File size: 24,015 Bytes
44ff2ed |
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 |
2023-10-13 08:43:02,471 ----------------------------------------------------------------------------------------------------
2023-10-13 08:43:02,472 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(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): 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()
)"
2023-10-13 08:43:02,472 ----------------------------------------------------------------------------------------------------
2023-10-13 08:43:02,472 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
2023-10-13 08:43:02,472 ----------------------------------------------------------------------------------------------------
2023-10-13 08:43:02,472 Train: 1100 sentences
2023-10-13 08:43:02,472 (train_with_dev=False, train_with_test=False)
2023-10-13 08:43:02,472 ----------------------------------------------------------------------------------------------------
2023-10-13 08:43:02,472 Training Params:
2023-10-13 08:43:02,472 - learning_rate: "5e-05"
2023-10-13 08:43:02,472 - mini_batch_size: "4"
2023-10-13 08:43:02,472 - max_epochs: "10"
2023-10-13 08:43:02,472 - shuffle: "True"
2023-10-13 08:43:02,472 ----------------------------------------------------------------------------------------------------
2023-10-13 08:43:02,473 Plugins:
2023-10-13 08:43:02,473 - LinearScheduler | warmup_fraction: '0.1'
2023-10-13 08:43:02,473 ----------------------------------------------------------------------------------------------------
2023-10-13 08:43:02,473 Final evaluation on model from best epoch (best-model.pt)
2023-10-13 08:43:02,473 - metric: "('micro avg', 'f1-score')"
2023-10-13 08:43:02,473 ----------------------------------------------------------------------------------------------------
2023-10-13 08:43:02,473 Computation:
2023-10-13 08:43:02,473 - compute on device: cuda:0
2023-10-13 08:43:02,473 - embedding storage: none
2023-10-13 08:43:02,473 ----------------------------------------------------------------------------------------------------
2023-10-13 08:43:02,473 Model training base path: "hmbench-ajmc/de-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-13 08:43:02,473 ----------------------------------------------------------------------------------------------------
2023-10-13 08:43:02,473 ----------------------------------------------------------------------------------------------------
2023-10-13 08:43:03,736 epoch 1 - iter 27/275 - loss 3.26762455 - time (sec): 1.26 - samples/sec: 1709.25 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:43:04,936 epoch 1 - iter 54/275 - loss 2.61858234 - time (sec): 2.46 - samples/sec: 1794.70 - lr: 0.000010 - momentum: 0.000000
2023-10-13 08:43:06,121 epoch 1 - iter 81/275 - loss 2.06451088 - time (sec): 3.65 - samples/sec: 1808.60 - lr: 0.000015 - momentum: 0.000000
2023-10-13 08:43:07,305 epoch 1 - iter 108/275 - loss 1.70976718 - time (sec): 4.83 - samples/sec: 1817.03 - lr: 0.000019 - momentum: 0.000000
2023-10-13 08:43:08,553 epoch 1 - iter 135/275 - loss 1.46718450 - time (sec): 6.08 - samples/sec: 1837.67 - lr: 0.000024 - momentum: 0.000000
2023-10-13 08:43:09,740 epoch 1 - iter 162/275 - loss 1.29811689 - time (sec): 7.27 - samples/sec: 1839.88 - lr: 0.000029 - momentum: 0.000000
2023-10-13 08:43:10,956 epoch 1 - iter 189/275 - loss 1.15204599 - time (sec): 8.48 - samples/sec: 1853.59 - lr: 0.000034 - momentum: 0.000000
2023-10-13 08:43:12,195 epoch 1 - iter 216/275 - loss 1.04030680 - time (sec): 9.72 - samples/sec: 1855.80 - lr: 0.000039 - momentum: 0.000000
2023-10-13 08:43:13,467 epoch 1 - iter 243/275 - loss 0.96060765 - time (sec): 10.99 - samples/sec: 1834.61 - lr: 0.000044 - momentum: 0.000000
2023-10-13 08:43:14,708 epoch 1 - iter 270/275 - loss 0.89047507 - time (sec): 12.23 - samples/sec: 1822.47 - lr: 0.000049 - momentum: 0.000000
2023-10-13 08:43:14,957 ----------------------------------------------------------------------------------------------------
2023-10-13 08:43:14,957 EPOCH 1 done: loss 0.8777 - lr: 0.000049
2023-10-13 08:43:15,936 DEV : loss 0.2361612170934677 - f1-score (micro avg) 0.6943
2023-10-13 08:43:15,945 saving best model
2023-10-13 08:43:16,296 ----------------------------------------------------------------------------------------------------
2023-10-13 08:43:17,666 epoch 2 - iter 27/275 - loss 0.25324016 - time (sec): 1.37 - samples/sec: 1521.92 - lr: 0.000049 - momentum: 0.000000
2023-10-13 08:43:19,054 epoch 2 - iter 54/275 - loss 0.21752413 - time (sec): 2.76 - samples/sec: 1589.66 - lr: 0.000049 - momentum: 0.000000
2023-10-13 08:43:20,273 epoch 2 - iter 81/275 - loss 0.20318272 - time (sec): 3.98 - samples/sec: 1633.17 - lr: 0.000048 - momentum: 0.000000
2023-10-13 08:43:21,423 epoch 2 - iter 108/275 - loss 0.18840612 - time (sec): 5.13 - samples/sec: 1725.84 - lr: 0.000048 - momentum: 0.000000
2023-10-13 08:43:22,590 epoch 2 - iter 135/275 - loss 0.17939364 - time (sec): 6.29 - samples/sec: 1758.63 - lr: 0.000047 - momentum: 0.000000
2023-10-13 08:43:23,760 epoch 2 - iter 162/275 - loss 0.17560780 - time (sec): 7.46 - samples/sec: 1809.56 - lr: 0.000047 - momentum: 0.000000
2023-10-13 08:43:24,917 epoch 2 - iter 189/275 - loss 0.16513976 - time (sec): 8.62 - samples/sec: 1813.69 - lr: 0.000046 - momentum: 0.000000
2023-10-13 08:43:26,123 epoch 2 - iter 216/275 - loss 0.16618392 - time (sec): 9.83 - samples/sec: 1836.15 - lr: 0.000046 - momentum: 0.000000
2023-10-13 08:43:27,341 epoch 2 - iter 243/275 - loss 0.16274574 - time (sec): 11.04 - samples/sec: 1835.74 - lr: 0.000045 - momentum: 0.000000
2023-10-13 08:43:28,528 epoch 2 - iter 270/275 - loss 0.16043632 - time (sec): 12.23 - samples/sec: 1821.92 - lr: 0.000045 - momentum: 0.000000
2023-10-13 08:43:28,760 ----------------------------------------------------------------------------------------------------
2023-10-13 08:43:28,760 EPOCH 2 done: loss 0.1631 - lr: 0.000045
2023-10-13 08:43:29,496 DEV : loss 0.13491378724575043 - f1-score (micro avg) 0.8185
2023-10-13 08:43:29,504 saving best model
2023-10-13 08:43:30,065 ----------------------------------------------------------------------------------------------------
2023-10-13 08:43:31,469 epoch 3 - iter 27/275 - loss 0.09269037 - time (sec): 1.40 - samples/sec: 1632.01 - lr: 0.000044 - momentum: 0.000000
2023-10-13 08:43:32,883 epoch 3 - iter 54/275 - loss 0.10066397 - time (sec): 2.82 - samples/sec: 1592.28 - lr: 0.000043 - momentum: 0.000000
2023-10-13 08:43:34,176 epoch 3 - iter 81/275 - loss 0.10912877 - time (sec): 4.11 - samples/sec: 1629.65 - lr: 0.000043 - momentum: 0.000000
2023-10-13 08:43:35,461 epoch 3 - iter 108/275 - loss 0.10894748 - time (sec): 5.39 - samples/sec: 1665.22 - lr: 0.000042 - momentum: 0.000000
2023-10-13 08:43:36,721 epoch 3 - iter 135/275 - loss 0.10675830 - time (sec): 6.65 - samples/sec: 1687.14 - lr: 0.000042 - momentum: 0.000000
2023-10-13 08:43:37,984 epoch 3 - iter 162/275 - loss 0.10865347 - time (sec): 7.92 - samples/sec: 1698.31 - lr: 0.000041 - momentum: 0.000000
2023-10-13 08:43:39,250 epoch 3 - iter 189/275 - loss 0.10904253 - time (sec): 9.18 - samples/sec: 1736.02 - lr: 0.000041 - momentum: 0.000000
2023-10-13 08:43:40,472 epoch 3 - iter 216/275 - loss 0.10403065 - time (sec): 10.40 - samples/sec: 1741.79 - lr: 0.000040 - momentum: 0.000000
2023-10-13 08:43:41,782 epoch 3 - iter 243/275 - loss 0.09802057 - time (sec): 11.71 - samples/sec: 1741.04 - lr: 0.000040 - momentum: 0.000000
2023-10-13 08:43:43,048 epoch 3 - iter 270/275 - loss 0.10304729 - time (sec): 12.98 - samples/sec: 1729.18 - lr: 0.000039 - momentum: 0.000000
2023-10-13 08:43:43,272 ----------------------------------------------------------------------------------------------------
2023-10-13 08:43:43,272 EPOCH 3 done: loss 0.1023 - lr: 0.000039
2023-10-13 08:43:43,929 DEV : loss 0.15517204999923706 - f1-score (micro avg) 0.8565
2023-10-13 08:43:43,935 saving best model
2023-10-13 08:43:44,434 ----------------------------------------------------------------------------------------------------
2023-10-13 08:43:45,562 epoch 4 - iter 27/275 - loss 0.04730955 - time (sec): 1.13 - samples/sec: 1946.03 - lr: 0.000038 - momentum: 0.000000
2023-10-13 08:43:46,720 epoch 4 - iter 54/275 - loss 0.05859413 - time (sec): 2.29 - samples/sec: 1886.10 - lr: 0.000038 - momentum: 0.000000
2023-10-13 08:43:47,890 epoch 4 - iter 81/275 - loss 0.07722230 - time (sec): 3.45 - samples/sec: 1839.70 - lr: 0.000037 - momentum: 0.000000
2023-10-13 08:43:49,046 epoch 4 - iter 108/275 - loss 0.08290820 - time (sec): 4.61 - samples/sec: 1839.41 - lr: 0.000037 - momentum: 0.000000
2023-10-13 08:43:50,209 epoch 4 - iter 135/275 - loss 0.07883405 - time (sec): 5.77 - samples/sec: 1853.08 - lr: 0.000036 - momentum: 0.000000
2023-10-13 08:43:51,350 epoch 4 - iter 162/275 - loss 0.07919026 - time (sec): 6.91 - samples/sec: 1874.21 - lr: 0.000036 - momentum: 0.000000
2023-10-13 08:43:52,473 epoch 4 - iter 189/275 - loss 0.07762734 - time (sec): 8.04 - samples/sec: 1856.64 - lr: 0.000035 - momentum: 0.000000
2023-10-13 08:43:53,639 epoch 4 - iter 216/275 - loss 0.07935127 - time (sec): 9.20 - samples/sec: 1865.31 - lr: 0.000035 - momentum: 0.000000
2023-10-13 08:43:54,847 epoch 4 - iter 243/275 - loss 0.08193059 - time (sec): 10.41 - samples/sec: 1894.59 - lr: 0.000034 - momentum: 0.000000
2023-10-13 08:43:56,056 epoch 4 - iter 270/275 - loss 0.08075966 - time (sec): 11.62 - samples/sec: 1920.53 - lr: 0.000034 - momentum: 0.000000
2023-10-13 08:43:56,269 ----------------------------------------------------------------------------------------------------
2023-10-13 08:43:56,269 EPOCH 4 done: loss 0.0794 - lr: 0.000034
2023-10-13 08:43:56,992 DEV : loss 0.15865449607372284 - f1-score (micro avg) 0.8635
2023-10-13 08:43:56,999 saving best model
2023-10-13 08:43:57,489 ----------------------------------------------------------------------------------------------------
2023-10-13 08:43:58,703 epoch 5 - iter 27/275 - loss 0.06511906 - time (sec): 1.21 - samples/sec: 1823.46 - lr: 0.000033 - momentum: 0.000000
2023-10-13 08:43:59,970 epoch 5 - iter 54/275 - loss 0.05423925 - time (sec): 2.48 - samples/sec: 1859.97 - lr: 0.000032 - momentum: 0.000000
2023-10-13 08:44:01,167 epoch 5 - iter 81/275 - loss 0.05481274 - time (sec): 3.68 - samples/sec: 1834.82 - lr: 0.000032 - momentum: 0.000000
2023-10-13 08:44:02,366 epoch 5 - iter 108/275 - loss 0.06460615 - time (sec): 4.87 - samples/sec: 1869.79 - lr: 0.000031 - momentum: 0.000000
2023-10-13 08:44:03,583 epoch 5 - iter 135/275 - loss 0.05906795 - time (sec): 6.09 - samples/sec: 1876.83 - lr: 0.000031 - momentum: 0.000000
2023-10-13 08:44:04,760 epoch 5 - iter 162/275 - loss 0.05559834 - time (sec): 7.27 - samples/sec: 1865.62 - lr: 0.000030 - momentum: 0.000000
2023-10-13 08:44:05,937 epoch 5 - iter 189/275 - loss 0.05630927 - time (sec): 8.45 - samples/sec: 1828.54 - lr: 0.000030 - momentum: 0.000000
2023-10-13 08:44:07,134 epoch 5 - iter 216/275 - loss 0.06238579 - time (sec): 9.64 - samples/sec: 1840.92 - lr: 0.000029 - momentum: 0.000000
2023-10-13 08:44:08,318 epoch 5 - iter 243/275 - loss 0.05897466 - time (sec): 10.83 - samples/sec: 1838.93 - lr: 0.000029 - momentum: 0.000000
2023-10-13 08:44:09,512 epoch 5 - iter 270/275 - loss 0.06050901 - time (sec): 12.02 - samples/sec: 1848.42 - lr: 0.000028 - momentum: 0.000000
2023-10-13 08:44:09,745 ----------------------------------------------------------------------------------------------------
2023-10-13 08:44:09,745 EPOCH 5 done: loss 0.0608 - lr: 0.000028
2023-10-13 08:44:10,438 DEV : loss 0.17816683650016785 - f1-score (micro avg) 0.8638
2023-10-13 08:44:10,445 saving best model
2023-10-13 08:44:10,922 ----------------------------------------------------------------------------------------------------
2023-10-13 08:44:12,282 epoch 6 - iter 27/275 - loss 0.02081309 - time (sec): 1.36 - samples/sec: 1660.71 - lr: 0.000027 - momentum: 0.000000
2023-10-13 08:44:13,665 epoch 6 - iter 54/275 - loss 0.03422600 - time (sec): 2.74 - samples/sec: 1680.74 - lr: 0.000027 - momentum: 0.000000
2023-10-13 08:44:15,026 epoch 6 - iter 81/275 - loss 0.05469774 - time (sec): 4.10 - samples/sec: 1680.05 - lr: 0.000026 - momentum: 0.000000
2023-10-13 08:44:16,379 epoch 6 - iter 108/275 - loss 0.05159438 - time (sec): 5.45 - samples/sec: 1636.99 - lr: 0.000026 - momentum: 0.000000
2023-10-13 08:44:17,759 epoch 6 - iter 135/275 - loss 0.04869846 - time (sec): 6.83 - samples/sec: 1659.27 - lr: 0.000025 - momentum: 0.000000
2023-10-13 08:44:19,111 epoch 6 - iter 162/275 - loss 0.04743475 - time (sec): 8.19 - samples/sec: 1654.66 - lr: 0.000025 - momentum: 0.000000
2023-10-13 08:44:20,480 epoch 6 - iter 189/275 - loss 0.04989248 - time (sec): 9.55 - samples/sec: 1658.84 - lr: 0.000024 - momentum: 0.000000
2023-10-13 08:44:21,869 epoch 6 - iter 216/275 - loss 0.04818983 - time (sec): 10.94 - samples/sec: 1639.75 - lr: 0.000024 - momentum: 0.000000
2023-10-13 08:44:23,230 epoch 6 - iter 243/275 - loss 0.04658753 - time (sec): 12.31 - samples/sec: 1645.97 - lr: 0.000023 - momentum: 0.000000
2023-10-13 08:44:24,614 epoch 6 - iter 270/275 - loss 0.04485098 - time (sec): 13.69 - samples/sec: 1639.87 - lr: 0.000022 - momentum: 0.000000
2023-10-13 08:44:24,888 ----------------------------------------------------------------------------------------------------
2023-10-13 08:44:24,888 EPOCH 6 done: loss 0.0456 - lr: 0.000022
2023-10-13 08:44:25,580 DEV : loss 0.1625245064496994 - f1-score (micro avg) 0.8709
2023-10-13 08:44:25,587 saving best model
2023-10-13 08:44:26,030 ----------------------------------------------------------------------------------------------------
2023-10-13 08:44:27,414 epoch 7 - iter 27/275 - loss 0.02007844 - time (sec): 1.38 - samples/sec: 1410.89 - lr: 0.000022 - momentum: 0.000000
2023-10-13 08:44:28,779 epoch 7 - iter 54/275 - loss 0.02694298 - time (sec): 2.75 - samples/sec: 1503.12 - lr: 0.000021 - momentum: 0.000000
2023-10-13 08:44:30,159 epoch 7 - iter 81/275 - loss 0.03953833 - time (sec): 4.13 - samples/sec: 1564.02 - lr: 0.000021 - momentum: 0.000000
2023-10-13 08:44:31,613 epoch 7 - iter 108/275 - loss 0.04578602 - time (sec): 5.58 - samples/sec: 1580.73 - lr: 0.000020 - momentum: 0.000000
2023-10-13 08:44:33,000 epoch 7 - iter 135/275 - loss 0.03901358 - time (sec): 6.97 - samples/sec: 1584.86 - lr: 0.000020 - momentum: 0.000000
2023-10-13 08:44:34,325 epoch 7 - iter 162/275 - loss 0.03827773 - time (sec): 8.29 - samples/sec: 1601.22 - lr: 0.000019 - momentum: 0.000000
2023-10-13 08:44:35,648 epoch 7 - iter 189/275 - loss 0.03517038 - time (sec): 9.62 - samples/sec: 1633.74 - lr: 0.000019 - momentum: 0.000000
2023-10-13 08:44:37,011 epoch 7 - iter 216/275 - loss 0.03656872 - time (sec): 10.98 - samples/sec: 1655.72 - lr: 0.000018 - momentum: 0.000000
2023-10-13 08:44:38,258 epoch 7 - iter 243/275 - loss 0.03458599 - time (sec): 12.23 - samples/sec: 1648.73 - lr: 0.000017 - momentum: 0.000000
2023-10-13 08:44:39,467 epoch 7 - iter 270/275 - loss 0.03211470 - time (sec): 13.44 - samples/sec: 1657.37 - lr: 0.000017 - momentum: 0.000000
2023-10-13 08:44:39,689 ----------------------------------------------------------------------------------------------------
2023-10-13 08:44:39,690 EPOCH 7 done: loss 0.0321 - lr: 0.000017
2023-10-13 08:44:40,378 DEV : loss 0.16852889955043793 - f1-score (micro avg) 0.8699
2023-10-13 08:44:40,385 ----------------------------------------------------------------------------------------------------
2023-10-13 08:44:41,675 epoch 8 - iter 27/275 - loss 0.01991863 - time (sec): 1.29 - samples/sec: 1682.09 - lr: 0.000016 - momentum: 0.000000
2023-10-13 08:44:43,008 epoch 8 - iter 54/275 - loss 0.02203765 - time (sec): 2.62 - samples/sec: 1617.53 - lr: 0.000016 - momentum: 0.000000
2023-10-13 08:44:44,372 epoch 8 - iter 81/275 - loss 0.01842223 - time (sec): 3.99 - samples/sec: 1611.20 - lr: 0.000015 - momentum: 0.000000
2023-10-13 08:44:45,591 epoch 8 - iter 108/275 - loss 0.02703636 - time (sec): 5.20 - samples/sec: 1682.24 - lr: 0.000015 - momentum: 0.000000
2023-10-13 08:44:46,783 epoch 8 - iter 135/275 - loss 0.02656983 - time (sec): 6.40 - samples/sec: 1708.78 - lr: 0.000014 - momentum: 0.000000
2023-10-13 08:44:47,956 epoch 8 - iter 162/275 - loss 0.02337674 - time (sec): 7.57 - samples/sec: 1718.38 - lr: 0.000014 - momentum: 0.000000
2023-10-13 08:44:49,110 epoch 8 - iter 189/275 - loss 0.02320125 - time (sec): 8.72 - samples/sec: 1751.86 - lr: 0.000013 - momentum: 0.000000
2023-10-13 08:44:50,279 epoch 8 - iter 216/275 - loss 0.02203139 - time (sec): 9.89 - samples/sec: 1793.87 - lr: 0.000012 - momentum: 0.000000
2023-10-13 08:44:51,495 epoch 8 - iter 243/275 - loss 0.02298342 - time (sec): 11.11 - samples/sec: 1810.66 - lr: 0.000012 - momentum: 0.000000
2023-10-13 08:44:52,735 epoch 8 - iter 270/275 - loss 0.02201397 - time (sec): 12.35 - samples/sec: 1803.23 - lr: 0.000011 - momentum: 0.000000
2023-10-13 08:44:52,964 ----------------------------------------------------------------------------------------------------
2023-10-13 08:44:52,964 EPOCH 8 done: loss 0.0262 - lr: 0.000011
2023-10-13 08:44:53,644 DEV : loss 0.1576564908027649 - f1-score (micro avg) 0.8867
2023-10-13 08:44:53,649 saving best model
2023-10-13 08:44:54,211 ----------------------------------------------------------------------------------------------------
2023-10-13 08:44:55,422 epoch 9 - iter 27/275 - loss 0.00305494 - time (sec): 1.21 - samples/sec: 1659.86 - lr: 0.000011 - momentum: 0.000000
2023-10-13 08:44:56,646 epoch 9 - iter 54/275 - loss 0.04012301 - time (sec): 2.43 - samples/sec: 1716.76 - lr: 0.000010 - momentum: 0.000000
2023-10-13 08:44:57,903 epoch 9 - iter 81/275 - loss 0.02902370 - time (sec): 3.69 - samples/sec: 1731.43 - lr: 0.000010 - momentum: 0.000000
2023-10-13 08:44:59,152 epoch 9 - iter 108/275 - loss 0.02389707 - time (sec): 4.94 - samples/sec: 1771.21 - lr: 0.000009 - momentum: 0.000000
2023-10-13 08:45:00,405 epoch 9 - iter 135/275 - loss 0.01980409 - time (sec): 6.19 - samples/sec: 1787.82 - lr: 0.000009 - momentum: 0.000000
2023-10-13 08:45:01,624 epoch 9 - iter 162/275 - loss 0.01933285 - time (sec): 7.41 - samples/sec: 1804.29 - lr: 0.000008 - momentum: 0.000000
2023-10-13 08:45:02,935 epoch 9 - iter 189/275 - loss 0.01697799 - time (sec): 8.72 - samples/sec: 1773.38 - lr: 0.000007 - momentum: 0.000000
2023-10-13 08:45:04,413 epoch 9 - iter 216/275 - loss 0.01651821 - time (sec): 10.20 - samples/sec: 1750.13 - lr: 0.000007 - momentum: 0.000000
2023-10-13 08:45:05,931 epoch 9 - iter 243/275 - loss 0.01563807 - time (sec): 11.72 - samples/sec: 1727.71 - lr: 0.000006 - momentum: 0.000000
2023-10-13 08:45:07,391 epoch 9 - iter 270/275 - loss 0.01736356 - time (sec): 13.18 - samples/sec: 1691.76 - lr: 0.000006 - momentum: 0.000000
2023-10-13 08:45:07,672 ----------------------------------------------------------------------------------------------------
2023-10-13 08:45:07,673 EPOCH 9 done: loss 0.0172 - lr: 0.000006
2023-10-13 08:45:08,365 DEV : loss 0.15656420588493347 - f1-score (micro avg) 0.8764
2023-10-13 08:45:08,372 ----------------------------------------------------------------------------------------------------
2023-10-13 08:45:09,826 epoch 10 - iter 27/275 - loss 0.00119089 - time (sec): 1.45 - samples/sec: 1396.00 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:45:11,229 epoch 10 - iter 54/275 - loss 0.01293146 - time (sec): 2.86 - samples/sec: 1532.63 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:45:12,587 epoch 10 - iter 81/275 - loss 0.00924602 - time (sec): 4.21 - samples/sec: 1543.93 - lr: 0.000004 - momentum: 0.000000
2023-10-13 08:45:13,955 epoch 10 - iter 108/275 - loss 0.01041877 - time (sec): 5.58 - samples/sec: 1563.54 - lr: 0.000004 - momentum: 0.000000
2023-10-13 08:45:15,386 epoch 10 - iter 135/275 - loss 0.01078816 - time (sec): 7.01 - samples/sec: 1613.49 - lr: 0.000003 - momentum: 0.000000
2023-10-13 08:45:16,722 epoch 10 - iter 162/275 - loss 0.01072570 - time (sec): 8.35 - samples/sec: 1583.39 - lr: 0.000002 - momentum: 0.000000
2023-10-13 08:45:17,913 epoch 10 - iter 189/275 - loss 0.00990525 - time (sec): 9.54 - samples/sec: 1638.19 - lr: 0.000002 - momentum: 0.000000
2023-10-13 08:45:19,088 epoch 10 - iter 216/275 - loss 0.01335076 - time (sec): 10.71 - samples/sec: 1671.76 - lr: 0.000001 - momentum: 0.000000
2023-10-13 08:45:20,259 epoch 10 - iter 243/275 - loss 0.01220671 - time (sec): 11.89 - samples/sec: 1676.88 - lr: 0.000001 - momentum: 0.000000
2023-10-13 08:45:21,443 epoch 10 - iter 270/275 - loss 0.01244867 - time (sec): 13.07 - samples/sec: 1708.53 - lr: 0.000000 - momentum: 0.000000
2023-10-13 08:45:21,663 ----------------------------------------------------------------------------------------------------
2023-10-13 08:45:21,664 EPOCH 10 done: loss 0.0126 - lr: 0.000000
2023-10-13 08:45:22,746 DEV : loss 0.1590714156627655 - f1-score (micro avg) 0.8803
2023-10-13 08:45:23,094 ----------------------------------------------------------------------------------------------------
2023-10-13 08:45:23,095 Loading model from best epoch ...
2023-10-13 08:45:24,509 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
2023-10-13 08:45:25,142
Results:
- F-score (micro) 0.9045
- F-score (macro) 0.8061
- Accuracy 0.8378
By class:
precision recall f1-score support
scope 0.8902 0.8750 0.8825 176
pers 0.9672 0.9219 0.9440 128
work 0.9167 0.8919 0.9041 74
loc 0.6667 1.0000 0.8000 2
object 0.5000 0.5000 0.5000 2
micro avg 0.9167 0.8927 0.9045 382
macro avg 0.7881 0.8378 0.8061 382
weighted avg 0.9179 0.8927 0.9049 382
2023-10-13 08:45:25,142 ----------------------------------------------------------------------------------------------------
|