,475 ----------------------------------------------------------------------------------------------------
2023-10-17 08:36:58,475 Train: 1100 sentences
2023-10-17 08:36:58,475 (train_with_dev=False, train_with_test=False)
2023-10-17 08:36:58,475 ----------------------------------------------------------------------------------------------------
2023-10-17 08:36:58,475 Training Params:
2023-10-17 08:36:58,475 - learning_rate: "3e-05"
2023-10-17 08:36:58,475 - mini_batch_size: "4"
2023-10-17 08:36:58,475 - max_epochs: "10"
2023-10-17 08:36:58,475 - shuffle: "True"
2023-10-17 08:36:58,476 ----------------------------------------------------------------------------------------------------
2023-10-17 08:36:58,476 Plugins:
2023-10-17 08:36:58,476 - TensorboardLogger
2023-10-17 08:36:58,476 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 08:36:58,476 ----------------------------------------------------------------------------------------------------
2023-10-17 08:36:58,476 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 08:36:58,476 - metric: "('micro avg', 'f1-score')"
2023-10-17 08:36:58,476 ----------------------------------------------------------------------------------------------------
2023-10-17 08:36:58,476 Computation:
2023-10-17 08:36:58,476 - compute on device: cuda:0
2023-10-17 08:36:58,476 - embedding storage: none
2023-10-17 08:36:58,476 ----------------------------------------------------------------------------------------------------
2023-10-17 08:36:58,476 Model training base path: "hmbench-ajmc/de-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3"
2023-10-17 08:36:58,476 ----------------------------------------------------------------------------------------------------
2023-10-17 08:36:58,476 ----------------------------------------------------------------------------------------------------
2023-10-17 08:36:58,476 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 08:36:59,775 epoch 1 - iter 27/275 - loss 4.18458578 - time (sec): 1.30 - samples/sec: 1607.98 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:37:01,092 epoch 1 - iter 54/275 - loss 3.52995056 - time (sec): 2.62 - samples/sec: 1665.60 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:37:02,364 epoch 1 - iter 81/275 - loss 2.73946158 - time (sec): 3.89 - samples/sec: 1742.61 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:37:03,575 epoch 1 - iter 108/275 - loss 2.32116704 - time (sec): 5.10 - samples/sec: 1717.25 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:37:04,802 epoch 1 - iter 135/275 - loss 1.97126016 - time (sec): 6.33 - samples/sec: 1750.19 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:37:06,020 epoch 1 - iter 162/275 - loss 1.71777691 - time (sec): 7.54 - samples/sec: 1773.32 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:37:07,229 epoch 1 - iter 189/275 - loss 1.54243129 - time (sec): 8.75 - samples/sec: 1772.43 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:37:08,460 epoch 1 - iter 216/275 - loss 1.39723452 - time (sec): 9.98 - samples/sec: 1779.60 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:37:09,699 epoch 1 - iter 243/275 - loss 1.27203012 - time (sec): 11.22 - samples/sec: 1793.76 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:37:10,955 epoch 1 - iter 270/275 - loss 1.17202515 - time (sec): 12.48 - samples/sec: 1794.93 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:37:11,175 ----------------------------------------------------------------------------------------------------
2023-10-17 08:37:11,175 EPOCH 1 done: loss 1.1580 - lr: 0.000029
2023-10-17 08:37:12,109 DEV : loss 0.24835138022899628 - f1-score (micro avg) 0.6724
2023-10-17 08:37:12,118 saving best model
2023-10-17 08:37:12,514 ----------------------------------------------------------------------------------------------------
2023-10-17 08:37:13,747 epoch 2 - iter 27/275 - loss 0.29820852 - time (sec): 1.23 - samples/sec: 1819.76 - lr: 0.000030 - momentum: 0.000000
2023-10-17 08:37:15,002 epoch 2 - iter 54/275 - loss 0.23526719 - time (sec): 2.49 - samples/sec: 1815.59 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:37:16,252 epoch 2 - iter 81/275 - loss 0.21389662 - time (sec): 3.74 - samples/sec: 1840.47 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:37:17,492 epoch 2 - iter 108/275 - loss 0.20199060 - time (sec): 4.98 - samples/sec: 1829.24 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:37:18,780 epoch 2 - iter 135/275 - loss 0.20714418 - time (sec): 6.26 - samples/sec: 1849.93 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:37:20,000 epoch 2 - iter 162/275 - loss 0.19899218 - time (sec): 7.48 - samples/sec: 1839.93 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:37:21,223 epoch 2 - iter 189/275 - loss 0.18640584 - time (sec): 8.71 - samples/sec: 1826.78 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:37:22,426 epoch 2 - iter 216/275 - loss 0.17495426 - time (sec): 9.91 - samples/sec: 1823.46 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:37:23,624 epoch 2 - iter 243/275 - loss 0.17167750 - time (sec): 11.11 - samples/sec: 1813.93 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:37:24,834 epoch 2 - iter 270/275 - loss 0.17281482 - time (sec): 12.32 - samples/sec: 1821.11 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:37:25,066 ----------------------------------------------------------------------------------------------------
2023-10-17 08:37:25,066 EPOCH 2 done: loss 0.1708 - lr: 0.000027
2023-10-17 08:37:25,708 DEV : loss 0.17084632813930511 - f1-score (micro avg) 0.7878
2023-10-17 08:37:25,713 saving best model
2023-10-17 08:37:26,168 ----------------------------------------------------------------------------------------------------
2023-10-17 08:37:27,514 epoch 3 - iter 27/275 - loss 0.09476884 - time (sec): 1.34 - samples/sec: 1858.77 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:37:28,804 epoch 3 - iter 54/275 - loss 0.10381617 - time (sec): 2.63 - samples/sec: 1816.48 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:37:30,061 epoch 3 - iter 81/275 - loss 0.08816859 - time (sec): 3.89 - samples/sec: 1809.77 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:37:31,279 epoch 3 - iter 108/275 - loss 0.09298207 - time (sec): 5.10 - samples/sec: 1858.42 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:37:32,446 epoch 3 - iter 135/275 - loss 0.08748654 - time (sec): 6.27 - samples/sec: 1829.77 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:37:33,636 epoch 3 - iter 162/275 - loss 0.08756638 - time (sec): 7.46 - samples/sec: 1832.04 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:37:34,860 epoch 3 - iter 189/275 - loss 0.09520057 - time (sec): 8.68 - samples/sec: 1832.53 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:37:36,073 epoch 3 - iter 216/275 - loss 0.09802843 - time (sec): 9.90 - samples/sec: 1837.75 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:37:37,279 epoch 3 - iter 243/275 - loss 0.10354344 - time (sec): 11.10 - samples/sec: 1832.55 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:37:38,492 epoch 3 - iter 270/275 - loss 0.10281829 - time (sec): 12.32 - samples/sec: 1822.19 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:37:38,722 ----------------------------------------------------------------------------------------------------
2023-10-17 08:37:38,722 EPOCH 3 done: loss 0.1016 - lr: 0.000023
2023-10-17 08:37:39,360 DEV : loss 0.1669451892375946 - f1-score (micro avg) 0.8578
2023-10-17 08:37:39,365 saving best model
2023-10-17 08:37:39,814 ----------------------------------------------------------------------------------------------------
2023-10-17 08:37:41,065 epoch 4 - iter 27/275 - loss 0.06433406 - time (sec): 1.25 - samples/sec: 1732.40 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:37:42,286 epoch 4 - iter 54/275 - loss 0.05325618 - time (sec): 2.47 - samples/sec: 1795.30 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:37:43,493 epoch 4 - iter 81/275 - loss 0.05697675 - time (sec): 3.68 - samples/sec: 1803.32 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:37:44,721 epoch 4 - iter 108/275 - loss 0.05585319 - time (sec): 4.91 - samples/sec: 1757.38 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:37:45,977 epoch 4 - iter 135/275 - loss 0.06478508 - time (sec): 6.16 - samples/sec: 1771.47 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:37:47,210 epoch 4 - iter 162/275 - loss 0.06539720 - time (sec): 7.39 - samples/sec: 1786.82 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:37:48,441 epoch 4 - iter 189/275 - loss 0.07160567 - time (sec): 8.63 - samples/sec: 1767.48 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:37:49,673 epoch 4 - iter 216/275 - loss 0.07816350 - time (sec): 9.86 - samples/sec: 1798.06 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:37:50,930 epoch 4 - iter 243/275 - loss 0.08021309 - time (sec): 11.11 - samples/sec: 1793.74 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:37:52,192 epoch 4 - iter 270/275 - loss 0.08200783 - time (sec): 12.38 - samples/sec: 1803.55 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:37:52,414 ----------------------------------------------------------------------------------------------------
2023-10-17 08:37:52,414 EPOCH 4 done: loss 0.0815 - lr: 0.000020
2023-10-17 08:37:53,069 DEV : loss 0.17194409668445587 - f1-score (micro avg) 0.8676
2023-10-17 08:37:53,074 saving best model
2023-10-17 08:37:53,493 ----------------------------------------------------------------------------------------------------
2023-10-17 08:37:54,661 epoch 5 - iter 27/275 - loss 0.10193073 - time (sec): 1.16 - samples/sec: 1980.50 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:37:55,826 epoch 5 - iter 54/275 - loss 0.10581887 - time (sec): 2.33 - samples/sec: 1992.98 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:37:56,994 epoch 5 - iter 81/275 - loss 0.08517821 - time (sec): 3.49 - samples/sec: 1958.62 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:37:58,283 epoch 5 - iter 108/275 - loss 0.08332972 - time (sec): 4.78 - samples/sec: 1885.61 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:37:59,502 epoch 5 - iter 135/275 - loss 0.07897071 - time (sec): 6.00 - samples/sec: 1878.56 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:38:00,721 epoch 5 - iter 162/275 - loss 0.08025946 - time (sec): 7.22 - samples/sec: 1863.81 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:38:01,973 epoch 5 - iter 189/275 - loss 0.07544818 - time (sec): 8.47 - samples/sec: 1862.92 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:38:03,224 epoch 5 - iter 216/275 - loss 0.07174048 - time (sec): 9.72 - samples/sec: 1860.61 - lr: 0.000017 - momentum: 0.000000
2023-10-17 08:38:04,430 epoch 5 - iter 243/275 - loss 0.06753309 - time (sec): 10.93 - samples/sec: 1856.49 - lr: 0.000017 - momentum: 0.000000
2023-10-17 08:38:05,649 epoch 5 - iter 270/275 - loss 0.06679138 - time (sec): 12.15 - samples/sec: 1841.03 - lr: 0.000017 - momentum: 0.000000
2023-10-17 08:38:05,872 ----------------------------------------------------------------------------------------------------
2023-10-17 08:38:05,872 EPOCH 5 done: loss 0.0656 - lr: 0.000017
2023-10-17 08:38:06,507 DEV : loss 0.18873989582061768 - f1-score (micro avg) 0.8697
2023-10-17 08:38:06,512 saving best model
2023-10-17 08:38:06,950 ----------------------------------------------------------------------------------------------------
2023-10-17 08:38:08,212 epoch 6 - iter 27/275 - loss 0.05035342 - time (sec): 1.26 - samples/sec: 1817.82 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:38:09,427 epoch 6 - iter 54/275 - loss 0.05035438 - time (sec): 2.48 - samples/sec: 1802.99 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:38:10,637 epoch 6 - iter 81/275 - loss 0.05778617 - time (sec): 3.69 - samples/sec: 1773.27 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:38:11,834 epoch 6 - iter 108/275 - loss 0.06517433 - time (sec): 4.88 - samples/sec: 1764.60 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:38:13,077 epoch 6 - iter 135/275 - loss 0.06395069 - time (sec): 6.13 - samples/sec: 1811.73 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:38:14,284 epoch 6 - iter 162/275 - loss 0.05909466 - time (sec): 7.33 - samples/sec: 1787.73 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:38:15,483 epoch 6 - iter 189/275 - loss 0.05858764 - time (sec): 8.53 - samples/sec: 1827.95 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:38:16,675 epoch 6 - iter 216/275 - loss 0.05612642 - time (sec): 9.72 - samples/sec: 1829.87 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:38:17,867 epoch 6 - iter 243/275 - loss 0.05375890 - time (sec): 10.92 - samples/sec: 1823.26 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:38:19,068 epoch 6 - iter 270/275 - loss 0.05000841 - time (sec): 12.12 - samples/sec: 1837.48 - lr: 0.000013 - momentum: 0.000000
2023-10-17 08:38:19,295 ----------------------------------------------------------------------------------------------------
2023-10-17 08:38:19,295 EPOCH 6 done: loss 0.0503 - lr: 0.000013
2023-10-17 08:38:19,926 DEV : loss 0.19087766110897064 - f1-score (micro avg) 0.8643
2023-10-17 08:38:19,931 ----------------------------------------------------------------------------------------------------
2023-10-17 08:38:21,162 epoch 7 - iter 27/275 - loss 0.04592519 - time (sec): 1.23 - samples/sec: 1883.07 - lr: 0.000013 - momentum: 0.000000
2023-10-17 08:38:22,366 epoch 7 - iter 54/275 - loss 0.03291311 - time (sec): 2.43 - samples/sec: 1838.45 - lr: 0.000013 - momentum: 0.000000
2023-10-17 08:38:23,561 epoch 7 - iter 81/275 - loss 0.02930348 - time (sec): 3.63 - samples/sec: 1820.31 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:38:24,776 epoch 7 - iter 108/275 - loss 0.04025652 - time (sec): 4.84 - samples/sec: 1783.80 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:38:26,025 epoch 7 - iter 135/275 - loss 0.03406485 - time (sec): 6.09 - samples/sec: 1832.62 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:38:27,240 epoch 7 - iter 162/275 - loss 0.04109484 - time (sec): 7.31 - samples/sec: 1852.06 - lr: 0.000011 - momentum: 0.000000
2023-10-17 08:38:28,446 epoch 7 - iter 189/275 - loss 0.04358684 - time (sec): 8.51 - samples/sec: 1840.09 - lr: 0.000011 - momentum: 0.000000
2023-10-17 08:38:29,663 epoch 7 - iter 216/275 - loss 0.04150936 - time (sec): 9.73 - samples/sec: 1848.67 - lr: 0.000011 - momentum: 0.000000
2023-10-17 08:38:30,890 epoch 7 - iter 243/275 - loss 0.04573996 - time (sec): 10.96 - samples/sec: 1843.23 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:38:32,107 epoch 7 - iter 270/275 - loss 0.04355072 - time (sec): 12.18 - samples/sec: 1846.29 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:38:32,333 ----------------------------------------------------------------------------------------------------
2023-10-17 08:38:32,333 EPOCH 7 done: loss 0.0451 - lr: 0.000010
2023-10-17 08:38:32,970 DEV : loss 0.18272067606449127 - f1-score (micro avg) 0.8841
2023-10-17 08:38:32,974 saving best model
2023-10-17 08:38:33,402 ----------------------------------------------------------------------------------------------------
2023-10-17 08:38:34,625 epoch 8 - iter 27/275 - loss 0.03791177 - time (sec): 1.22 - samples/sec: 1818.03 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:38:35,843 epoch 8 - iter 54/275 - loss 0.03884052 - time (sec): 2.44 - samples/sec: 1798.40 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:38:37,056 epoch 8 - iter 81/275 - loss 0.03522629 - time (sec): 3.65 - samples/sec: 1876.18 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:38:38,213 epoch 8 - iter 108/275 - loss 0.03898247 - time (sec): 4.81 - samples/sec: 1900.36 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:38:39,385 epoch 8 - iter 135/275 - loss 0.03363928 - time (sec): 5.98 - samples/sec: 1897.76 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:38:40,576 epoch 8 - iter 162/275 - loss 0.02922465 - time (sec): 7.17 - samples/sec: 1852.74 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:38:41,825 epoch 8 - iter 189/275 - loss 0.02981224 - time (sec): 8.42 - samples/sec: 1850.17 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:38:43,040 epoch 8 - iter 216/275 - loss 0.03064469 - time (sec): 9.64 - samples/sec: 1851.37 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:38:44,247 epoch 8 - iter 243/275 - loss 0.02957048 - time (sec): 10.84 - samples/sec: 1850.42 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:38:45,469 epoch 8 - iter 270/275 - loss 0.03328664 - time (sec): 12.06 - samples/sec: 1855.69 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:38:45,697 ----------------------------------------------------------------------------------------------------
2023-10-17 08:38:45,697 EPOCH 8 done: loss 0.0331 - lr: 0.000007
2023-10-17 08:38:46,406 DEV : loss 0.17866075038909912 - f1-score (micro avg) 0.8835
2023-10-17 08:38:46,412 ----------------------------------------------------------------------------------------------------
2023-10-17 08:38:47,639 epoch 9 - iter 27/275 - loss 0.02070222 - time (sec): 1.23 - samples/sec: 1816.68 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:38:48,857 epoch 9 - iter 54/275 - loss 0.02809420 - time (sec): 2.44 - samples/sec: 1912.52 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:38:50,059 epoch 9 - iter 81/275 - loss 0.03070519 - time (sec): 3.65 - samples/sec: 1858.56 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:38:51,254 epoch 9 - iter 108/275 - loss 0.03042463 - time (sec): 4.84 - samples/sec: 1851.92 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:38:52,474 epoch 9 - iter 135/275 - loss 0.02817818 - time (sec): 6.06 - samples/sec: 1860.00 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:38:53,696 epoch 9 - iter 162/275 - loss 0.02493376 - time (sec): 7.28 - samples/sec: 1851.21 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:38:54,935 epoch 9 - iter 189/275 - loss 0.03113342 - time (sec): 8.52 - samples/sec: 1840.47 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:38:56,177 epoch 9 - iter 216/275 - loss 0.03047445 - time (sec): 9.76 - samples/sec: 1839.37 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:38:57,419 epoch 9 - iter 243/275 - loss 0.02770802 - time (sec): 11.01 - samples/sec: 1817.40 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:38:58,661 epoch 9 - iter 270/275 - loss 0.02697792 - time (sec): 12.25 - samples/sec: 1822.10 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:38:58,881 ----------------------------------------------------------------------------------------------------
2023-10-17 08:38:58,881 EPOCH 9 done: loss 0.0269 - lr: 0.000003
2023-10-17 08:38:59,631 DEV : loss 0.17246393859386444 - f1-score (micro avg) 0.8803
2023-10-17 08:38:59,636 ----------------------------------------------------------------------------------------------------
2023-10-17 08:39:00,822 epoch 10 - iter 27/275 - loss 0.06560676 - time (sec): 1.19 - samples/sec: 2181.30 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:39:02,049 epoch 10 - iter 54/275 - loss 0.04650226 - time (sec): 2.41 - samples/sec: 2062.30 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:39:03,285 epoch 10 - iter 81/275 - loss 0.03455730 - time (sec): 3.65 - samples/sec: 2025.03 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:39:04,480 epoch 10 - iter 108/275 - loss 0.02764019 - time (sec): 4.84 - samples/sec: 1950.65 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:39:05,691 epoch 10 - iter 135/275 - loss 0.02670571 - time (sec): 6.05 - samples/sec: 1911.94 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:39:06,908 epoch 10 - iter 162/275 - loss 0.02432927 - time (sec): 7.27 - samples/sec: 1886.79 - lr: 0.000001 - momentum: 0.000000
2023-10-17 08:39:08,144 epoch 10 - iter 189/275 - loss 0.02362532 - time (sec): 8.51 - samples/sec: 1860.15 - lr: 0.000001 - momentum: 0.000000
2023-10-17 08:39:09,374 epoch 10 - iter 216/275 - loss 0.02263105 - time (sec): 9.74 - samples/sec: 1841.53 - lr: 0.000001 - momentum: 0.000000
2023-10-17 08:39:10,593 epoch 10 - iter 243/275 - loss 0.02183048 - time (sec): 10.96 - samples/sec: 1847.62 - lr: 0.000000 - momentum: 0.000000
2023-10-17 08:39:11,842 epoch 10 - iter 270/275 - loss 0.02161543 - time (sec): 12.21 - samples/sec: 1832.25 - lr: 0.000000 - momentum: 0.000000
2023-10-17 08:39:12,068 ----------------------------------------------------------------------------------------------------
2023-10-17 08:39:12,068 EPOCH 10 done: loss 0.0212 - lr: 0.000000
2023-10-17 08:39:12,729 DEV : loss 0.17962802946567535 - f1-score (micro avg) 0.8816
2023-10-17 08:39:13,079 ----------------------------------------------------------------------------------------------------
2023-10-17 08:39:13,080 Loading model from best epoch ...
2023-10-17 08:39:14,442 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 08:39:15,253
Results:
- F-score (micro) 0.8863
- F-score (macro) 0.6622
- Accuracy 0.8091
By class:
precision recall f1-score support
scope 0.8708 0.8807 0.8757 176
pers 0.9603 0.9453 0.9528 128
work 0.7949 0.8378 0.8158 74
loc 1.0000 0.5000 0.6667 2
object 0.0000 0.0000 0.0000 2
micro avg 0.8851 0.8874 0.8863 382
macro avg 0.7252 0.6328 0.6622 382
weighted avg 0.8822 0.8874 0.8842 382
2023-10-17 08:39:15,253 ----------------------------------------------------------------------------------------------------
|