File size: 24,065 Bytes
bff1c25 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 |
2023-10-13 09:24:54,182 ----------------------------------------------------------------------------------------------------
2023-10-13 09:24:54,183 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 09:24:54,183 ----------------------------------------------------------------------------------------------------
2023-10-13 09:24:54,183 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
2023-10-13 09:24:54,183 ----------------------------------------------------------------------------------------------------
2023-10-13 09:24:54,183 Train: 1214 sentences
2023-10-13 09:24:54,183 (train_with_dev=False, train_with_test=False)
2023-10-13 09:24:54,183 ----------------------------------------------------------------------------------------------------
2023-10-13 09:24:54,183 Training Params:
2023-10-13 09:24:54,183 - learning_rate: "5e-05"
2023-10-13 09:24:54,183 - mini_batch_size: "4"
2023-10-13 09:24:54,183 - max_epochs: "10"
2023-10-13 09:24:54,183 - shuffle: "True"
2023-10-13 09:24:54,183 ----------------------------------------------------------------------------------------------------
2023-10-13 09:24:54,183 Plugins:
2023-10-13 09:24:54,184 - LinearScheduler | warmup_fraction: '0.1'
2023-10-13 09:24:54,184 ----------------------------------------------------------------------------------------------------
2023-10-13 09:24:54,184 Final evaluation on model from best epoch (best-model.pt)
2023-10-13 09:24:54,184 - metric: "('micro avg', 'f1-score')"
2023-10-13 09:24:54,184 ----------------------------------------------------------------------------------------------------
2023-10-13 09:24:54,184 Computation:
2023-10-13 09:24:54,184 - compute on device: cuda:0
2023-10-13 09:24:54,184 - embedding storage: none
2023-10-13 09:24:54,184 ----------------------------------------------------------------------------------------------------
2023-10-13 09:24:54,184 Model training base path: "hmbench-ajmc/en-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3"
2023-10-13 09:24:54,184 ----------------------------------------------------------------------------------------------------
2023-10-13 09:24:54,184 ----------------------------------------------------------------------------------------------------
2023-10-13 09:24:55,668 epoch 1 - iter 30/304 - loss 3.06399738 - time (sec): 1.48 - samples/sec: 2127.62 - lr: 0.000005 - momentum: 0.000000
2023-10-13 09:24:57,203 epoch 1 - iter 60/304 - loss 2.26158163 - time (sec): 3.02 - samples/sec: 2025.00 - lr: 0.000010 - momentum: 0.000000
2023-10-13 09:24:58,809 epoch 1 - iter 90/304 - loss 1.75579198 - time (sec): 4.62 - samples/sec: 1964.93 - lr: 0.000015 - momentum: 0.000000
2023-10-13 09:25:00,392 epoch 1 - iter 120/304 - loss 1.46459955 - time (sec): 6.21 - samples/sec: 1979.38 - lr: 0.000020 - momentum: 0.000000
2023-10-13 09:25:02,019 epoch 1 - iter 150/304 - loss 1.24942060 - time (sec): 7.83 - samples/sec: 1974.58 - lr: 0.000025 - momentum: 0.000000
2023-10-13 09:25:03,606 epoch 1 - iter 180/304 - loss 1.08919783 - time (sec): 9.42 - samples/sec: 1964.07 - lr: 0.000029 - momentum: 0.000000
2023-10-13 09:25:05,190 epoch 1 - iter 210/304 - loss 0.97371069 - time (sec): 11.00 - samples/sec: 1993.42 - lr: 0.000034 - momentum: 0.000000
2023-10-13 09:25:06,718 epoch 1 - iter 240/304 - loss 0.90345855 - time (sec): 12.53 - samples/sec: 1963.22 - lr: 0.000039 - momentum: 0.000000
2023-10-13 09:25:08,149 epoch 1 - iter 270/304 - loss 0.83443013 - time (sec): 13.96 - samples/sec: 1969.09 - lr: 0.000044 - momentum: 0.000000
2023-10-13 09:25:09,678 epoch 1 - iter 300/304 - loss 0.77115655 - time (sec): 15.49 - samples/sec: 1980.61 - lr: 0.000049 - momentum: 0.000000
2023-10-13 09:25:09,898 ----------------------------------------------------------------------------------------------------
2023-10-13 09:25:09,898 EPOCH 1 done: loss 0.7659 - lr: 0.000049
2023-10-13 09:25:10,999 DEV : loss 0.1917836219072342 - f1-score (micro avg) 0.6731
2023-10-13 09:25:11,006 saving best model
2023-10-13 09:25:11,357 ----------------------------------------------------------------------------------------------------
2023-10-13 09:25:12,871 epoch 2 - iter 30/304 - loss 0.19001675 - time (sec): 1.51 - samples/sec: 1817.12 - lr: 0.000049 - momentum: 0.000000
2023-10-13 09:25:14,370 epoch 2 - iter 60/304 - loss 0.16332851 - time (sec): 3.01 - samples/sec: 1913.15 - lr: 0.000049 - momentum: 0.000000
2023-10-13 09:25:15,833 epoch 2 - iter 90/304 - loss 0.17518968 - time (sec): 4.47 - samples/sec: 1930.72 - lr: 0.000048 - momentum: 0.000000
2023-10-13 09:25:17,205 epoch 2 - iter 120/304 - loss 0.16320078 - time (sec): 5.85 - samples/sec: 2039.49 - lr: 0.000048 - momentum: 0.000000
2023-10-13 09:25:18,550 epoch 2 - iter 150/304 - loss 0.15631235 - time (sec): 7.19 - samples/sec: 2112.46 - lr: 0.000047 - momentum: 0.000000
2023-10-13 09:25:19,859 epoch 2 - iter 180/304 - loss 0.15100832 - time (sec): 8.50 - samples/sec: 2141.52 - lr: 0.000047 - momentum: 0.000000
2023-10-13 09:25:21,180 epoch 2 - iter 210/304 - loss 0.14814376 - time (sec): 9.82 - samples/sec: 2181.33 - lr: 0.000046 - momentum: 0.000000
2023-10-13 09:25:22,527 epoch 2 - iter 240/304 - loss 0.13961372 - time (sec): 11.17 - samples/sec: 2181.57 - lr: 0.000046 - momentum: 0.000000
2023-10-13 09:25:23,843 epoch 2 - iter 270/304 - loss 0.13518424 - time (sec): 12.48 - samples/sec: 2209.33 - lr: 0.000045 - momentum: 0.000000
2023-10-13 09:25:25,172 epoch 2 - iter 300/304 - loss 0.14038691 - time (sec): 13.81 - samples/sec: 2221.89 - lr: 0.000045 - momentum: 0.000000
2023-10-13 09:25:25,342 ----------------------------------------------------------------------------------------------------
2023-10-13 09:25:25,343 EPOCH 2 done: loss 0.1393 - lr: 0.000045
2023-10-13 09:25:26,383 DEV : loss 0.16002638638019562 - f1-score (micro avg) 0.7846
2023-10-13 09:25:26,395 saving best model
2023-10-13 09:25:26,852 ----------------------------------------------------------------------------------------------------
2023-10-13 09:25:28,435 epoch 3 - iter 30/304 - loss 0.06790292 - time (sec): 1.58 - samples/sec: 1897.09 - lr: 0.000044 - momentum: 0.000000
2023-10-13 09:25:29,990 epoch 3 - iter 60/304 - loss 0.08312541 - time (sec): 3.14 - samples/sec: 1889.26 - lr: 0.000043 - momentum: 0.000000
2023-10-13 09:25:31,541 epoch 3 - iter 90/304 - loss 0.07538342 - time (sec): 4.69 - samples/sec: 1894.44 - lr: 0.000043 - momentum: 0.000000
2023-10-13 09:25:33,194 epoch 3 - iter 120/304 - loss 0.07350489 - time (sec): 6.34 - samples/sec: 1875.91 - lr: 0.000042 - momentum: 0.000000
2023-10-13 09:25:34,793 epoch 3 - iter 150/304 - loss 0.08069146 - time (sec): 7.94 - samples/sec: 1908.79 - lr: 0.000042 - momentum: 0.000000
2023-10-13 09:25:36,339 epoch 3 - iter 180/304 - loss 0.08343527 - time (sec): 9.48 - samples/sec: 1940.93 - lr: 0.000041 - momentum: 0.000000
2023-10-13 09:25:37,918 epoch 3 - iter 210/304 - loss 0.08407638 - time (sec): 11.06 - samples/sec: 1914.86 - lr: 0.000041 - momentum: 0.000000
2023-10-13 09:25:39,421 epoch 3 - iter 240/304 - loss 0.08240695 - time (sec): 12.57 - samples/sec: 1932.36 - lr: 0.000040 - momentum: 0.000000
2023-10-13 09:25:40,786 epoch 3 - iter 270/304 - loss 0.08431923 - time (sec): 13.93 - samples/sec: 1970.97 - lr: 0.000040 - momentum: 0.000000
2023-10-13 09:25:42,132 epoch 3 - iter 300/304 - loss 0.08278814 - time (sec): 15.28 - samples/sec: 2004.54 - lr: 0.000039 - momentum: 0.000000
2023-10-13 09:25:42,303 ----------------------------------------------------------------------------------------------------
2023-10-13 09:25:42,303 EPOCH 3 done: loss 0.0823 - lr: 0.000039
2023-10-13 09:25:43,252 DEV : loss 0.17538176476955414 - f1-score (micro avg) 0.8167
2023-10-13 09:25:43,262 saving best model
2023-10-13 09:25:43,838 ----------------------------------------------------------------------------------------------------
2023-10-13 09:25:45,202 epoch 4 - iter 30/304 - loss 0.06052743 - time (sec): 1.36 - samples/sec: 2137.33 - lr: 0.000038 - momentum: 0.000000
2023-10-13 09:25:46,545 epoch 4 - iter 60/304 - loss 0.04582401 - time (sec): 2.70 - samples/sec: 2176.13 - lr: 0.000038 - momentum: 0.000000
2023-10-13 09:25:47,896 epoch 4 - iter 90/304 - loss 0.06348942 - time (sec): 4.06 - samples/sec: 2217.37 - lr: 0.000037 - momentum: 0.000000
2023-10-13 09:25:49,209 epoch 4 - iter 120/304 - loss 0.06958931 - time (sec): 5.37 - samples/sec: 2286.77 - lr: 0.000037 - momentum: 0.000000
2023-10-13 09:25:50,506 epoch 4 - iter 150/304 - loss 0.06465443 - time (sec): 6.66 - samples/sec: 2293.94 - lr: 0.000036 - momentum: 0.000000
2023-10-13 09:25:51,861 epoch 4 - iter 180/304 - loss 0.05834462 - time (sec): 8.02 - samples/sec: 2263.26 - lr: 0.000036 - momentum: 0.000000
2023-10-13 09:25:53,208 epoch 4 - iter 210/304 - loss 0.05591947 - time (sec): 9.37 - samples/sec: 2274.03 - lr: 0.000035 - momentum: 0.000000
2023-10-13 09:25:54,577 epoch 4 - iter 240/304 - loss 0.06178706 - time (sec): 10.74 - samples/sec: 2276.87 - lr: 0.000035 - momentum: 0.000000
2023-10-13 09:25:55,923 epoch 4 - iter 270/304 - loss 0.06195653 - time (sec): 12.08 - samples/sec: 2268.21 - lr: 0.000034 - momentum: 0.000000
2023-10-13 09:25:57,273 epoch 4 - iter 300/304 - loss 0.06233097 - time (sec): 13.43 - samples/sec: 2274.92 - lr: 0.000033 - momentum: 0.000000
2023-10-13 09:25:57,454 ----------------------------------------------------------------------------------------------------
2023-10-13 09:25:57,454 EPOCH 4 done: loss 0.0639 - lr: 0.000033
2023-10-13 09:25:58,407 DEV : loss 0.21819397807121277 - f1-score (micro avg) 0.8037
2023-10-13 09:25:58,416 ----------------------------------------------------------------------------------------------------
2023-10-13 09:25:59,792 epoch 5 - iter 30/304 - loss 0.06028989 - time (sec): 1.37 - samples/sec: 2232.46 - lr: 0.000033 - momentum: 0.000000
2023-10-13 09:26:01,374 epoch 5 - iter 60/304 - loss 0.04886804 - time (sec): 2.96 - samples/sec: 2132.90 - lr: 0.000032 - momentum: 0.000000
2023-10-13 09:26:02,931 epoch 5 - iter 90/304 - loss 0.04157150 - time (sec): 4.51 - samples/sec: 2031.33 - lr: 0.000032 - momentum: 0.000000
2023-10-13 09:26:04,453 epoch 5 - iter 120/304 - loss 0.04799012 - time (sec): 6.04 - samples/sec: 2026.96 - lr: 0.000031 - momentum: 0.000000
2023-10-13 09:26:05,965 epoch 5 - iter 150/304 - loss 0.04485221 - time (sec): 7.55 - samples/sec: 2045.94 - lr: 0.000031 - momentum: 0.000000
2023-10-13 09:26:07,487 epoch 5 - iter 180/304 - loss 0.05156410 - time (sec): 9.07 - samples/sec: 2041.58 - lr: 0.000030 - momentum: 0.000000
2023-10-13 09:26:09,009 epoch 5 - iter 210/304 - loss 0.05474022 - time (sec): 10.59 - samples/sec: 2041.70 - lr: 0.000030 - momentum: 0.000000
2023-10-13 09:26:10,532 epoch 5 - iter 240/304 - loss 0.05293665 - time (sec): 12.11 - samples/sec: 2045.66 - lr: 0.000029 - momentum: 0.000000
2023-10-13 09:26:12,016 epoch 5 - iter 270/304 - loss 0.04984029 - time (sec): 13.60 - samples/sec: 2023.60 - lr: 0.000028 - momentum: 0.000000
2023-10-13 09:26:13,500 epoch 5 - iter 300/304 - loss 0.04969734 - time (sec): 15.08 - samples/sec: 2033.61 - lr: 0.000028 - momentum: 0.000000
2023-10-13 09:26:13,681 ----------------------------------------------------------------------------------------------------
2023-10-13 09:26:13,681 EPOCH 5 done: loss 0.0503 - lr: 0.000028
2023-10-13 09:26:14,665 DEV : loss 0.2147764265537262 - f1-score (micro avg) 0.8219
2023-10-13 09:26:14,674 saving best model
2023-10-13 09:26:15,374 ----------------------------------------------------------------------------------------------------
2023-10-13 09:26:17,145 epoch 6 - iter 30/304 - loss 0.03857010 - time (sec): 1.77 - samples/sec: 2027.32 - lr: 0.000027 - momentum: 0.000000
2023-10-13 09:26:18,495 epoch 6 - iter 60/304 - loss 0.03065086 - time (sec): 3.12 - samples/sec: 2067.04 - lr: 0.000027 - momentum: 0.000000
2023-10-13 09:26:19,867 epoch 6 - iter 90/304 - loss 0.02994557 - time (sec): 4.49 - samples/sec: 2091.38 - lr: 0.000026 - momentum: 0.000000
2023-10-13 09:26:21,188 epoch 6 - iter 120/304 - loss 0.02589307 - time (sec): 5.81 - samples/sec: 2162.40 - lr: 0.000026 - momentum: 0.000000
2023-10-13 09:26:22,502 epoch 6 - iter 150/304 - loss 0.02925331 - time (sec): 7.13 - samples/sec: 2177.44 - lr: 0.000025 - momentum: 0.000000
2023-10-13 09:26:23,830 epoch 6 - iter 180/304 - loss 0.03045963 - time (sec): 8.45 - samples/sec: 2200.63 - lr: 0.000025 - momentum: 0.000000
2023-10-13 09:26:25,156 epoch 6 - iter 210/304 - loss 0.03381298 - time (sec): 9.78 - samples/sec: 2213.88 - lr: 0.000024 - momentum: 0.000000
2023-10-13 09:26:26,482 epoch 6 - iter 240/304 - loss 0.04034034 - time (sec): 11.11 - samples/sec: 2221.95 - lr: 0.000023 - momentum: 0.000000
2023-10-13 09:26:28,098 epoch 6 - iter 270/304 - loss 0.04229335 - time (sec): 12.72 - samples/sec: 2184.74 - lr: 0.000023 - momentum: 0.000000
2023-10-13 09:26:29,647 epoch 6 - iter 300/304 - loss 0.03913055 - time (sec): 14.27 - samples/sec: 2143.57 - lr: 0.000022 - momentum: 0.000000
2023-10-13 09:26:29,844 ----------------------------------------------------------------------------------------------------
2023-10-13 09:26:29,844 EPOCH 6 done: loss 0.0387 - lr: 0.000022
2023-10-13 09:26:30,791 DEV : loss 0.2286389172077179 - f1-score (micro avg) 0.8308
2023-10-13 09:26:30,799 saving best model
2023-10-13 09:26:31,330 ----------------------------------------------------------------------------------------------------
2023-10-13 09:26:32,757 epoch 7 - iter 30/304 - loss 0.01753161 - time (sec): 1.43 - samples/sec: 2105.43 - lr: 0.000022 - momentum: 0.000000
2023-10-13 09:26:34,149 epoch 7 - iter 60/304 - loss 0.02359718 - time (sec): 2.82 - samples/sec: 2127.05 - lr: 0.000021 - momentum: 0.000000
2023-10-13 09:26:35,712 epoch 7 - iter 90/304 - loss 0.02659548 - time (sec): 4.38 - samples/sec: 2061.77 - lr: 0.000021 - momentum: 0.000000
2023-10-13 09:26:37,401 epoch 7 - iter 120/304 - loss 0.02435644 - time (sec): 6.07 - samples/sec: 1988.68 - lr: 0.000020 - momentum: 0.000000
2023-10-13 09:26:38,887 epoch 7 - iter 150/304 - loss 0.02337589 - time (sec): 7.56 - samples/sec: 2001.08 - lr: 0.000020 - momentum: 0.000000
2023-10-13 09:26:40,301 epoch 7 - iter 180/304 - loss 0.02205755 - time (sec): 8.97 - samples/sec: 2016.95 - lr: 0.000019 - momentum: 0.000000
2023-10-13 09:26:41,704 epoch 7 - iter 210/304 - loss 0.02284182 - time (sec): 10.37 - samples/sec: 2035.89 - lr: 0.000018 - momentum: 0.000000
2023-10-13 09:26:43,080 epoch 7 - iter 240/304 - loss 0.02205146 - time (sec): 11.75 - samples/sec: 2068.84 - lr: 0.000018 - momentum: 0.000000
2023-10-13 09:26:44,430 epoch 7 - iter 270/304 - loss 0.02110901 - time (sec): 13.10 - samples/sec: 2110.95 - lr: 0.000017 - momentum: 0.000000
2023-10-13 09:26:45,801 epoch 7 - iter 300/304 - loss 0.02295478 - time (sec): 14.47 - samples/sec: 2121.90 - lr: 0.000017 - momentum: 0.000000
2023-10-13 09:26:45,968 ----------------------------------------------------------------------------------------------------
2023-10-13 09:26:45,968 EPOCH 7 done: loss 0.0243 - lr: 0.000017
2023-10-13 09:26:47,032 DEV : loss 0.22793522477149963 - f1-score (micro avg) 0.82
2023-10-13 09:26:47,041 ----------------------------------------------------------------------------------------------------
2023-10-13 09:26:48,649 epoch 8 - iter 30/304 - loss 0.02012554 - time (sec): 1.61 - samples/sec: 1908.27 - lr: 0.000016 - momentum: 0.000000
2023-10-13 09:26:50,121 epoch 8 - iter 60/304 - loss 0.03760001 - time (sec): 3.08 - samples/sec: 2113.10 - lr: 0.000016 - momentum: 0.000000
2023-10-13 09:26:51,446 epoch 8 - iter 90/304 - loss 0.03699049 - time (sec): 4.40 - samples/sec: 2134.31 - lr: 0.000015 - momentum: 0.000000
2023-10-13 09:26:52,785 epoch 8 - iter 120/304 - loss 0.02855817 - time (sec): 5.74 - samples/sec: 2134.70 - lr: 0.000015 - momentum: 0.000000
2023-10-13 09:26:54,136 epoch 8 - iter 150/304 - loss 0.02421875 - time (sec): 7.09 - samples/sec: 2139.98 - lr: 0.000014 - momentum: 0.000000
2023-10-13 09:26:55,501 epoch 8 - iter 180/304 - loss 0.02334823 - time (sec): 8.46 - samples/sec: 2171.59 - lr: 0.000013 - momentum: 0.000000
2023-10-13 09:26:56,844 epoch 8 - iter 210/304 - loss 0.02066826 - time (sec): 9.80 - samples/sec: 2177.53 - lr: 0.000013 - momentum: 0.000000
2023-10-13 09:26:58,225 epoch 8 - iter 240/304 - loss 0.02086912 - time (sec): 11.18 - samples/sec: 2183.75 - lr: 0.000012 - momentum: 0.000000
2023-10-13 09:26:59,612 epoch 8 - iter 270/304 - loss 0.01956457 - time (sec): 12.57 - samples/sec: 2191.52 - lr: 0.000012 - momentum: 0.000000
2023-10-13 09:27:00,937 epoch 8 - iter 300/304 - loss 0.01786150 - time (sec): 13.89 - samples/sec: 2201.82 - lr: 0.000011 - momentum: 0.000000
2023-10-13 09:27:01,119 ----------------------------------------------------------------------------------------------------
2023-10-13 09:27:01,119 EPOCH 8 done: loss 0.0176 - lr: 0.000011
2023-10-13 09:27:02,108 DEV : loss 0.23265127837657928 - f1-score (micro avg) 0.8329
2023-10-13 09:27:02,116 saving best model
2023-10-13 09:27:02,688 ----------------------------------------------------------------------------------------------------
2023-10-13 09:27:04,022 epoch 9 - iter 30/304 - loss 0.00303313 - time (sec): 1.33 - samples/sec: 2189.12 - lr: 0.000011 - momentum: 0.000000
2023-10-13 09:27:05,339 epoch 9 - iter 60/304 - loss 0.00552285 - time (sec): 2.65 - samples/sec: 2302.98 - lr: 0.000010 - momentum: 0.000000
2023-10-13 09:27:06,919 epoch 9 - iter 90/304 - loss 0.01349350 - time (sec): 4.23 - samples/sec: 2171.18 - lr: 0.000010 - momentum: 0.000000
2023-10-13 09:27:08,480 epoch 9 - iter 120/304 - loss 0.01134152 - time (sec): 5.79 - samples/sec: 2095.53 - lr: 0.000009 - momentum: 0.000000
2023-10-13 09:27:09,969 epoch 9 - iter 150/304 - loss 0.01043367 - time (sec): 7.28 - samples/sec: 2084.32 - lr: 0.000008 - momentum: 0.000000
2023-10-13 09:27:11,341 epoch 9 - iter 180/304 - loss 0.01395282 - time (sec): 8.65 - samples/sec: 2143.20 - lr: 0.000008 - momentum: 0.000000
2023-10-13 09:27:12,656 epoch 9 - iter 210/304 - loss 0.01191362 - time (sec): 9.96 - samples/sec: 2186.36 - lr: 0.000007 - momentum: 0.000000
2023-10-13 09:27:13,983 epoch 9 - iter 240/304 - loss 0.01338057 - time (sec): 11.29 - samples/sec: 2200.15 - lr: 0.000007 - momentum: 0.000000
2023-10-13 09:27:15,300 epoch 9 - iter 270/304 - loss 0.01401480 - time (sec): 12.61 - samples/sec: 2205.21 - lr: 0.000006 - momentum: 0.000000
2023-10-13 09:27:16,611 epoch 9 - iter 300/304 - loss 0.01441779 - time (sec): 13.92 - samples/sec: 2204.16 - lr: 0.000006 - momentum: 0.000000
2023-10-13 09:27:16,780 ----------------------------------------------------------------------------------------------------
2023-10-13 09:27:16,780 EPOCH 9 done: loss 0.0144 - lr: 0.000006
2023-10-13 09:27:17,805 DEV : loss 0.23394043743610382 - f1-score (micro avg) 0.8382
2023-10-13 09:27:17,813 saving best model
2023-10-13 09:27:18,306 ----------------------------------------------------------------------------------------------------
2023-10-13 09:27:19,615 epoch 10 - iter 30/304 - loss 0.00012268 - time (sec): 1.31 - samples/sec: 2299.01 - lr: 0.000005 - momentum: 0.000000
2023-10-13 09:27:21,071 epoch 10 - iter 60/304 - loss 0.00771760 - time (sec): 2.76 - samples/sec: 2169.58 - lr: 0.000005 - momentum: 0.000000
2023-10-13 09:27:22,447 epoch 10 - iter 90/304 - loss 0.00707183 - time (sec): 4.14 - samples/sec: 2237.43 - lr: 0.000004 - momentum: 0.000000
2023-10-13 09:27:23,716 epoch 10 - iter 120/304 - loss 0.00573922 - time (sec): 5.41 - samples/sec: 2269.24 - lr: 0.000003 - momentum: 0.000000
2023-10-13 09:27:25,018 epoch 10 - iter 150/304 - loss 0.00512078 - time (sec): 6.71 - samples/sec: 2269.99 - lr: 0.000003 - momentum: 0.000000
2023-10-13 09:27:26,486 epoch 10 - iter 180/304 - loss 0.00554864 - time (sec): 8.18 - samples/sec: 2230.80 - lr: 0.000002 - momentum: 0.000000
2023-10-13 09:27:27,854 epoch 10 - iter 210/304 - loss 0.01123371 - time (sec): 9.55 - samples/sec: 2231.16 - lr: 0.000002 - momentum: 0.000000
2023-10-13 09:27:29,159 epoch 10 - iter 240/304 - loss 0.01078583 - time (sec): 10.85 - samples/sec: 2263.55 - lr: 0.000001 - momentum: 0.000000
2023-10-13 09:27:30,514 epoch 10 - iter 270/304 - loss 0.00961352 - time (sec): 12.21 - samples/sec: 2263.25 - lr: 0.000001 - momentum: 0.000000
2023-10-13 09:27:31,830 epoch 10 - iter 300/304 - loss 0.01062488 - time (sec): 13.52 - samples/sec: 2267.98 - lr: 0.000000 - momentum: 0.000000
2023-10-13 09:27:31,999 ----------------------------------------------------------------------------------------------------
2023-10-13 09:27:32,000 EPOCH 10 done: loss 0.0105 - lr: 0.000000
2023-10-13 09:27:33,026 DEV : loss 0.22849875688552856 - f1-score (micro avg) 0.8405
2023-10-13 09:27:33,033 saving best model
2023-10-13 09:27:33,864 ----------------------------------------------------------------------------------------------------
2023-10-13 09:27:33,865 Loading model from best epoch ...
2023-10-13 09:27:35,338 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
2023-10-13 09:27:36,493
Results:
- F-score (micro) 0.7887
- F-score (macro) 0.6409
- Accuracy 0.6614
By class:
precision recall f1-score support
scope 0.7469 0.8013 0.7732 151
work 0.7130 0.8632 0.7810 95
pers 0.7928 0.9167 0.8502 96
date 0.0000 0.0000 0.0000 3
loc 1.0000 0.6667 0.8000 3
micro avg 0.7418 0.8420 0.7887 348
macro avg 0.6505 0.6496 0.6409 348
weighted avg 0.7461 0.8420 0.7901 348
2023-10-13 09:27:36,493 ----------------------------------------------------------------------------------------------------
|