File size: 24,059 Bytes
e807e5b |
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 |
2023-10-23 18:16:14,534 ----------------------------------------------------------------------------------------------------
2023-10-23 18:16:14,535 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 18:16:14,535 ----------------------------------------------------------------------------------------------------
2023-10-23 18:16:14,535 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 18:16:14,535 ----------------------------------------------------------------------------------------------------
2023-10-23 18:16:14,535 Train: 1214 sentences
2023-10-23 18:16:14,536 (train_with_dev=False, train_with_test=False)
2023-10-23 18:16:14,536 ----------------------------------------------------------------------------------------------------
2023-10-23 18:16:14,536 Training Params:
2023-10-23 18:16:14,536 - learning_rate: "5e-05"
2023-10-23 18:16:14,536 - mini_batch_size: "4"
2023-10-23 18:16:14,536 - max_epochs: "10"
2023-10-23 18:16:14,536 - shuffle: "True"
2023-10-23 18:16:14,536 ----------------------------------------------------------------------------------------------------
2023-10-23 18:16:14,536 Plugins:
2023-10-23 18:16:14,536 - TensorboardLogger
2023-10-23 18:16:14,536 - LinearScheduler | warmup_fraction: '0.1'
2023-10-23 18:16:14,536 ----------------------------------------------------------------------------------------------------
2023-10-23 18:16:14,536 Final evaluation on model from best epoch (best-model.pt)
2023-10-23 18:16:14,536 - metric: "('micro avg', 'f1-score')"
2023-10-23 18:16:14,536 ----------------------------------------------------------------------------------------------------
2023-10-23 18:16:14,536 Computation:
2023-10-23 18:16:14,536 - compute on device: cuda:0
2023-10-23 18:16:14,536 - embedding storage: none
2023-10-23 18:16:14,536 ----------------------------------------------------------------------------------------------------
2023-10-23 18:16:14,536 Model training base path: "hmbench-ajmc/en-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3"
2023-10-23 18:16:14,536 ----------------------------------------------------------------------------------------------------
2023-10-23 18:16:14,536 ----------------------------------------------------------------------------------------------------
2023-10-23 18:16:14,537 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-23 18:16:16,168 epoch 1 - iter 30/304 - loss 2.81106393 - time (sec): 1.63 - samples/sec: 1868.20 - lr: 0.000005 - momentum: 0.000000
2023-10-23 18:16:17,803 epoch 1 - iter 60/304 - loss 1.93076019 - time (sec): 3.26 - samples/sec: 1809.64 - lr: 0.000010 - momentum: 0.000000
2023-10-23 18:16:19,444 epoch 1 - iter 90/304 - loss 1.46147412 - time (sec): 4.91 - samples/sec: 1814.82 - lr: 0.000015 - momentum: 0.000000
2023-10-23 18:16:21,083 epoch 1 - iter 120/304 - loss 1.20065349 - time (sec): 6.54 - samples/sec: 1830.61 - lr: 0.000020 - momentum: 0.000000
2023-10-23 18:16:22,722 epoch 1 - iter 150/304 - loss 1.01442418 - time (sec): 8.18 - samples/sec: 1840.20 - lr: 0.000025 - momentum: 0.000000
2023-10-23 18:16:24,360 epoch 1 - iter 180/304 - loss 0.89232529 - time (sec): 9.82 - samples/sec: 1850.31 - lr: 0.000029 - momentum: 0.000000
2023-10-23 18:16:25,998 epoch 1 - iter 210/304 - loss 0.78694129 - time (sec): 11.46 - samples/sec: 1858.68 - lr: 0.000034 - momentum: 0.000000
2023-10-23 18:16:27,642 epoch 1 - iter 240/304 - loss 0.71551296 - time (sec): 13.10 - samples/sec: 1858.63 - lr: 0.000039 - momentum: 0.000000
2023-10-23 18:16:29,280 epoch 1 - iter 270/304 - loss 0.64584501 - time (sec): 14.74 - samples/sec: 1871.71 - lr: 0.000044 - momentum: 0.000000
2023-10-23 18:16:30,918 epoch 1 - iter 300/304 - loss 0.60271094 - time (sec): 16.38 - samples/sec: 1866.99 - lr: 0.000049 - momentum: 0.000000
2023-10-23 18:16:31,135 ----------------------------------------------------------------------------------------------------
2023-10-23 18:16:31,135 EPOCH 1 done: loss 0.5980 - lr: 0.000049
2023-10-23 18:16:31,913 DEV : loss 0.16384108364582062 - f1-score (micro avg) 0.704
2023-10-23 18:16:31,921 saving best model
2023-10-23 18:16:32,319 ----------------------------------------------------------------------------------------------------
2023-10-23 18:16:33,933 epoch 2 - iter 30/304 - loss 0.13187891 - time (sec): 1.61 - samples/sec: 1921.94 - lr: 0.000049 - momentum: 0.000000
2023-10-23 18:16:35,566 epoch 2 - iter 60/304 - loss 0.12737012 - time (sec): 3.25 - samples/sec: 1942.32 - lr: 0.000049 - momentum: 0.000000
2023-10-23 18:16:37,200 epoch 2 - iter 90/304 - loss 0.12304473 - time (sec): 4.88 - samples/sec: 1887.11 - lr: 0.000048 - momentum: 0.000000
2023-10-23 18:16:38,830 epoch 2 - iter 120/304 - loss 0.12231441 - time (sec): 6.51 - samples/sec: 1875.57 - lr: 0.000048 - momentum: 0.000000
2023-10-23 18:16:40,451 epoch 2 - iter 150/304 - loss 0.12142969 - time (sec): 8.13 - samples/sec: 1860.22 - lr: 0.000047 - momentum: 0.000000
2023-10-23 18:16:42,079 epoch 2 - iter 180/304 - loss 0.12168571 - time (sec): 9.76 - samples/sec: 1866.79 - lr: 0.000047 - momentum: 0.000000
2023-10-23 18:16:43,699 epoch 2 - iter 210/304 - loss 0.12781452 - time (sec): 11.38 - samples/sec: 1840.27 - lr: 0.000046 - momentum: 0.000000
2023-10-23 18:16:45,340 epoch 2 - iter 240/304 - loss 0.12865335 - time (sec): 13.02 - samples/sec: 1857.53 - lr: 0.000046 - momentum: 0.000000
2023-10-23 18:16:46,971 epoch 2 - iter 270/304 - loss 0.12960732 - time (sec): 14.65 - samples/sec: 1863.18 - lr: 0.000045 - momentum: 0.000000
2023-10-23 18:16:48,610 epoch 2 - iter 300/304 - loss 0.12332332 - time (sec): 16.29 - samples/sec: 1880.23 - lr: 0.000045 - momentum: 0.000000
2023-10-23 18:16:48,825 ----------------------------------------------------------------------------------------------------
2023-10-23 18:16:48,825 EPOCH 2 done: loss 0.1226 - lr: 0.000045
2023-10-23 18:16:49,711 DEV : loss 0.14167819917201996 - f1-score (micro avg) 0.8188
2023-10-23 18:16:49,718 saving best model
2023-10-23 18:16:50,248 ----------------------------------------------------------------------------------------------------
2023-10-23 18:16:51,875 epoch 3 - iter 30/304 - loss 0.07030755 - time (sec): 1.63 - samples/sec: 1785.11 - lr: 0.000044 - momentum: 0.000000
2023-10-23 18:16:53,518 epoch 3 - iter 60/304 - loss 0.07494672 - time (sec): 3.27 - samples/sec: 1911.69 - lr: 0.000043 - momentum: 0.000000
2023-10-23 18:16:55,156 epoch 3 - iter 90/304 - loss 0.08781157 - time (sec): 4.91 - samples/sec: 1927.02 - lr: 0.000043 - momentum: 0.000000
2023-10-23 18:16:56,786 epoch 3 - iter 120/304 - loss 0.08801500 - time (sec): 6.54 - samples/sec: 1902.16 - lr: 0.000042 - momentum: 0.000000
2023-10-23 18:16:58,422 epoch 3 - iter 150/304 - loss 0.08096093 - time (sec): 8.17 - samples/sec: 1922.15 - lr: 0.000042 - momentum: 0.000000
2023-10-23 18:17:00,061 epoch 3 - iter 180/304 - loss 0.08068663 - time (sec): 9.81 - samples/sec: 1898.78 - lr: 0.000041 - momentum: 0.000000
2023-10-23 18:17:01,697 epoch 3 - iter 210/304 - loss 0.08424806 - time (sec): 11.45 - samples/sec: 1878.13 - lr: 0.000041 - momentum: 0.000000
2023-10-23 18:17:03,312 epoch 3 - iter 240/304 - loss 0.08231945 - time (sec): 13.06 - samples/sec: 1900.07 - lr: 0.000040 - momentum: 0.000000
2023-10-23 18:17:04,951 epoch 3 - iter 270/304 - loss 0.07987312 - time (sec): 14.70 - samples/sec: 1880.03 - lr: 0.000040 - momentum: 0.000000
2023-10-23 18:17:06,596 epoch 3 - iter 300/304 - loss 0.08623195 - time (sec): 16.35 - samples/sec: 1876.47 - lr: 0.000039 - momentum: 0.000000
2023-10-23 18:17:06,810 ----------------------------------------------------------------------------------------------------
2023-10-23 18:17:06,810 EPOCH 3 done: loss 0.0874 - lr: 0.000039
2023-10-23 18:17:07,670 DEV : loss 0.1868170201778412 - f1-score (micro avg) 0.8171
2023-10-23 18:17:07,678 ----------------------------------------------------------------------------------------------------
2023-10-23 18:17:09,306 epoch 4 - iter 30/304 - loss 0.03795868 - time (sec): 1.63 - samples/sec: 2070.92 - lr: 0.000038 - momentum: 0.000000
2023-10-23 18:17:10,945 epoch 4 - iter 60/304 - loss 0.03436106 - time (sec): 3.27 - samples/sec: 1990.77 - lr: 0.000038 - momentum: 0.000000
2023-10-23 18:17:12,583 epoch 4 - iter 90/304 - loss 0.05396697 - time (sec): 4.90 - samples/sec: 1900.22 - lr: 0.000037 - momentum: 0.000000
2023-10-23 18:17:14,225 epoch 4 - iter 120/304 - loss 0.04411729 - time (sec): 6.55 - samples/sec: 1907.64 - lr: 0.000037 - momentum: 0.000000
2023-10-23 18:17:15,858 epoch 4 - iter 150/304 - loss 0.04620010 - time (sec): 8.18 - samples/sec: 1892.75 - lr: 0.000036 - momentum: 0.000000
2023-10-23 18:17:17,502 epoch 4 - iter 180/304 - loss 0.05291311 - time (sec): 9.82 - samples/sec: 1887.18 - lr: 0.000036 - momentum: 0.000000
2023-10-23 18:17:19,138 epoch 4 - iter 210/304 - loss 0.06392533 - time (sec): 11.46 - samples/sec: 1888.42 - lr: 0.000035 - momentum: 0.000000
2023-10-23 18:17:20,771 epoch 4 - iter 240/304 - loss 0.06336046 - time (sec): 13.09 - samples/sec: 1883.10 - lr: 0.000035 - momentum: 0.000000
2023-10-23 18:17:22,404 epoch 4 - iter 270/304 - loss 0.06184609 - time (sec): 14.73 - samples/sec: 1879.24 - lr: 0.000034 - momentum: 0.000000
2023-10-23 18:17:24,030 epoch 4 - iter 300/304 - loss 0.05887253 - time (sec): 16.35 - samples/sec: 1870.50 - lr: 0.000033 - momentum: 0.000000
2023-10-23 18:17:24,245 ----------------------------------------------------------------------------------------------------
2023-10-23 18:17:24,246 EPOCH 4 done: loss 0.0581 - lr: 0.000033
2023-10-23 18:17:25,250 DEV : loss 0.1909182071685791 - f1-score (micro avg) 0.8271
2023-10-23 18:17:25,257 saving best model
2023-10-23 18:17:25,783 ----------------------------------------------------------------------------------------------------
2023-10-23 18:17:27,406 epoch 5 - iter 30/304 - loss 0.01744137 - time (sec): 1.62 - samples/sec: 1937.30 - lr: 0.000033 - momentum: 0.000000
2023-10-23 18:17:29,031 epoch 5 - iter 60/304 - loss 0.03338940 - time (sec): 3.25 - samples/sec: 1879.20 - lr: 0.000032 - momentum: 0.000000
2023-10-23 18:17:30,666 epoch 5 - iter 90/304 - loss 0.04690048 - time (sec): 4.88 - samples/sec: 1914.80 - lr: 0.000032 - momentum: 0.000000
2023-10-23 18:17:32,300 epoch 5 - iter 120/304 - loss 0.04346722 - time (sec): 6.52 - samples/sec: 1895.35 - lr: 0.000031 - momentum: 0.000000
2023-10-23 18:17:33,937 epoch 5 - iter 150/304 - loss 0.04056557 - time (sec): 8.15 - samples/sec: 1924.73 - lr: 0.000031 - momentum: 0.000000
2023-10-23 18:17:35,568 epoch 5 - iter 180/304 - loss 0.03650648 - time (sec): 9.78 - samples/sec: 1908.95 - lr: 0.000030 - momentum: 0.000000
2023-10-23 18:17:37,208 epoch 5 - iter 210/304 - loss 0.03565747 - time (sec): 11.42 - samples/sec: 1898.72 - lr: 0.000030 - momentum: 0.000000
2023-10-23 18:17:38,842 epoch 5 - iter 240/304 - loss 0.03624900 - time (sec): 13.06 - samples/sec: 1893.31 - lr: 0.000029 - momentum: 0.000000
2023-10-23 18:17:40,473 epoch 5 - iter 270/304 - loss 0.03997973 - time (sec): 14.69 - samples/sec: 1879.27 - lr: 0.000028 - momentum: 0.000000
2023-10-23 18:17:41,959 epoch 5 - iter 300/304 - loss 0.04238227 - time (sec): 16.17 - samples/sec: 1887.18 - lr: 0.000028 - momentum: 0.000000
2023-10-23 18:17:42,137 ----------------------------------------------------------------------------------------------------
2023-10-23 18:17:42,137 EPOCH 5 done: loss 0.0438 - lr: 0.000028
2023-10-23 18:17:42,996 DEV : loss 0.20630058646202087 - f1-score (micro avg) 0.8014
2023-10-23 18:17:43,003 ----------------------------------------------------------------------------------------------------
2023-10-23 18:17:44,652 epoch 6 - iter 30/304 - loss 0.07524903 - time (sec): 1.65 - samples/sec: 1896.84 - lr: 0.000027 - momentum: 0.000000
2023-10-23 18:17:46,280 epoch 6 - iter 60/304 - loss 0.05025896 - time (sec): 3.28 - samples/sec: 1832.90 - lr: 0.000027 - momentum: 0.000000
2023-10-23 18:17:47,919 epoch 6 - iter 90/304 - loss 0.04756550 - time (sec): 4.91 - samples/sec: 1879.40 - lr: 0.000026 - momentum: 0.000000
2023-10-23 18:17:49,557 epoch 6 - iter 120/304 - loss 0.03855495 - time (sec): 6.55 - samples/sec: 1856.87 - lr: 0.000026 - momentum: 0.000000
2023-10-23 18:17:51,183 epoch 6 - iter 150/304 - loss 0.04185719 - time (sec): 8.18 - samples/sec: 1851.78 - lr: 0.000025 - momentum: 0.000000
2023-10-23 18:17:52,810 epoch 6 - iter 180/304 - loss 0.03650979 - time (sec): 9.81 - samples/sec: 1838.74 - lr: 0.000025 - momentum: 0.000000
2023-10-23 18:17:54,443 epoch 6 - iter 210/304 - loss 0.03600914 - time (sec): 11.44 - samples/sec: 1845.91 - lr: 0.000024 - momentum: 0.000000
2023-10-23 18:17:56,084 epoch 6 - iter 240/304 - loss 0.03276981 - time (sec): 13.08 - samples/sec: 1865.06 - lr: 0.000023 - momentum: 0.000000
2023-10-23 18:17:57,720 epoch 6 - iter 270/304 - loss 0.03306713 - time (sec): 14.72 - samples/sec: 1862.62 - lr: 0.000023 - momentum: 0.000000
2023-10-23 18:17:59,360 epoch 6 - iter 300/304 - loss 0.03105096 - time (sec): 16.36 - samples/sec: 1876.42 - lr: 0.000022 - momentum: 0.000000
2023-10-23 18:17:59,574 ----------------------------------------------------------------------------------------------------
2023-10-23 18:17:59,574 EPOCH 6 done: loss 0.0308 - lr: 0.000022
2023-10-23 18:18:00,427 DEV : loss 0.22646591067314148 - f1-score (micro avg) 0.8223
2023-10-23 18:18:00,434 ----------------------------------------------------------------------------------------------------
2023-10-23 18:18:02,066 epoch 7 - iter 30/304 - loss 0.02843019 - time (sec): 1.63 - samples/sec: 1922.19 - lr: 0.000022 - momentum: 0.000000
2023-10-23 18:18:03,702 epoch 7 - iter 60/304 - loss 0.02264544 - time (sec): 3.27 - samples/sec: 1886.65 - lr: 0.000021 - momentum: 0.000000
2023-10-23 18:18:05,338 epoch 7 - iter 90/304 - loss 0.02267803 - time (sec): 4.90 - samples/sec: 1896.52 - lr: 0.000021 - momentum: 0.000000
2023-10-23 18:18:06,978 epoch 7 - iter 120/304 - loss 0.02033071 - time (sec): 6.54 - samples/sec: 1923.53 - lr: 0.000020 - momentum: 0.000000
2023-10-23 18:18:08,606 epoch 7 - iter 150/304 - loss 0.01831766 - time (sec): 8.17 - samples/sec: 1905.74 - lr: 0.000020 - momentum: 0.000000
2023-10-23 18:18:10,240 epoch 7 - iter 180/304 - loss 0.01927033 - time (sec): 9.80 - samples/sec: 1898.80 - lr: 0.000019 - momentum: 0.000000
2023-10-23 18:18:11,864 epoch 7 - iter 210/304 - loss 0.02071687 - time (sec): 11.43 - samples/sec: 1888.28 - lr: 0.000018 - momentum: 0.000000
2023-10-23 18:18:13,494 epoch 7 - iter 240/304 - loss 0.02263808 - time (sec): 13.06 - samples/sec: 1867.41 - lr: 0.000018 - momentum: 0.000000
2023-10-23 18:18:15,133 epoch 7 - iter 270/304 - loss 0.02451109 - time (sec): 14.70 - samples/sec: 1891.99 - lr: 0.000017 - momentum: 0.000000
2023-10-23 18:18:16,761 epoch 7 - iter 300/304 - loss 0.02486620 - time (sec): 16.33 - samples/sec: 1880.23 - lr: 0.000017 - momentum: 0.000000
2023-10-23 18:18:16,975 ----------------------------------------------------------------------------------------------------
2023-10-23 18:18:16,975 EPOCH 7 done: loss 0.0246 - lr: 0.000017
2023-10-23 18:18:17,816 DEV : loss 0.20806437730789185 - f1-score (micro avg) 0.8262
2023-10-23 18:18:17,823 ----------------------------------------------------------------------------------------------------
2023-10-23 18:18:19,440 epoch 8 - iter 30/304 - loss 0.03017504 - time (sec): 1.62 - samples/sec: 1665.61 - lr: 0.000016 - momentum: 0.000000
2023-10-23 18:18:21,061 epoch 8 - iter 60/304 - loss 0.01475323 - time (sec): 3.24 - samples/sec: 1791.60 - lr: 0.000016 - momentum: 0.000000
2023-10-23 18:18:22,689 epoch 8 - iter 90/304 - loss 0.01529115 - time (sec): 4.86 - samples/sec: 1824.73 - lr: 0.000015 - momentum: 0.000000
2023-10-23 18:18:24,301 epoch 8 - iter 120/304 - loss 0.01367787 - time (sec): 6.48 - samples/sec: 1797.21 - lr: 0.000015 - momentum: 0.000000
2023-10-23 18:18:25,946 epoch 8 - iter 150/304 - loss 0.01317154 - time (sec): 8.12 - samples/sec: 1874.76 - lr: 0.000014 - momentum: 0.000000
2023-10-23 18:18:27,541 epoch 8 - iter 180/304 - loss 0.01616983 - time (sec): 9.72 - samples/sec: 1901.04 - lr: 0.000013 - momentum: 0.000000
2023-10-23 18:18:29,175 epoch 8 - iter 210/304 - loss 0.01609663 - time (sec): 11.35 - samples/sec: 1892.89 - lr: 0.000013 - momentum: 0.000000
2023-10-23 18:18:30,814 epoch 8 - iter 240/304 - loss 0.01486684 - time (sec): 12.99 - samples/sec: 1926.77 - lr: 0.000012 - momentum: 0.000000
2023-10-23 18:18:32,438 epoch 8 - iter 270/304 - loss 0.01438102 - time (sec): 14.61 - samples/sec: 1916.14 - lr: 0.000012 - momentum: 0.000000
2023-10-23 18:18:34,068 epoch 8 - iter 300/304 - loss 0.01467520 - time (sec): 16.24 - samples/sec: 1888.31 - lr: 0.000011 - momentum: 0.000000
2023-10-23 18:18:34,281 ----------------------------------------------------------------------------------------------------
2023-10-23 18:18:34,281 EPOCH 8 done: loss 0.0145 - lr: 0.000011
2023-10-23 18:18:35,114 DEV : loss 0.21577374637126923 - f1-score (micro avg) 0.8517
2023-10-23 18:18:35,121 saving best model
2023-10-23 18:18:35,654 ----------------------------------------------------------------------------------------------------
2023-10-23 18:18:37,296 epoch 9 - iter 30/304 - loss 0.02217251 - time (sec): 1.64 - samples/sec: 2016.73 - lr: 0.000011 - momentum: 0.000000
2023-10-23 18:18:38,925 epoch 9 - iter 60/304 - loss 0.01548730 - time (sec): 3.27 - samples/sec: 1914.58 - lr: 0.000010 - momentum: 0.000000
2023-10-23 18:18:40,571 epoch 9 - iter 90/304 - loss 0.01781690 - time (sec): 4.91 - samples/sec: 1919.16 - lr: 0.000010 - momentum: 0.000000
2023-10-23 18:18:42,193 epoch 9 - iter 120/304 - loss 0.01604254 - time (sec): 6.54 - samples/sec: 1837.29 - lr: 0.000009 - momentum: 0.000000
2023-10-23 18:18:43,837 epoch 9 - iter 150/304 - loss 0.01483913 - time (sec): 8.18 - samples/sec: 1874.20 - lr: 0.000008 - momentum: 0.000000
2023-10-23 18:18:45,472 epoch 9 - iter 180/304 - loss 0.01278405 - time (sec): 9.82 - samples/sec: 1863.31 - lr: 0.000008 - momentum: 0.000000
2023-10-23 18:18:47,113 epoch 9 - iter 210/304 - loss 0.01183613 - time (sec): 11.46 - samples/sec: 1865.42 - lr: 0.000007 - momentum: 0.000000
2023-10-23 18:18:48,744 epoch 9 - iter 240/304 - loss 0.01173901 - time (sec): 13.09 - samples/sec: 1862.89 - lr: 0.000007 - momentum: 0.000000
2023-10-23 18:18:50,377 epoch 9 - iter 270/304 - loss 0.01107654 - time (sec): 14.72 - samples/sec: 1868.29 - lr: 0.000006 - momentum: 0.000000
2023-10-23 18:18:52,020 epoch 9 - iter 300/304 - loss 0.01093778 - time (sec): 16.36 - samples/sec: 1872.44 - lr: 0.000006 - momentum: 0.000000
2023-10-23 18:18:52,237 ----------------------------------------------------------------------------------------------------
2023-10-23 18:18:52,237 EPOCH 9 done: loss 0.0108 - lr: 0.000006
2023-10-23 18:18:53,118 DEV : loss 0.2204783409833908 - f1-score (micro avg) 0.8473
2023-10-23 18:18:53,125 ----------------------------------------------------------------------------------------------------
2023-10-23 18:18:54,762 epoch 10 - iter 30/304 - loss 0.00097800 - time (sec): 1.64 - samples/sec: 1856.40 - lr: 0.000005 - momentum: 0.000000
2023-10-23 18:18:56,397 epoch 10 - iter 60/304 - loss 0.00905964 - time (sec): 3.27 - samples/sec: 1873.85 - lr: 0.000005 - momentum: 0.000000
2023-10-23 18:18:58,043 epoch 10 - iter 90/304 - loss 0.00780886 - time (sec): 4.92 - samples/sec: 1900.28 - lr: 0.000004 - momentum: 0.000000
2023-10-23 18:18:59,671 epoch 10 - iter 120/304 - loss 0.00866515 - time (sec): 6.55 - samples/sec: 1865.78 - lr: 0.000003 - momentum: 0.000000
2023-10-23 18:19:01,310 epoch 10 - iter 150/304 - loss 0.00766781 - time (sec): 8.18 - samples/sec: 1880.41 - lr: 0.000003 - momentum: 0.000000
2023-10-23 18:19:02,940 epoch 10 - iter 180/304 - loss 0.00744633 - time (sec): 9.81 - samples/sec: 1878.27 - lr: 0.000002 - momentum: 0.000000
2023-10-23 18:19:04,571 epoch 10 - iter 210/304 - loss 0.00780657 - time (sec): 11.44 - samples/sec: 1870.69 - lr: 0.000002 - momentum: 0.000000
2023-10-23 18:19:06,209 epoch 10 - iter 240/304 - loss 0.00784100 - time (sec): 13.08 - samples/sec: 1892.77 - lr: 0.000001 - momentum: 0.000000
2023-10-23 18:19:07,839 epoch 10 - iter 270/304 - loss 0.00724030 - time (sec): 14.71 - samples/sec: 1877.08 - lr: 0.000001 - momentum: 0.000000
2023-10-23 18:19:09,472 epoch 10 - iter 300/304 - loss 0.00659561 - time (sec): 16.35 - samples/sec: 1872.36 - lr: 0.000000 - momentum: 0.000000
2023-10-23 18:19:09,686 ----------------------------------------------------------------------------------------------------
2023-10-23 18:19:09,686 EPOCH 10 done: loss 0.0065 - lr: 0.000000
2023-10-23 18:19:10,517 DEV : loss 0.22362594306468964 - f1-score (micro avg) 0.8427
2023-10-23 18:19:10,914 ----------------------------------------------------------------------------------------------------
2023-10-23 18:19:10,915 Loading model from best epoch ...
2023-10-23 18:19:12,690 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 18:19:13,529
Results:
- F-score (micro) 0.8081
- F-score (macro) 0.7142
- Accuracy 0.6874
By class:
precision recall f1-score support
scope 0.8013 0.8278 0.8143 151
work 0.6855 0.8947 0.7763 95
pers 0.8037 0.8958 0.8473 96
date 0.3333 0.3333 0.3333 3
loc 1.0000 0.6667 0.8000 3
micro avg 0.7628 0.8592 0.8081 348
macro avg 0.7248 0.7237 0.7142 348
weighted avg 0.7680 0.8592 0.8088 348
2023-10-23 18:19:13,529 ----------------------------------------------------------------------------------------------------
|