File size: 24,099 Bytes
24c6d15 |
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 |
2023-10-23 18:05:18,170 ----------------------------------------------------------------------------------------------------
2023-10-23 18:05:18,172 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(64001, 768)
(position_embeddings): Embedding(512, 768)
(token_type_embeddings): Embedding(2, 768)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): BertEncoder(
(layer): ModuleList(
(0-11): 12 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=768, out_features=768, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-23 18:05:18,172 ----------------------------------------------------------------------------------------------------
2023-10-23 18:05:18,172 MultiCorpus: 1214 train + 266 dev + 251 test sentences
- NER_HIPE_2022 Corpus: 1214 train + 266 dev + 251 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/en/with_doc_seperator
2023-10-23 18:05:18,172 ----------------------------------------------------------------------------------------------------
2023-10-23 18:05:18,172 Train: 1214 sentences
2023-10-23 18:05:18,172 (train_with_dev=False, train_with_test=False)
2023-10-23 18:05:18,172 ----------------------------------------------------------------------------------------------------
2023-10-23 18:05:18,172 Training Params:
2023-10-23 18:05:18,172 - learning_rate: "5e-05"
2023-10-23 18:05:18,172 - mini_batch_size: "4"
2023-10-23 18:05:18,172 - max_epochs: "10"
2023-10-23 18:05:18,172 - shuffle: "True"
2023-10-23 18:05:18,172 ----------------------------------------------------------------------------------------------------
2023-10-23 18:05:18,172 Plugins:
2023-10-23 18:05:18,173 - TensorboardLogger
2023-10-23 18:05:18,173 - LinearScheduler | warmup_fraction: '0.1'
2023-10-23 18:05:18,173 ----------------------------------------------------------------------------------------------------
2023-10-23 18:05:18,173 Final evaluation on model from best epoch (best-model.pt)
2023-10-23 18:05:18,173 - metric: "('micro avg', 'f1-score')"
2023-10-23 18:05:18,173 ----------------------------------------------------------------------------------------------------
2023-10-23 18:05:18,173 Computation:
2023-10-23 18:05:18,173 - compute on device: cuda:0
2023-10-23 18:05:18,173 - embedding storage: none
2023-10-23 18:05:18,173 ----------------------------------------------------------------------------------------------------
2023-10-23 18:05:18,173 Model training base path: "hmbench-ajmc/en-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-23 18:05:18,173 ----------------------------------------------------------------------------------------------------
2023-10-23 18:05:18,173 ----------------------------------------------------------------------------------------------------
2023-10-23 18:05:18,173 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-23 18:05:19,738 epoch 1 - iter 30/304 - loss 3.02351216 - time (sec): 1.56 - samples/sec: 2092.12 - lr: 0.000005 - momentum: 0.000000
2023-10-23 18:05:21,366 epoch 1 - iter 60/304 - loss 2.16279826 - time (sec): 3.19 - samples/sec: 1911.82 - lr: 0.000010 - momentum: 0.000000
2023-10-23 18:05:23,007 epoch 1 - iter 90/304 - loss 1.64732925 - time (sec): 4.83 - samples/sec: 1911.64 - lr: 0.000015 - momentum: 0.000000
2023-10-23 18:05:24,648 epoch 1 - iter 120/304 - loss 1.30376680 - time (sec): 6.47 - samples/sec: 1966.46 - lr: 0.000020 - momentum: 0.000000
2023-10-23 18:05:26,288 epoch 1 - iter 150/304 - loss 1.11797657 - time (sec): 8.11 - samples/sec: 1933.49 - lr: 0.000025 - momentum: 0.000000
2023-10-23 18:05:27,926 epoch 1 - iter 180/304 - loss 0.98825947 - time (sec): 9.75 - samples/sec: 1913.77 - lr: 0.000029 - momentum: 0.000000
2023-10-23 18:05:29,564 epoch 1 - iter 210/304 - loss 0.88981727 - time (sec): 11.39 - samples/sec: 1897.65 - lr: 0.000034 - momentum: 0.000000
2023-10-23 18:05:31,197 epoch 1 - iter 240/304 - loss 0.80330881 - time (sec): 13.02 - samples/sec: 1882.05 - lr: 0.000039 - momentum: 0.000000
2023-10-23 18:05:32,835 epoch 1 - iter 270/304 - loss 0.74394110 - time (sec): 14.66 - samples/sec: 1872.07 - lr: 0.000044 - momentum: 0.000000
2023-10-23 18:05:34,471 epoch 1 - iter 300/304 - loss 0.68489685 - time (sec): 16.30 - samples/sec: 1880.48 - lr: 0.000049 - momentum: 0.000000
2023-10-23 18:05:34,685 ----------------------------------------------------------------------------------------------------
2023-10-23 18:05:34,685 EPOCH 1 done: loss 0.6805 - lr: 0.000049
2023-10-23 18:05:35,473 DEV : loss 0.15541432797908783 - f1-score (micro avg) 0.726
2023-10-23 18:05:35,481 saving best model
2023-10-23 18:05:35,857 ----------------------------------------------------------------------------------------------------
2023-10-23 18:05:37,496 epoch 2 - iter 30/304 - loss 0.19423395 - time (sec): 1.64 - samples/sec: 1971.04 - lr: 0.000049 - momentum: 0.000000
2023-10-23 18:05:39,135 epoch 2 - iter 60/304 - loss 0.15037454 - time (sec): 3.28 - samples/sec: 1867.22 - lr: 0.000049 - momentum: 0.000000
2023-10-23 18:05:40,779 epoch 2 - iter 90/304 - loss 0.14114291 - time (sec): 4.92 - samples/sec: 1855.21 - lr: 0.000048 - momentum: 0.000000
2023-10-23 18:05:42,426 epoch 2 - iter 120/304 - loss 0.13326513 - time (sec): 6.57 - samples/sec: 1870.75 - lr: 0.000048 - momentum: 0.000000
2023-10-23 18:05:44,070 epoch 2 - iter 150/304 - loss 0.12683531 - time (sec): 8.21 - samples/sec: 1878.75 - lr: 0.000047 - momentum: 0.000000
2023-10-23 18:05:45,710 epoch 2 - iter 180/304 - loss 0.12499108 - time (sec): 9.85 - samples/sec: 1863.27 - lr: 0.000047 - momentum: 0.000000
2023-10-23 18:05:47,357 epoch 2 - iter 210/304 - loss 0.12849916 - time (sec): 11.50 - samples/sec: 1857.08 - lr: 0.000046 - momentum: 0.000000
2023-10-23 18:05:49,009 epoch 2 - iter 240/304 - loss 0.13158152 - time (sec): 13.15 - samples/sec: 1864.61 - lr: 0.000046 - momentum: 0.000000
2023-10-23 18:05:50,657 epoch 2 - iter 270/304 - loss 0.12718118 - time (sec): 14.80 - samples/sec: 1861.95 - lr: 0.000045 - momentum: 0.000000
2023-10-23 18:05:52,291 epoch 2 - iter 300/304 - loss 0.12696736 - time (sec): 16.43 - samples/sec: 1865.31 - lr: 0.000045 - momentum: 0.000000
2023-10-23 18:05:52,504 ----------------------------------------------------------------------------------------------------
2023-10-23 18:05:52,504 EPOCH 2 done: loss 0.1293 - lr: 0.000045
2023-10-23 18:05:53,359 DEV : loss 0.17078670859336853 - f1-score (micro avg) 0.76
2023-10-23 18:05:53,367 saving best model
2023-10-23 18:05:53,888 ----------------------------------------------------------------------------------------------------
2023-10-23 18:05:55,526 epoch 3 - iter 30/304 - loss 0.08939435 - time (sec): 1.64 - samples/sec: 1863.49 - lr: 0.000044 - momentum: 0.000000
2023-10-23 18:05:57,173 epoch 3 - iter 60/304 - loss 0.07897526 - time (sec): 3.28 - samples/sec: 1808.72 - lr: 0.000043 - momentum: 0.000000
2023-10-23 18:05:58,810 epoch 3 - iter 90/304 - loss 0.08706321 - time (sec): 4.92 - samples/sec: 1820.30 - lr: 0.000043 - momentum: 0.000000
2023-10-23 18:06:00,456 epoch 3 - iter 120/304 - loss 0.07668656 - time (sec): 6.57 - samples/sec: 1832.75 - lr: 0.000042 - momentum: 0.000000
2023-10-23 18:06:02,105 epoch 3 - iter 150/304 - loss 0.07354657 - time (sec): 8.21 - samples/sec: 1826.17 - lr: 0.000042 - momentum: 0.000000
2023-10-23 18:06:03,754 epoch 3 - iter 180/304 - loss 0.08189361 - time (sec): 9.86 - samples/sec: 1846.34 - lr: 0.000041 - momentum: 0.000000
2023-10-23 18:06:05,403 epoch 3 - iter 210/304 - loss 0.07732853 - time (sec): 11.51 - samples/sec: 1871.12 - lr: 0.000041 - momentum: 0.000000
2023-10-23 18:06:07,047 epoch 3 - iter 240/304 - loss 0.08312185 - time (sec): 13.16 - samples/sec: 1868.54 - lr: 0.000040 - momentum: 0.000000
2023-10-23 18:06:08,681 epoch 3 - iter 270/304 - loss 0.08169576 - time (sec): 14.79 - samples/sec: 1877.51 - lr: 0.000040 - momentum: 0.000000
2023-10-23 18:06:10,296 epoch 3 - iter 300/304 - loss 0.08673653 - time (sec): 16.41 - samples/sec: 1870.86 - lr: 0.000039 - momentum: 0.000000
2023-10-23 18:06:10,506 ----------------------------------------------------------------------------------------------------
2023-10-23 18:06:10,507 EPOCH 3 done: loss 0.0861 - lr: 0.000039
2023-10-23 18:06:11,363 DEV : loss 0.17484019696712494 - f1-score (micro avg) 0.8167
2023-10-23 18:06:11,371 saving best model
2023-10-23 18:06:11,897 ----------------------------------------------------------------------------------------------------
2023-10-23 18:06:13,537 epoch 4 - iter 30/304 - loss 0.04368220 - time (sec): 1.64 - samples/sec: 1658.87 - lr: 0.000038 - momentum: 0.000000
2023-10-23 18:06:15,181 epoch 4 - iter 60/304 - loss 0.05730760 - time (sec): 3.28 - samples/sec: 1825.30 - lr: 0.000038 - momentum: 0.000000
2023-10-23 18:06:16,788 epoch 4 - iter 90/304 - loss 0.06114841 - time (sec): 4.89 - samples/sec: 1849.70 - lr: 0.000037 - momentum: 0.000000
2023-10-23 18:06:18,420 epoch 4 - iter 120/304 - loss 0.05553906 - time (sec): 6.52 - samples/sec: 1850.44 - lr: 0.000037 - momentum: 0.000000
2023-10-23 18:06:20,054 epoch 4 - iter 150/304 - loss 0.06034393 - time (sec): 8.15 - samples/sec: 1885.51 - lr: 0.000036 - momentum: 0.000000
2023-10-23 18:06:21,679 epoch 4 - iter 180/304 - loss 0.06426993 - time (sec): 9.78 - samples/sec: 1865.59 - lr: 0.000036 - momentum: 0.000000
2023-10-23 18:06:23,330 epoch 4 - iter 210/304 - loss 0.06643169 - time (sec): 11.43 - samples/sec: 1891.08 - lr: 0.000035 - momentum: 0.000000
2023-10-23 18:06:24,979 epoch 4 - iter 240/304 - loss 0.06216304 - time (sec): 13.08 - samples/sec: 1900.70 - lr: 0.000035 - momentum: 0.000000
2023-10-23 18:06:26,611 epoch 4 - iter 270/304 - loss 0.06119980 - time (sec): 14.71 - samples/sec: 1877.13 - lr: 0.000034 - momentum: 0.000000
2023-10-23 18:06:28,247 epoch 4 - iter 300/304 - loss 0.05807122 - time (sec): 16.35 - samples/sec: 1874.32 - lr: 0.000033 - momentum: 0.000000
2023-10-23 18:06:28,461 ----------------------------------------------------------------------------------------------------
2023-10-23 18:06:28,461 EPOCH 4 done: loss 0.0590 - lr: 0.000033
2023-10-23 18:06:29,313 DEV : loss 0.20702120661735535 - f1-score (micro avg) 0.8028
2023-10-23 18:06:29,320 ----------------------------------------------------------------------------------------------------
2023-10-23 18:06:30,953 epoch 5 - iter 30/304 - loss 0.07189814 - time (sec): 1.63 - samples/sec: 1711.36 - lr: 0.000033 - momentum: 0.000000
2023-10-23 18:06:32,593 epoch 5 - iter 60/304 - loss 0.04765151 - time (sec): 3.27 - samples/sec: 1787.29 - lr: 0.000032 - momentum: 0.000000
2023-10-23 18:06:34,231 epoch 5 - iter 90/304 - loss 0.03673195 - time (sec): 4.91 - samples/sec: 1826.48 - lr: 0.000032 - momentum: 0.000000
2023-10-23 18:06:35,868 epoch 5 - iter 120/304 - loss 0.04243436 - time (sec): 6.55 - samples/sec: 1839.20 - lr: 0.000031 - momentum: 0.000000
2023-10-23 18:06:37,507 epoch 5 - iter 150/304 - loss 0.04100820 - time (sec): 8.19 - samples/sec: 1862.05 - lr: 0.000031 - momentum: 0.000000
2023-10-23 18:06:39,146 epoch 5 - iter 180/304 - loss 0.04005715 - time (sec): 9.82 - samples/sec: 1876.76 - lr: 0.000030 - momentum: 0.000000
2023-10-23 18:06:40,785 epoch 5 - iter 210/304 - loss 0.04081348 - time (sec): 11.46 - samples/sec: 1890.67 - lr: 0.000030 - momentum: 0.000000
2023-10-23 18:06:42,432 epoch 5 - iter 240/304 - loss 0.04483258 - time (sec): 13.11 - samples/sec: 1886.03 - lr: 0.000029 - momentum: 0.000000
2023-10-23 18:06:44,078 epoch 5 - iter 270/304 - loss 0.04381410 - time (sec): 14.76 - samples/sec: 1888.61 - lr: 0.000028 - momentum: 0.000000
2023-10-23 18:06:45,716 epoch 5 - iter 300/304 - loss 0.04177830 - time (sec): 16.39 - samples/sec: 1868.39 - lr: 0.000028 - momentum: 0.000000
2023-10-23 18:06:45,928 ----------------------------------------------------------------------------------------------------
2023-10-23 18:06:45,929 EPOCH 5 done: loss 0.0418 - lr: 0.000028
2023-10-23 18:06:46,772 DEV : loss 0.19316518306732178 - f1-score (micro avg) 0.8464
2023-10-23 18:06:46,779 saving best model
2023-10-23 18:06:47,338 ----------------------------------------------------------------------------------------------------
2023-10-23 18:06:48,973 epoch 6 - iter 30/304 - loss 0.05467245 - time (sec): 1.63 - samples/sec: 1886.52 - lr: 0.000027 - momentum: 0.000000
2023-10-23 18:06:50,612 epoch 6 - iter 60/304 - loss 0.03021958 - time (sec): 3.27 - samples/sec: 1888.73 - lr: 0.000027 - momentum: 0.000000
2023-10-23 18:06:52,244 epoch 6 - iter 90/304 - loss 0.02852577 - time (sec): 4.90 - samples/sec: 1844.84 - lr: 0.000026 - momentum: 0.000000
2023-10-23 18:06:53,884 epoch 6 - iter 120/304 - loss 0.02827307 - time (sec): 6.54 - samples/sec: 1856.42 - lr: 0.000026 - momentum: 0.000000
2023-10-23 18:06:55,532 epoch 6 - iter 150/304 - loss 0.03479194 - time (sec): 8.19 - samples/sec: 1873.89 - lr: 0.000025 - momentum: 0.000000
2023-10-23 18:06:57,167 epoch 6 - iter 180/304 - loss 0.03172006 - time (sec): 9.83 - samples/sec: 1845.55 - lr: 0.000025 - momentum: 0.000000
2023-10-23 18:06:58,806 epoch 6 - iter 210/304 - loss 0.03968617 - time (sec): 11.47 - samples/sec: 1847.34 - lr: 0.000024 - momentum: 0.000000
2023-10-23 18:07:00,442 epoch 6 - iter 240/304 - loss 0.04188832 - time (sec): 13.10 - samples/sec: 1848.07 - lr: 0.000023 - momentum: 0.000000
2023-10-23 18:07:02,083 epoch 6 - iter 270/304 - loss 0.04332220 - time (sec): 14.74 - samples/sec: 1846.32 - lr: 0.000023 - momentum: 0.000000
2023-10-23 18:07:03,721 epoch 6 - iter 300/304 - loss 0.05048982 - time (sec): 16.38 - samples/sec: 1873.09 - lr: 0.000022 - momentum: 0.000000
2023-10-23 18:07:03,936 ----------------------------------------------------------------------------------------------------
2023-10-23 18:07:03,936 EPOCH 6 done: loss 0.0501 - lr: 0.000022
2023-10-23 18:07:04,785 DEV : loss 0.22458811104297638 - f1-score (micro avg) 0.6659
2023-10-23 18:07:04,792 ----------------------------------------------------------------------------------------------------
2023-10-23 18:07:06,425 epoch 7 - iter 30/304 - loss 0.04567402 - time (sec): 1.63 - samples/sec: 1846.80 - lr: 0.000022 - momentum: 0.000000
2023-10-23 18:07:08,068 epoch 7 - iter 60/304 - loss 0.04013202 - time (sec): 3.27 - samples/sec: 1887.08 - lr: 0.000021 - momentum: 0.000000
2023-10-23 18:07:09,712 epoch 7 - iter 90/304 - loss 0.04103430 - time (sec): 4.92 - samples/sec: 1867.22 - lr: 0.000021 - momentum: 0.000000
2023-10-23 18:07:11,354 epoch 7 - iter 120/304 - loss 0.04689034 - time (sec): 6.56 - samples/sec: 1879.38 - lr: 0.000020 - momentum: 0.000000
2023-10-23 18:07:13,007 epoch 7 - iter 150/304 - loss 0.04243080 - time (sec): 8.21 - samples/sec: 1887.35 - lr: 0.000020 - momentum: 0.000000
2023-10-23 18:07:14,626 epoch 7 - iter 180/304 - loss 0.03739778 - time (sec): 9.83 - samples/sec: 1884.23 - lr: 0.000019 - momentum: 0.000000
2023-10-23 18:07:16,244 epoch 7 - iter 210/304 - loss 0.03572045 - time (sec): 11.45 - samples/sec: 1889.73 - lr: 0.000018 - momentum: 0.000000
2023-10-23 18:07:17,860 epoch 7 - iter 240/304 - loss 0.03268610 - time (sec): 13.07 - samples/sec: 1891.94 - lr: 0.000018 - momentum: 0.000000
2023-10-23 18:07:19,499 epoch 7 - iter 270/304 - loss 0.03137129 - time (sec): 14.71 - samples/sec: 1870.60 - lr: 0.000017 - momentum: 0.000000
2023-10-23 18:07:21,141 epoch 7 - iter 300/304 - loss 0.03089265 - time (sec): 16.35 - samples/sec: 1873.73 - lr: 0.000017 - momentum: 0.000000
2023-10-23 18:07:21,356 ----------------------------------------------------------------------------------------------------
2023-10-23 18:07:21,356 EPOCH 7 done: loss 0.0305 - lr: 0.000017
2023-10-23 18:07:22,378 DEV : loss 0.19062967598438263 - f1-score (micro avg) 0.8435
2023-10-23 18:07:22,386 ----------------------------------------------------------------------------------------------------
2023-10-23 18:07:24,031 epoch 8 - iter 30/304 - loss 0.03284706 - time (sec): 1.64 - samples/sec: 1796.18 - lr: 0.000016 - momentum: 0.000000
2023-10-23 18:07:25,666 epoch 8 - iter 60/304 - loss 0.02779091 - time (sec): 3.28 - samples/sec: 1744.32 - lr: 0.000016 - momentum: 0.000000
2023-10-23 18:07:27,302 epoch 8 - iter 90/304 - loss 0.01855685 - time (sec): 4.91 - samples/sec: 1746.35 - lr: 0.000015 - momentum: 0.000000
2023-10-23 18:07:28,948 epoch 8 - iter 120/304 - loss 0.01484052 - time (sec): 6.56 - samples/sec: 1789.62 - lr: 0.000015 - momentum: 0.000000
2023-10-23 18:07:30,589 epoch 8 - iter 150/304 - loss 0.01229122 - time (sec): 8.20 - samples/sec: 1841.21 - lr: 0.000014 - momentum: 0.000000
2023-10-23 18:07:32,228 epoch 8 - iter 180/304 - loss 0.01257633 - time (sec): 9.84 - samples/sec: 1836.27 - lr: 0.000013 - momentum: 0.000000
2023-10-23 18:07:33,865 epoch 8 - iter 210/304 - loss 0.01408353 - time (sec): 11.48 - samples/sec: 1834.05 - lr: 0.000013 - momentum: 0.000000
2023-10-23 18:07:35,504 epoch 8 - iter 240/304 - loss 0.01384160 - time (sec): 13.12 - samples/sec: 1836.24 - lr: 0.000012 - momentum: 0.000000
2023-10-23 18:07:37,149 epoch 8 - iter 270/304 - loss 0.01387053 - time (sec): 14.76 - samples/sec: 1853.15 - lr: 0.000012 - momentum: 0.000000
2023-10-23 18:07:38,791 epoch 8 - iter 300/304 - loss 0.01359108 - time (sec): 16.40 - samples/sec: 1866.58 - lr: 0.000011 - momentum: 0.000000
2023-10-23 18:07:39,007 ----------------------------------------------------------------------------------------------------
2023-10-23 18:07:39,007 EPOCH 8 done: loss 0.0134 - lr: 0.000011
2023-10-23 18:07:39,856 DEV : loss 0.2041168510913849 - f1-score (micro avg) 0.852
2023-10-23 18:07:39,863 saving best model
2023-10-23 18:07:40,387 ----------------------------------------------------------------------------------------------------
2023-10-23 18:07:42,032 epoch 9 - iter 30/304 - loss 0.00600479 - time (sec): 1.64 - samples/sec: 1838.36 - lr: 0.000011 - momentum: 0.000000
2023-10-23 18:07:43,670 epoch 9 - iter 60/304 - loss 0.00346556 - time (sec): 3.28 - samples/sec: 1817.14 - lr: 0.000010 - momentum: 0.000000
2023-10-23 18:07:45,307 epoch 9 - iter 90/304 - loss 0.00522093 - time (sec): 4.92 - samples/sec: 1831.14 - lr: 0.000010 - momentum: 0.000000
2023-10-23 18:07:46,946 epoch 9 - iter 120/304 - loss 0.00635893 - time (sec): 6.56 - samples/sec: 1857.20 - lr: 0.000009 - momentum: 0.000000
2023-10-23 18:07:48,561 epoch 9 - iter 150/304 - loss 0.00829404 - time (sec): 8.17 - samples/sec: 1843.97 - lr: 0.000008 - momentum: 0.000000
2023-10-23 18:07:50,202 epoch 9 - iter 180/304 - loss 0.01123978 - time (sec): 9.81 - samples/sec: 1863.28 - lr: 0.000008 - momentum: 0.000000
2023-10-23 18:07:51,829 epoch 9 - iter 210/304 - loss 0.00958663 - time (sec): 11.44 - samples/sec: 1875.46 - lr: 0.000007 - momentum: 0.000000
2023-10-23 18:07:53,451 epoch 9 - iter 240/304 - loss 0.00999350 - time (sec): 13.06 - samples/sec: 1858.64 - lr: 0.000007 - momentum: 0.000000
2023-10-23 18:07:55,085 epoch 9 - iter 270/304 - loss 0.00938350 - time (sec): 14.70 - samples/sec: 1862.46 - lr: 0.000006 - momentum: 0.000000
2023-10-23 18:07:56,735 epoch 9 - iter 300/304 - loss 0.00917477 - time (sec): 16.35 - samples/sec: 1877.87 - lr: 0.000006 - momentum: 0.000000
2023-10-23 18:07:56,950 ----------------------------------------------------------------------------------------------------
2023-10-23 18:07:56,950 EPOCH 9 done: loss 0.0091 - lr: 0.000006
2023-10-23 18:07:57,812 DEV : loss 0.2051309198141098 - f1-score (micro avg) 0.8456
2023-10-23 18:07:57,820 ----------------------------------------------------------------------------------------------------
2023-10-23 18:07:59,462 epoch 10 - iter 30/304 - loss 0.00983964 - time (sec): 1.64 - samples/sec: 1911.88 - lr: 0.000005 - momentum: 0.000000
2023-10-23 18:08:01,094 epoch 10 - iter 60/304 - loss 0.01371081 - time (sec): 3.27 - samples/sec: 1855.77 - lr: 0.000005 - momentum: 0.000000
2023-10-23 18:08:02,731 epoch 10 - iter 90/304 - loss 0.00942075 - time (sec): 4.91 - samples/sec: 1815.96 - lr: 0.000004 - momentum: 0.000000
2023-10-23 18:08:04,367 epoch 10 - iter 120/304 - loss 0.00928705 - time (sec): 6.55 - samples/sec: 1844.97 - lr: 0.000003 - momentum: 0.000000
2023-10-23 18:08:06,004 epoch 10 - iter 150/304 - loss 0.00814334 - time (sec): 8.18 - samples/sec: 1827.76 - lr: 0.000003 - momentum: 0.000000
2023-10-23 18:08:07,642 epoch 10 - iter 180/304 - loss 0.00692152 - time (sec): 9.82 - samples/sec: 1856.28 - lr: 0.000002 - momentum: 0.000000
2023-10-23 18:08:09,282 epoch 10 - iter 210/304 - loss 0.00802509 - time (sec): 11.46 - samples/sec: 1878.13 - lr: 0.000002 - momentum: 0.000000
2023-10-23 18:08:10,927 epoch 10 - iter 240/304 - loss 0.00720115 - time (sec): 13.11 - samples/sec: 1888.06 - lr: 0.000001 - momentum: 0.000000
2023-10-23 18:08:12,565 epoch 10 - iter 270/304 - loss 0.00642390 - time (sec): 14.74 - samples/sec: 1885.17 - lr: 0.000001 - momentum: 0.000000
2023-10-23 18:08:14,194 epoch 10 - iter 300/304 - loss 0.00718668 - time (sec): 16.37 - samples/sec: 1875.88 - lr: 0.000000 - momentum: 0.000000
2023-10-23 18:08:14,406 ----------------------------------------------------------------------------------------------------
2023-10-23 18:08:14,406 EPOCH 10 done: loss 0.0071 - lr: 0.000000
2023-10-23 18:08:15,325 DEV : loss 0.20948979258537292 - f1-score (micro avg) 0.8476
2023-10-23 18:08:15,746 ----------------------------------------------------------------------------------------------------
2023-10-23 18:08:15,747 Loading model from best epoch ...
2023-10-23 18:08:17,540 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-date, B-date, E-date, I-date, S-object, B-object, E-object, I-object
2023-10-23 18:08:18,429
Results:
- F-score (micro) 0.8016
- F-score (macro) 0.6217
- Accuracy 0.6781
By class:
precision recall f1-score support
scope 0.7546 0.8146 0.7834 151
work 0.7368 0.8842 0.8038 95
pers 0.8000 0.9167 0.8544 96
loc 0.6667 0.6667 0.6667 3
date 0.0000 0.0000 0.0000 3
micro avg 0.7557 0.8534 0.8016 348
macro avg 0.5916 0.6564 0.6217 348
weighted avg 0.7550 0.8534 0.8008 348
2023-10-23 18:08:18,429 ----------------------------------------------------------------------------------------------------
|