File size: 24,101 Bytes
44a675e |
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 18:12:46,441 ----------------------------------------------------------------------------------------------------
2023-10-23 18:12:46,443 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 18:12:46,443 ----------------------------------------------------------------------------------------------------
2023-10-23 18:12:46,443 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 18:12:46,443 ----------------------------------------------------------------------------------------------------
2023-10-23 18:12:46,443 Train: 1214 sentences
2023-10-23 18:12:46,443 (train_with_dev=False, train_with_test=False)
2023-10-23 18:12:46,443 ----------------------------------------------------------------------------------------------------
2023-10-23 18:12:46,443 Training Params:
2023-10-23 18:12:46,443 - learning_rate: "3e-05"
2023-10-23 18:12:46,443 - mini_batch_size: "4"
2023-10-23 18:12:46,443 - max_epochs: "10"
2023-10-23 18:12:46,443 - shuffle: "True"
2023-10-23 18:12:46,443 ----------------------------------------------------------------------------------------------------
2023-10-23 18:12:46,443 Plugins:
2023-10-23 18:12:46,443 - TensorboardLogger
2023-10-23 18:12:46,443 - LinearScheduler | warmup_fraction: '0.1'
2023-10-23 18:12:46,443 ----------------------------------------------------------------------------------------------------
2023-10-23 18:12:46,444 Final evaluation on model from best epoch (best-model.pt)
2023-10-23 18:12:46,444 - metric: "('micro avg', 'f1-score')"
2023-10-23 18:12:46,444 ----------------------------------------------------------------------------------------------------
2023-10-23 18:12:46,444 Computation:
2023-10-23 18:12:46,444 - compute on device: cuda:0
2023-10-23 18:12:46,444 - embedding storage: none
2023-10-23 18:12:46,444 ----------------------------------------------------------------------------------------------------
2023-10-23 18:12:46,444 Model training base path: "hmbench-ajmc/en-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3"
2023-10-23 18:12:46,444 ----------------------------------------------------------------------------------------------------
2023-10-23 18:12:46,444 ----------------------------------------------------------------------------------------------------
2023-10-23 18:12:46,444 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-23 18:12:48,068 epoch 1 - iter 30/304 - loss 2.98445592 - time (sec): 1.62 - samples/sec: 1875.96 - lr: 0.000003 - momentum: 0.000000
2023-10-23 18:12:49,686 epoch 1 - iter 60/304 - loss 2.22348022 - time (sec): 3.24 - samples/sec: 1822.80 - lr: 0.000006 - momentum: 0.000000
2023-10-23 18:12:51,322 epoch 1 - iter 90/304 - loss 1.68960332 - time (sec): 4.88 - samples/sec: 1825.97 - lr: 0.000009 - momentum: 0.000000
2023-10-23 18:12:52,950 epoch 1 - iter 120/304 - loss 1.40192310 - time (sec): 6.51 - samples/sec: 1841.80 - lr: 0.000012 - momentum: 0.000000
2023-10-23 18:12:54,582 epoch 1 - iter 150/304 - loss 1.19239185 - time (sec): 8.14 - samples/sec: 1850.78 - lr: 0.000015 - momentum: 0.000000
2023-10-23 18:12:56,204 epoch 1 - iter 180/304 - loss 1.05158409 - time (sec): 9.76 - samples/sec: 1862.20 - lr: 0.000018 - momentum: 0.000000
2023-10-23 18:12:57,837 epoch 1 - iter 210/304 - loss 0.93428074 - time (sec): 11.39 - samples/sec: 1869.74 - lr: 0.000021 - momentum: 0.000000
2023-10-23 18:12:59,457 epoch 1 - iter 240/304 - loss 0.85197867 - time (sec): 13.01 - samples/sec: 1871.71 - lr: 0.000024 - momentum: 0.000000
2023-10-23 18:13:01,033 epoch 1 - iter 270/304 - loss 0.76874976 - time (sec): 14.59 - samples/sec: 1891.39 - lr: 0.000027 - momentum: 0.000000
2023-10-23 18:13:02,662 epoch 1 - iter 300/304 - loss 0.71452414 - time (sec): 16.22 - samples/sec: 1885.87 - lr: 0.000030 - momentum: 0.000000
2023-10-23 18:13:02,876 ----------------------------------------------------------------------------------------------------
2023-10-23 18:13:02,876 EPOCH 1 done: loss 0.7077 - lr: 0.000030
2023-10-23 18:13:03,837 DEV : loss 0.1715477705001831 - f1-score (micro avg) 0.6521
2023-10-23 18:13:03,846 saving best model
2023-10-23 18:13:04,262 ----------------------------------------------------------------------------------------------------
2023-10-23 18:13:05,887 epoch 2 - iter 30/304 - loss 0.16454595 - time (sec): 1.62 - samples/sec: 1908.65 - lr: 0.000030 - momentum: 0.000000
2023-10-23 18:13:07,528 epoch 2 - iter 60/304 - loss 0.14465478 - time (sec): 3.26 - samples/sec: 1930.97 - lr: 0.000029 - momentum: 0.000000
2023-10-23 18:13:09,167 epoch 2 - iter 90/304 - loss 0.13948154 - time (sec): 4.90 - samples/sec: 1877.81 - lr: 0.000029 - momentum: 0.000000
2023-10-23 18:13:10,811 epoch 2 - iter 120/304 - loss 0.13895752 - time (sec): 6.55 - samples/sec: 1864.80 - lr: 0.000029 - momentum: 0.000000
2023-10-23 18:13:12,406 epoch 2 - iter 150/304 - loss 0.14096606 - time (sec): 8.14 - samples/sec: 1857.62 - lr: 0.000028 - momentum: 0.000000
2023-10-23 18:13:14,052 epoch 2 - iter 180/304 - loss 0.13400557 - time (sec): 9.79 - samples/sec: 1860.88 - lr: 0.000028 - momentum: 0.000000
2023-10-23 18:13:15,685 epoch 2 - iter 210/304 - loss 0.13838672 - time (sec): 11.42 - samples/sec: 1833.38 - lr: 0.000028 - momentum: 0.000000
2023-10-23 18:13:17,338 epoch 2 - iter 240/304 - loss 0.13721522 - time (sec): 13.07 - samples/sec: 1849.68 - lr: 0.000027 - momentum: 0.000000
2023-10-23 18:13:18,976 epoch 2 - iter 270/304 - loss 0.13752883 - time (sec): 14.71 - samples/sec: 1855.39 - lr: 0.000027 - momentum: 0.000000
2023-10-23 18:13:20,607 epoch 2 - iter 300/304 - loss 0.13090121 - time (sec): 16.34 - samples/sec: 1874.02 - lr: 0.000027 - momentum: 0.000000
2023-10-23 18:13:20,823 ----------------------------------------------------------------------------------------------------
2023-10-23 18:13:20,823 EPOCH 2 done: loss 0.1302 - lr: 0.000027
2023-10-23 18:13:21,706 DEV : loss 0.16064199805259705 - f1-score (micro avg) 0.8138
2023-10-23 18:13:21,714 saving best model
2023-10-23 18:13:22,216 ----------------------------------------------------------------------------------------------------
2023-10-23 18:13:23,811 epoch 3 - iter 30/304 - loss 0.05475800 - time (sec): 1.59 - samples/sec: 1820.83 - lr: 0.000026 - momentum: 0.000000
2023-10-23 18:13:25,458 epoch 3 - iter 60/304 - loss 0.06229097 - time (sec): 3.24 - samples/sec: 1928.05 - lr: 0.000026 - momentum: 0.000000
2023-10-23 18:13:27,110 epoch 3 - iter 90/304 - loss 0.08416342 - time (sec): 4.89 - samples/sec: 1932.50 - lr: 0.000026 - momentum: 0.000000
2023-10-23 18:13:28,749 epoch 3 - iter 120/304 - loss 0.08385153 - time (sec): 6.53 - samples/sec: 1903.56 - lr: 0.000025 - momentum: 0.000000
2023-10-23 18:13:30,395 epoch 3 - iter 150/304 - loss 0.07573603 - time (sec): 8.18 - samples/sec: 1920.97 - lr: 0.000025 - momentum: 0.000000
2023-10-23 18:13:32,020 epoch 3 - iter 180/304 - loss 0.07699958 - time (sec): 9.80 - samples/sec: 1900.63 - lr: 0.000025 - momentum: 0.000000
2023-10-23 18:13:33,617 epoch 3 - iter 210/304 - loss 0.07824690 - time (sec): 11.40 - samples/sec: 1885.96 - lr: 0.000024 - momentum: 0.000000
2023-10-23 18:13:35,267 epoch 3 - iter 240/304 - loss 0.07568889 - time (sec): 13.05 - samples/sec: 1901.98 - lr: 0.000024 - momentum: 0.000000
2023-10-23 18:13:36,909 epoch 3 - iter 270/304 - loss 0.07464948 - time (sec): 14.69 - samples/sec: 1881.26 - lr: 0.000024 - momentum: 0.000000
2023-10-23 18:13:38,563 epoch 3 - iter 300/304 - loss 0.08036614 - time (sec): 16.35 - samples/sec: 1876.61 - lr: 0.000023 - momentum: 0.000000
2023-10-23 18:13:38,779 ----------------------------------------------------------------------------------------------------
2023-10-23 18:13:38,780 EPOCH 3 done: loss 0.0823 - lr: 0.000023
2023-10-23 18:13:39,794 DEV : loss 0.15495248138904572 - f1-score (micro avg) 0.8122
2023-10-23 18:13:39,801 ----------------------------------------------------------------------------------------------------
2023-10-23 18:13:41,439 epoch 4 - iter 30/304 - loss 0.04707249 - time (sec): 1.64 - samples/sec: 2059.62 - lr: 0.000023 - momentum: 0.000000
2023-10-23 18:13:43,080 epoch 4 - iter 60/304 - loss 0.04271187 - time (sec): 3.28 - samples/sec: 1983.84 - lr: 0.000023 - momentum: 0.000000
2023-10-23 18:13:44,719 epoch 4 - iter 90/304 - loss 0.05633248 - time (sec): 4.92 - samples/sec: 1895.38 - lr: 0.000022 - momentum: 0.000000
2023-10-23 18:13:46,363 epoch 4 - iter 120/304 - loss 0.04563425 - time (sec): 6.56 - samples/sec: 1903.34 - lr: 0.000022 - momentum: 0.000000
2023-10-23 18:13:48,003 epoch 4 - iter 150/304 - loss 0.04720628 - time (sec): 8.20 - samples/sec: 1887.96 - lr: 0.000022 - momentum: 0.000000
2023-10-23 18:13:49,652 epoch 4 - iter 180/304 - loss 0.05155600 - time (sec): 9.85 - samples/sec: 1882.05 - lr: 0.000021 - momentum: 0.000000
2023-10-23 18:13:51,294 epoch 4 - iter 210/304 - loss 0.05654426 - time (sec): 11.49 - samples/sec: 1883.09 - lr: 0.000021 - momentum: 0.000000
2023-10-23 18:13:52,942 epoch 4 - iter 240/304 - loss 0.05514680 - time (sec): 13.14 - samples/sec: 1876.42 - lr: 0.000021 - momentum: 0.000000
2023-10-23 18:13:54,584 epoch 4 - iter 270/304 - loss 0.05396017 - time (sec): 14.78 - samples/sec: 1872.14 - lr: 0.000020 - momentum: 0.000000
2023-10-23 18:13:56,220 epoch 4 - iter 300/304 - loss 0.05228844 - time (sec): 16.42 - samples/sec: 1863.03 - lr: 0.000020 - momentum: 0.000000
2023-10-23 18:13:56,435 ----------------------------------------------------------------------------------------------------
2023-10-23 18:13:56,436 EPOCH 4 done: loss 0.0518 - lr: 0.000020
2023-10-23 18:13:57,309 DEV : loss 0.19193391501903534 - f1-score (micro avg) 0.809
2023-10-23 18:13:57,316 ----------------------------------------------------------------------------------------------------
2023-10-23 18:13:58,960 epoch 5 - iter 30/304 - loss 0.03390289 - time (sec): 1.64 - samples/sec: 1913.03 - lr: 0.000020 - momentum: 0.000000
2023-10-23 18:14:00,600 epoch 5 - iter 60/304 - loss 0.04606974 - time (sec): 3.28 - samples/sec: 1858.67 - lr: 0.000019 - momentum: 0.000000
2023-10-23 18:14:02,257 epoch 5 - iter 90/304 - loss 0.04695528 - time (sec): 4.94 - samples/sec: 1892.28 - lr: 0.000019 - momentum: 0.000000
2023-10-23 18:14:03,883 epoch 5 - iter 120/304 - loss 0.04235888 - time (sec): 6.57 - samples/sec: 1880.87 - lr: 0.000019 - momentum: 0.000000
2023-10-23 18:14:05,526 epoch 5 - iter 150/304 - loss 0.04071583 - time (sec): 8.21 - samples/sec: 1911.76 - lr: 0.000018 - momentum: 0.000000
2023-10-23 18:14:07,153 epoch 5 - iter 180/304 - loss 0.03583492 - time (sec): 9.84 - samples/sec: 1899.00 - lr: 0.000018 - momentum: 0.000000
2023-10-23 18:14:08,767 epoch 5 - iter 210/304 - loss 0.03645264 - time (sec): 11.45 - samples/sec: 1894.36 - lr: 0.000018 - momentum: 0.000000
2023-10-23 18:14:10,387 epoch 5 - iter 240/304 - loss 0.03453267 - time (sec): 13.07 - samples/sec: 1891.71 - lr: 0.000017 - momentum: 0.000000
2023-10-23 18:14:12,015 epoch 5 - iter 270/304 - loss 0.03727819 - time (sec): 14.70 - samples/sec: 1878.18 - lr: 0.000017 - momentum: 0.000000
2023-10-23 18:14:13,650 epoch 5 - iter 300/304 - loss 0.03875291 - time (sec): 16.33 - samples/sec: 1869.00 - lr: 0.000017 - momentum: 0.000000
2023-10-23 18:14:13,866 ----------------------------------------------------------------------------------------------------
2023-10-23 18:14:13,866 EPOCH 5 done: loss 0.0393 - lr: 0.000017
2023-10-23 18:14:14,722 DEV : loss 0.20043757557868958 - f1-score (micro avg) 0.8349
2023-10-23 18:14:14,728 saving best model
2023-10-23 18:14:15,256 ----------------------------------------------------------------------------------------------------
2023-10-23 18:14:16,884 epoch 6 - iter 30/304 - loss 0.07028796 - time (sec): 1.63 - samples/sec: 1921.25 - lr: 0.000016 - momentum: 0.000000
2023-10-23 18:14:18,514 epoch 6 - iter 60/304 - loss 0.04781504 - time (sec): 3.26 - samples/sec: 1843.92 - lr: 0.000016 - momentum: 0.000000
2023-10-23 18:14:20,154 epoch 6 - iter 90/304 - loss 0.04475473 - time (sec): 4.90 - samples/sec: 1885.95 - lr: 0.000016 - momentum: 0.000000
2023-10-23 18:14:21,791 epoch 6 - iter 120/304 - loss 0.03578933 - time (sec): 6.53 - samples/sec: 1862.26 - lr: 0.000015 - momentum: 0.000000
2023-10-23 18:14:23,415 epoch 6 - iter 150/304 - loss 0.03712405 - time (sec): 8.16 - samples/sec: 1856.72 - lr: 0.000015 - momentum: 0.000000
2023-10-23 18:14:25,043 epoch 6 - iter 180/304 - loss 0.03222092 - time (sec): 9.79 - samples/sec: 1842.63 - lr: 0.000015 - momentum: 0.000000
2023-10-23 18:14:26,672 epoch 6 - iter 210/304 - loss 0.03272180 - time (sec): 11.41 - samples/sec: 1849.78 - lr: 0.000014 - momentum: 0.000000
2023-10-23 18:14:28,305 epoch 6 - iter 240/304 - loss 0.03111402 - time (sec): 13.05 - samples/sec: 1869.64 - lr: 0.000014 - momentum: 0.000000
2023-10-23 18:14:29,931 epoch 6 - iter 270/304 - loss 0.03162902 - time (sec): 14.67 - samples/sec: 1867.90 - lr: 0.000014 - momentum: 0.000000
2023-10-23 18:14:31,565 epoch 6 - iter 300/304 - loss 0.02944895 - time (sec): 16.31 - samples/sec: 1881.99 - lr: 0.000013 - momentum: 0.000000
2023-10-23 18:14:31,779 ----------------------------------------------------------------------------------------------------
2023-10-23 18:14:31,779 EPOCH 6 done: loss 0.0292 - lr: 0.000013
2023-10-23 18:14:32,621 DEV : loss 0.20006847381591797 - f1-score (micro avg) 0.8369
2023-10-23 18:14:32,628 saving best model
2023-10-23 18:14:33,142 ----------------------------------------------------------------------------------------------------
2023-10-23 18:14:34,780 epoch 7 - iter 30/304 - loss 0.01748835 - time (sec): 1.64 - samples/sec: 1916.44 - lr: 0.000013 - momentum: 0.000000
2023-10-23 18:14:36,421 epoch 7 - iter 60/304 - loss 0.01697280 - time (sec): 3.28 - samples/sec: 1880.73 - lr: 0.000013 - momentum: 0.000000
2023-10-23 18:14:38,068 epoch 7 - iter 90/304 - loss 0.01478209 - time (sec): 4.92 - samples/sec: 1888.36 - lr: 0.000012 - momentum: 0.000000
2023-10-23 18:14:39,726 epoch 7 - iter 120/304 - loss 0.01321084 - time (sec): 6.58 - samples/sec: 1912.13 - lr: 0.000012 - momentum: 0.000000
2023-10-23 18:14:41,368 epoch 7 - iter 150/304 - loss 0.01353593 - time (sec): 8.22 - samples/sec: 1893.45 - lr: 0.000012 - momentum: 0.000000
2023-10-23 18:14:43,000 epoch 7 - iter 180/304 - loss 0.01357105 - time (sec): 9.86 - samples/sec: 1888.99 - lr: 0.000011 - momentum: 0.000000
2023-10-23 18:14:44,619 epoch 7 - iter 210/304 - loss 0.01577979 - time (sec): 11.47 - samples/sec: 1880.71 - lr: 0.000011 - momentum: 0.000000
2023-10-23 18:14:46,231 epoch 7 - iter 240/304 - loss 0.01970489 - time (sec): 13.09 - samples/sec: 1863.38 - lr: 0.000011 - momentum: 0.000000
2023-10-23 18:14:47,836 epoch 7 - iter 270/304 - loss 0.02213724 - time (sec): 14.69 - samples/sec: 1892.70 - lr: 0.000010 - momentum: 0.000000
2023-10-23 18:14:49,364 epoch 7 - iter 300/304 - loss 0.02161108 - time (sec): 16.22 - samples/sec: 1892.57 - lr: 0.000010 - momentum: 0.000000
2023-10-23 18:14:49,574 ----------------------------------------------------------------------------------------------------
2023-10-23 18:14:49,575 EPOCH 7 done: loss 0.0214 - lr: 0.000010
2023-10-23 18:14:50,431 DEV : loss 0.20860883593559265 - f1-score (micro avg) 0.8347
2023-10-23 18:14:50,439 ----------------------------------------------------------------------------------------------------
2023-10-23 18:14:52,065 epoch 8 - iter 30/304 - loss 0.02968926 - time (sec): 1.63 - samples/sec: 1654.90 - lr: 0.000010 - momentum: 0.000000
2023-10-23 18:14:53,645 epoch 8 - iter 60/304 - loss 0.01406728 - time (sec): 3.21 - samples/sec: 1808.97 - lr: 0.000009 - momentum: 0.000000
2023-10-23 18:14:55,281 epoch 8 - iter 90/304 - loss 0.01843177 - time (sec): 4.84 - samples/sec: 1833.79 - lr: 0.000009 - momentum: 0.000000
2023-10-23 18:14:56,915 epoch 8 - iter 120/304 - loss 0.01640842 - time (sec): 6.47 - samples/sec: 1797.72 - lr: 0.000009 - momentum: 0.000000
2023-10-23 18:14:58,573 epoch 8 - iter 150/304 - loss 0.01538265 - time (sec): 8.13 - samples/sec: 1872.16 - lr: 0.000008 - momentum: 0.000000
2023-10-23 18:15:00,219 epoch 8 - iter 180/304 - loss 0.01963783 - time (sec): 9.78 - samples/sec: 1888.96 - lr: 0.000008 - momentum: 0.000000
2023-10-23 18:15:01,861 epoch 8 - iter 210/304 - loss 0.01959002 - time (sec): 11.42 - samples/sec: 1881.24 - lr: 0.000008 - momentum: 0.000000
2023-10-23 18:15:03,520 epoch 8 - iter 240/304 - loss 0.01694540 - time (sec): 13.08 - samples/sec: 1913.49 - lr: 0.000007 - momentum: 0.000000
2023-10-23 18:15:05,163 epoch 8 - iter 270/304 - loss 0.01613531 - time (sec): 14.72 - samples/sec: 1901.88 - lr: 0.000007 - momentum: 0.000000
2023-10-23 18:15:06,793 epoch 8 - iter 300/304 - loss 0.01531872 - time (sec): 16.35 - samples/sec: 1875.64 - lr: 0.000007 - momentum: 0.000000
2023-10-23 18:15:07,007 ----------------------------------------------------------------------------------------------------
2023-10-23 18:15:07,007 EPOCH 8 done: loss 0.0155 - lr: 0.000007
2023-10-23 18:15:07,895 DEV : loss 0.21332137286663055 - f1-score (micro avg) 0.8513
2023-10-23 18:15:07,902 saving best model
2023-10-23 18:15:08,377 ----------------------------------------------------------------------------------------------------
2023-10-23 18:15:10,015 epoch 9 - iter 30/304 - loss 0.02369919 - time (sec): 1.64 - samples/sec: 2021.13 - lr: 0.000006 - momentum: 0.000000
2023-10-23 18:15:11,655 epoch 9 - iter 60/304 - loss 0.01598962 - time (sec): 3.28 - samples/sec: 1910.25 - lr: 0.000006 - momentum: 0.000000
2023-10-23 18:15:13,275 epoch 9 - iter 90/304 - loss 0.01792024 - time (sec): 4.90 - samples/sec: 1926.43 - lr: 0.000006 - momentum: 0.000000
2023-10-23 18:15:14,901 epoch 9 - iter 120/304 - loss 0.01711638 - time (sec): 6.52 - samples/sec: 1841.30 - lr: 0.000005 - momentum: 0.000000
2023-10-23 18:15:16,537 epoch 9 - iter 150/304 - loss 0.01403237 - time (sec): 8.16 - samples/sec: 1879.29 - lr: 0.000005 - momentum: 0.000000
2023-10-23 18:15:18,162 epoch 9 - iter 180/304 - loss 0.01255566 - time (sec): 9.78 - samples/sec: 1869.43 - lr: 0.000005 - momentum: 0.000000
2023-10-23 18:15:19,810 epoch 9 - iter 210/304 - loss 0.01332711 - time (sec): 11.43 - samples/sec: 1869.47 - lr: 0.000004 - momentum: 0.000000
2023-10-23 18:15:21,460 epoch 9 - iter 240/304 - loss 0.01325790 - time (sec): 13.08 - samples/sec: 1863.74 - lr: 0.000004 - momentum: 0.000000
2023-10-23 18:15:23,108 epoch 9 - iter 270/304 - loss 0.01199154 - time (sec): 14.73 - samples/sec: 1867.08 - lr: 0.000004 - momentum: 0.000000
2023-10-23 18:15:24,755 epoch 9 - iter 300/304 - loss 0.01200321 - time (sec): 16.38 - samples/sec: 1870.93 - lr: 0.000003 - momentum: 0.000000
2023-10-23 18:15:24,972 ----------------------------------------------------------------------------------------------------
2023-10-23 18:15:24,973 EPOCH 9 done: loss 0.0119 - lr: 0.000003
2023-10-23 18:15:25,835 DEV : loss 0.21348616480827332 - f1-score (micro avg) 0.8415
2023-10-23 18:15:25,842 ----------------------------------------------------------------------------------------------------
2023-10-23 18:15:27,477 epoch 10 - iter 30/304 - loss 0.01226683 - time (sec): 1.63 - samples/sec: 1858.91 - lr: 0.000003 - momentum: 0.000000
2023-10-23 18:15:29,124 epoch 10 - iter 60/304 - loss 0.01317451 - time (sec): 3.28 - samples/sec: 1868.24 - lr: 0.000003 - momentum: 0.000000
2023-10-23 18:15:30,776 epoch 10 - iter 90/304 - loss 0.01065899 - time (sec): 4.93 - samples/sec: 1894.11 - lr: 0.000002 - momentum: 0.000000
2023-10-23 18:15:32,405 epoch 10 - iter 120/304 - loss 0.01158533 - time (sec): 6.56 - samples/sec: 1861.05 - lr: 0.000002 - momentum: 0.000000
2023-10-23 18:15:34,044 epoch 10 - iter 150/304 - loss 0.01081262 - time (sec): 8.20 - samples/sec: 1876.54 - lr: 0.000002 - momentum: 0.000000
2023-10-23 18:15:35,682 epoch 10 - iter 180/304 - loss 0.01070216 - time (sec): 9.84 - samples/sec: 1873.61 - lr: 0.000001 - momentum: 0.000000
2023-10-23 18:15:37,317 epoch 10 - iter 210/304 - loss 0.01196506 - time (sec): 11.47 - samples/sec: 1865.90 - lr: 0.000001 - momentum: 0.000000
2023-10-23 18:15:38,958 epoch 10 - iter 240/304 - loss 0.01186978 - time (sec): 13.12 - samples/sec: 1888.09 - lr: 0.000001 - momentum: 0.000000
2023-10-23 18:15:40,534 epoch 10 - iter 270/304 - loss 0.01092989 - time (sec): 14.69 - samples/sec: 1879.86 - lr: 0.000000 - momentum: 0.000000
2023-10-23 18:15:42,171 epoch 10 - iter 300/304 - loss 0.00998623 - time (sec): 16.33 - samples/sec: 1874.49 - lr: 0.000000 - momentum: 0.000000
2023-10-23 18:15:42,387 ----------------------------------------------------------------------------------------------------
2023-10-23 18:15:42,387 EPOCH 10 done: loss 0.0099 - lr: 0.000000
2023-10-23 18:15:43,251 DEV : loss 0.216822549700737 - f1-score (micro avg) 0.8431
2023-10-23 18:15:43,667 ----------------------------------------------------------------------------------------------------
2023-10-23 18:15:43,669 Loading model from best epoch ...
2023-10-23 18:15:45,445 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 18:15:46,251
Results:
- F-score (micro) 0.7967
- F-score (macro) 0.6456
- Accuracy 0.6728
By class:
precision recall f1-score support
scope 0.7407 0.7947 0.7668 151
pers 0.7459 0.9479 0.8349 96
work 0.8020 0.8526 0.8265 95
date 0.0000 0.0000 0.0000 3
loc 1.0000 0.6667 0.8000 3
micro avg 0.7538 0.8448 0.7967 348
macro avg 0.6577 0.6524 0.6456 348
weighted avg 0.7547 0.8448 0.7955 348
2023-10-23 18:15:46,251 ----------------------------------------------------------------------------------------------------
|