File size: 24,096 Bytes
d3a365d |
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 12:23:10,194 ----------------------------------------------------------------------------------------------------
2023-10-13 12:23:10,195 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=21, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-13 12:23:10,195 ----------------------------------------------------------------------------------------------------
2023-10-13 12:23:10,195 MultiCorpus: 3575 train + 1235 dev + 1266 test sentences
- NER_HIPE_2022 Corpus: 3575 train + 1235 dev + 1266 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/de/with_doc_seperator
2023-10-13 12:23:10,195 ----------------------------------------------------------------------------------------------------
2023-10-13 12:23:10,195 Train: 3575 sentences
2023-10-13 12:23:10,195 (train_with_dev=False, train_with_test=False)
2023-10-13 12:23:10,195 ----------------------------------------------------------------------------------------------------
2023-10-13 12:23:10,195 Training Params:
2023-10-13 12:23:10,195 - learning_rate: "3e-05"
2023-10-13 12:23:10,195 - mini_batch_size: "4"
2023-10-13 12:23:10,195 - max_epochs: "10"
2023-10-13 12:23:10,195 - shuffle: "True"
2023-10-13 12:23:10,196 ----------------------------------------------------------------------------------------------------
2023-10-13 12:23:10,196 Plugins:
2023-10-13 12:23:10,196 - LinearScheduler | warmup_fraction: '0.1'
2023-10-13 12:23:10,196 ----------------------------------------------------------------------------------------------------
2023-10-13 12:23:10,196 Final evaluation on model from best epoch (best-model.pt)
2023-10-13 12:23:10,196 - metric: "('micro avg', 'f1-score')"
2023-10-13 12:23:10,196 ----------------------------------------------------------------------------------------------------
2023-10-13 12:23:10,196 Computation:
2023-10-13 12:23:10,196 - compute on device: cuda:0
2023-10-13 12:23:10,196 - embedding storage: none
2023-10-13 12:23:10,196 ----------------------------------------------------------------------------------------------------
2023-10-13 12:23:10,196 Model training base path: "hmbench-hipe2020/de-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-13 12:23:10,196 ----------------------------------------------------------------------------------------------------
2023-10-13 12:23:10,196 ----------------------------------------------------------------------------------------------------
2023-10-13 12:23:14,561 epoch 1 - iter 89/894 - loss 3.10167065 - time (sec): 4.36 - samples/sec: 1836.58 - lr: 0.000003 - momentum: 0.000000
2023-10-13 12:23:18,929 epoch 1 - iter 178/894 - loss 2.04262052 - time (sec): 8.73 - samples/sec: 1838.99 - lr: 0.000006 - momentum: 0.000000
2023-10-13 12:23:23,191 epoch 1 - iter 267/894 - loss 1.47032662 - time (sec): 12.99 - samples/sec: 1919.00 - lr: 0.000009 - momentum: 0.000000
2023-10-13 12:23:27,431 epoch 1 - iter 356/894 - loss 1.21051529 - time (sec): 17.23 - samples/sec: 1916.83 - lr: 0.000012 - momentum: 0.000000
2023-10-13 12:23:31,569 epoch 1 - iter 445/894 - loss 1.02631486 - time (sec): 21.37 - samples/sec: 1959.98 - lr: 0.000015 - momentum: 0.000000
2023-10-13 12:23:36,164 epoch 1 - iter 534/894 - loss 0.88774045 - time (sec): 25.97 - samples/sec: 2010.02 - lr: 0.000018 - momentum: 0.000000
2023-10-13 12:23:40,328 epoch 1 - iter 623/894 - loss 0.80641379 - time (sec): 30.13 - samples/sec: 2007.12 - lr: 0.000021 - momentum: 0.000000
2023-10-13 12:23:44,841 epoch 1 - iter 712/894 - loss 0.73642288 - time (sec): 34.64 - samples/sec: 1998.57 - lr: 0.000024 - momentum: 0.000000
2023-10-13 12:23:49,049 epoch 1 - iter 801/894 - loss 0.68713278 - time (sec): 38.85 - samples/sec: 1987.57 - lr: 0.000027 - momentum: 0.000000
2023-10-13 12:23:53,392 epoch 1 - iter 890/894 - loss 0.63696695 - time (sec): 43.19 - samples/sec: 1991.95 - lr: 0.000030 - momentum: 0.000000
2023-10-13 12:23:53,573 ----------------------------------------------------------------------------------------------------
2023-10-13 12:23:53,574 EPOCH 1 done: loss 0.6340 - lr: 0.000030
2023-10-13 12:23:58,634 DEV : loss 0.1835888773202896 - f1-score (micro avg) 0.5989
2023-10-13 12:23:58,663 saving best model
2023-10-13 12:23:59,015 ----------------------------------------------------------------------------------------------------
2023-10-13 12:24:03,466 epoch 2 - iter 89/894 - loss 0.20312042 - time (sec): 4.45 - samples/sec: 1933.81 - lr: 0.000030 - momentum: 0.000000
2023-10-13 12:24:08,010 epoch 2 - iter 178/894 - loss 0.19771800 - time (sec): 8.99 - samples/sec: 1893.98 - lr: 0.000029 - momentum: 0.000000
2023-10-13 12:24:12,088 epoch 2 - iter 267/894 - loss 0.18381905 - time (sec): 13.07 - samples/sec: 1929.95 - lr: 0.000029 - momentum: 0.000000
2023-10-13 12:24:16,167 epoch 2 - iter 356/894 - loss 0.18005035 - time (sec): 17.15 - samples/sec: 1986.03 - lr: 0.000029 - momentum: 0.000000
2023-10-13 12:24:20,416 epoch 2 - iter 445/894 - loss 0.17303291 - time (sec): 21.40 - samples/sec: 1970.70 - lr: 0.000028 - momentum: 0.000000
2023-10-13 12:24:24,735 epoch 2 - iter 534/894 - loss 0.17208210 - time (sec): 25.72 - samples/sec: 2001.50 - lr: 0.000028 - momentum: 0.000000
2023-10-13 12:24:28,879 epoch 2 - iter 623/894 - loss 0.16599635 - time (sec): 29.86 - samples/sec: 2002.87 - lr: 0.000028 - momentum: 0.000000
2023-10-13 12:24:33,171 epoch 2 - iter 712/894 - loss 0.16263272 - time (sec): 34.15 - samples/sec: 2024.19 - lr: 0.000027 - momentum: 0.000000
2023-10-13 12:24:37,429 epoch 2 - iter 801/894 - loss 0.15971559 - time (sec): 38.41 - samples/sec: 2027.57 - lr: 0.000027 - momentum: 0.000000
2023-10-13 12:24:41,579 epoch 2 - iter 890/894 - loss 0.15954822 - time (sec): 42.56 - samples/sec: 2024.08 - lr: 0.000027 - momentum: 0.000000
2023-10-13 12:24:41,766 ----------------------------------------------------------------------------------------------------
2023-10-13 12:24:41,766 EPOCH 2 done: loss 0.1594 - lr: 0.000027
2023-10-13 12:24:50,035 DEV : loss 0.13900581002235413 - f1-score (micro avg) 0.717
2023-10-13 12:24:50,066 saving best model
2023-10-13 12:24:50,907 ----------------------------------------------------------------------------------------------------
2023-10-13 12:24:55,312 epoch 3 - iter 89/894 - loss 0.08955154 - time (sec): 4.40 - samples/sec: 1959.52 - lr: 0.000026 - momentum: 0.000000
2023-10-13 12:24:59,630 epoch 3 - iter 178/894 - loss 0.08525409 - time (sec): 8.72 - samples/sec: 2094.68 - lr: 0.000026 - momentum: 0.000000
2023-10-13 12:25:03,783 epoch 3 - iter 267/894 - loss 0.08743891 - time (sec): 12.87 - samples/sec: 2136.94 - lr: 0.000026 - momentum: 0.000000
2023-10-13 12:25:07,837 epoch 3 - iter 356/894 - loss 0.08203209 - time (sec): 16.93 - samples/sec: 2154.98 - lr: 0.000025 - momentum: 0.000000
2023-10-13 12:25:12,061 epoch 3 - iter 445/894 - loss 0.08840063 - time (sec): 21.15 - samples/sec: 2149.93 - lr: 0.000025 - momentum: 0.000000
2023-10-13 12:25:16,072 epoch 3 - iter 534/894 - loss 0.08896715 - time (sec): 25.16 - samples/sec: 2112.47 - lr: 0.000025 - momentum: 0.000000
2023-10-13 12:25:20,538 epoch 3 - iter 623/894 - loss 0.08701667 - time (sec): 29.63 - samples/sec: 2076.88 - lr: 0.000024 - momentum: 0.000000
2023-10-13 12:25:25,148 epoch 3 - iter 712/894 - loss 0.08815847 - time (sec): 34.24 - samples/sec: 2030.04 - lr: 0.000024 - momentum: 0.000000
2023-10-13 12:25:29,745 epoch 3 - iter 801/894 - loss 0.08990165 - time (sec): 38.84 - samples/sec: 2005.45 - lr: 0.000024 - momentum: 0.000000
2023-10-13 12:25:34,431 epoch 3 - iter 890/894 - loss 0.08911325 - time (sec): 43.52 - samples/sec: 1979.02 - lr: 0.000023 - momentum: 0.000000
2023-10-13 12:25:34,625 ----------------------------------------------------------------------------------------------------
2023-10-13 12:25:34,625 EPOCH 3 done: loss 0.0887 - lr: 0.000023
2023-10-13 12:25:43,288 DEV : loss 0.14982320368289948 - f1-score (micro avg) 0.7447
2023-10-13 12:25:43,319 saving best model
2023-10-13 12:25:43,780 ----------------------------------------------------------------------------------------------------
2023-10-13 12:25:47,748 epoch 4 - iter 89/894 - loss 0.04933864 - time (sec): 3.96 - samples/sec: 1926.49 - lr: 0.000023 - momentum: 0.000000
2023-10-13 12:25:52,068 epoch 4 - iter 178/894 - loss 0.04682326 - time (sec): 8.28 - samples/sec: 2051.15 - lr: 0.000023 - momentum: 0.000000
2023-10-13 12:25:56,138 epoch 4 - iter 267/894 - loss 0.05684560 - time (sec): 12.35 - samples/sec: 2049.73 - lr: 0.000022 - momentum: 0.000000
2023-10-13 12:26:00,190 epoch 4 - iter 356/894 - loss 0.05827360 - time (sec): 16.40 - samples/sec: 2070.32 - lr: 0.000022 - momentum: 0.000000
2023-10-13 12:26:04,426 epoch 4 - iter 445/894 - loss 0.05752293 - time (sec): 20.63 - samples/sec: 2014.80 - lr: 0.000022 - momentum: 0.000000
2023-10-13 12:26:09,200 epoch 4 - iter 534/894 - loss 0.05616595 - time (sec): 25.41 - samples/sec: 2044.01 - lr: 0.000021 - momentum: 0.000000
2023-10-13 12:26:13,441 epoch 4 - iter 623/894 - loss 0.05742081 - time (sec): 29.65 - samples/sec: 2040.15 - lr: 0.000021 - momentum: 0.000000
2023-10-13 12:26:17,568 epoch 4 - iter 712/894 - loss 0.05840359 - time (sec): 33.78 - samples/sec: 2031.02 - lr: 0.000021 - momentum: 0.000000
2023-10-13 12:26:21,693 epoch 4 - iter 801/894 - loss 0.05795915 - time (sec): 37.90 - samples/sec: 2052.24 - lr: 0.000020 - momentum: 0.000000
2023-10-13 12:26:25,851 epoch 4 - iter 890/894 - loss 0.05741204 - time (sec): 42.06 - samples/sec: 2050.78 - lr: 0.000020 - momentum: 0.000000
2023-10-13 12:26:26,043 ----------------------------------------------------------------------------------------------------
2023-10-13 12:26:26,043 EPOCH 4 done: loss 0.0572 - lr: 0.000020
2023-10-13 12:26:34,825 DEV : loss 0.16696369647979736 - f1-score (micro avg) 0.7515
2023-10-13 12:26:34,854 saving best model
2023-10-13 12:26:35,218 ----------------------------------------------------------------------------------------------------
2023-10-13 12:26:39,447 epoch 5 - iter 89/894 - loss 0.08290902 - time (sec): 4.23 - samples/sec: 1924.41 - lr: 0.000020 - momentum: 0.000000
2023-10-13 12:26:43,739 epoch 5 - iter 178/894 - loss 0.05714557 - time (sec): 8.52 - samples/sec: 1883.19 - lr: 0.000019 - momentum: 0.000000
2023-10-13 12:26:47,911 epoch 5 - iter 267/894 - loss 0.05015082 - time (sec): 12.69 - samples/sec: 1949.88 - lr: 0.000019 - momentum: 0.000000
2023-10-13 12:26:52,185 epoch 5 - iter 356/894 - loss 0.04668232 - time (sec): 16.97 - samples/sec: 2009.99 - lr: 0.000019 - momentum: 0.000000
2023-10-13 12:26:56,346 epoch 5 - iter 445/894 - loss 0.04376156 - time (sec): 21.13 - samples/sec: 2035.25 - lr: 0.000018 - momentum: 0.000000
2023-10-13 12:27:00,517 epoch 5 - iter 534/894 - loss 0.04233902 - time (sec): 25.30 - samples/sec: 2032.31 - lr: 0.000018 - momentum: 0.000000
2023-10-13 12:27:04,633 epoch 5 - iter 623/894 - loss 0.04487302 - time (sec): 29.41 - samples/sec: 2055.83 - lr: 0.000018 - momentum: 0.000000
2023-10-13 12:27:08,861 epoch 5 - iter 712/894 - loss 0.04491408 - time (sec): 33.64 - samples/sec: 2075.25 - lr: 0.000017 - momentum: 0.000000
2023-10-13 12:27:12,963 epoch 5 - iter 801/894 - loss 0.04271529 - time (sec): 37.74 - samples/sec: 2078.21 - lr: 0.000017 - momentum: 0.000000
2023-10-13 12:27:17,117 epoch 5 - iter 890/894 - loss 0.04193496 - time (sec): 41.90 - samples/sec: 2057.49 - lr: 0.000017 - momentum: 0.000000
2023-10-13 12:27:17,315 ----------------------------------------------------------------------------------------------------
2023-10-13 12:27:17,316 EPOCH 5 done: loss 0.0418 - lr: 0.000017
2023-10-13 12:27:26,372 DEV : loss 0.18145021796226501 - f1-score (micro avg) 0.7689
2023-10-13 12:27:26,403 saving best model
2023-10-13 12:27:26,950 ----------------------------------------------------------------------------------------------------
2023-10-13 12:27:31,158 epoch 6 - iter 89/894 - loss 0.03384747 - time (sec): 4.21 - samples/sec: 2063.80 - lr: 0.000016 - momentum: 0.000000
2023-10-13 12:27:35,381 epoch 6 - iter 178/894 - loss 0.03067694 - time (sec): 8.43 - samples/sec: 2009.41 - lr: 0.000016 - momentum: 0.000000
2023-10-13 12:27:39,388 epoch 6 - iter 267/894 - loss 0.02591000 - time (sec): 12.44 - samples/sec: 2017.20 - lr: 0.000016 - momentum: 0.000000
2023-10-13 12:27:43,453 epoch 6 - iter 356/894 - loss 0.02772844 - time (sec): 16.50 - samples/sec: 2044.51 - lr: 0.000015 - momentum: 0.000000
2023-10-13 12:27:47,471 epoch 6 - iter 445/894 - loss 0.02576901 - time (sec): 20.52 - samples/sec: 2046.17 - lr: 0.000015 - momentum: 0.000000
2023-10-13 12:27:51,589 epoch 6 - iter 534/894 - loss 0.02535571 - time (sec): 24.64 - samples/sec: 2047.25 - lr: 0.000015 - momentum: 0.000000
2023-10-13 12:27:55,665 epoch 6 - iter 623/894 - loss 0.02569978 - time (sec): 28.71 - samples/sec: 2039.91 - lr: 0.000014 - momentum: 0.000000
2023-10-13 12:27:59,772 epoch 6 - iter 712/894 - loss 0.02713449 - time (sec): 32.82 - samples/sec: 2040.92 - lr: 0.000014 - momentum: 0.000000
2023-10-13 12:28:03,995 epoch 6 - iter 801/894 - loss 0.02621730 - time (sec): 37.04 - samples/sec: 2050.38 - lr: 0.000014 - momentum: 0.000000
2023-10-13 12:28:08,477 epoch 6 - iter 890/894 - loss 0.02491693 - time (sec): 41.52 - samples/sec: 2070.94 - lr: 0.000013 - momentum: 0.000000
2023-10-13 12:28:08,690 ----------------------------------------------------------------------------------------------------
2023-10-13 12:28:08,690 EPOCH 6 done: loss 0.0249 - lr: 0.000013
2023-10-13 12:28:17,448 DEV : loss 0.19793881475925446 - f1-score (micro avg) 0.7708
2023-10-13 12:28:17,486 saving best model
2023-10-13 12:28:17,952 ----------------------------------------------------------------------------------------------------
2023-10-13 12:28:22,165 epoch 7 - iter 89/894 - loss 0.01290534 - time (sec): 4.21 - samples/sec: 2063.90 - lr: 0.000013 - momentum: 0.000000
2023-10-13 12:28:26,299 epoch 7 - iter 178/894 - loss 0.01077333 - time (sec): 8.34 - samples/sec: 2068.13 - lr: 0.000013 - momentum: 0.000000
2023-10-13 12:28:30,531 epoch 7 - iter 267/894 - loss 0.01262376 - time (sec): 12.58 - samples/sec: 2091.58 - lr: 0.000012 - momentum: 0.000000
2023-10-13 12:28:34,912 epoch 7 - iter 356/894 - loss 0.01632707 - time (sec): 16.96 - samples/sec: 2069.76 - lr: 0.000012 - momentum: 0.000000
2023-10-13 12:28:38,942 epoch 7 - iter 445/894 - loss 0.01462341 - time (sec): 20.99 - samples/sec: 2058.62 - lr: 0.000012 - momentum: 0.000000
2023-10-13 12:28:43,173 epoch 7 - iter 534/894 - loss 0.01445585 - time (sec): 25.22 - samples/sec: 2054.37 - lr: 0.000011 - momentum: 0.000000
2023-10-13 12:28:47,307 epoch 7 - iter 623/894 - loss 0.01536119 - time (sec): 29.35 - samples/sec: 2046.50 - lr: 0.000011 - momentum: 0.000000
2023-10-13 12:28:51,469 epoch 7 - iter 712/894 - loss 0.01608270 - time (sec): 33.51 - samples/sec: 2042.72 - lr: 0.000011 - momentum: 0.000000
2023-10-13 12:28:55,582 epoch 7 - iter 801/894 - loss 0.01607712 - time (sec): 37.63 - samples/sec: 2025.83 - lr: 0.000010 - momentum: 0.000000
2023-10-13 12:29:00,577 epoch 7 - iter 890/894 - loss 0.01659841 - time (sec): 42.62 - samples/sec: 2019.13 - lr: 0.000010 - momentum: 0.000000
2023-10-13 12:29:00,789 ----------------------------------------------------------------------------------------------------
2023-10-13 12:29:00,789 EPOCH 7 done: loss 0.0165 - lr: 0.000010
2023-10-13 12:29:09,562 DEV : loss 0.21279636025428772 - f1-score (micro avg) 0.7867
2023-10-13 12:29:09,594 saving best model
2023-10-13 12:29:10,053 ----------------------------------------------------------------------------------------------------
2023-10-13 12:29:14,068 epoch 8 - iter 89/894 - loss 0.00681915 - time (sec): 4.01 - samples/sec: 2093.44 - lr: 0.000010 - momentum: 0.000000
2023-10-13 12:29:18,659 epoch 8 - iter 178/894 - loss 0.01159880 - time (sec): 8.60 - samples/sec: 2102.46 - lr: 0.000009 - momentum: 0.000000
2023-10-13 12:29:22,841 epoch 8 - iter 267/894 - loss 0.01058804 - time (sec): 12.78 - samples/sec: 2064.98 - lr: 0.000009 - momentum: 0.000000
2023-10-13 12:29:27,156 epoch 8 - iter 356/894 - loss 0.00878485 - time (sec): 17.10 - samples/sec: 2048.58 - lr: 0.000009 - momentum: 0.000000
2023-10-13 12:29:31,282 epoch 8 - iter 445/894 - loss 0.00963028 - time (sec): 21.22 - samples/sec: 2016.59 - lr: 0.000008 - momentum: 0.000000
2023-10-13 12:29:35,452 epoch 8 - iter 534/894 - loss 0.00874697 - time (sec): 25.39 - samples/sec: 2038.40 - lr: 0.000008 - momentum: 0.000000
2023-10-13 12:29:39,587 epoch 8 - iter 623/894 - loss 0.00881807 - time (sec): 29.53 - samples/sec: 2052.68 - lr: 0.000008 - momentum: 0.000000
2023-10-13 12:29:43,671 epoch 8 - iter 712/894 - loss 0.00946149 - time (sec): 33.61 - samples/sec: 2049.34 - lr: 0.000007 - momentum: 0.000000
2023-10-13 12:29:47,674 epoch 8 - iter 801/894 - loss 0.00989863 - time (sec): 37.62 - samples/sec: 2060.11 - lr: 0.000007 - momentum: 0.000000
2023-10-13 12:29:51,882 epoch 8 - iter 890/894 - loss 0.00969459 - time (sec): 41.82 - samples/sec: 2060.95 - lr: 0.000007 - momentum: 0.000000
2023-10-13 12:29:52,062 ----------------------------------------------------------------------------------------------------
2023-10-13 12:29:52,062 EPOCH 8 done: loss 0.0097 - lr: 0.000007
2023-10-13 12:30:00,901 DEV : loss 0.21466292440891266 - f1-score (micro avg) 0.7978
2023-10-13 12:30:00,932 saving best model
2023-10-13 12:30:01,395 ----------------------------------------------------------------------------------------------------
2023-10-13 12:30:05,934 epoch 9 - iter 89/894 - loss 0.00422191 - time (sec): 4.54 - samples/sec: 1907.01 - lr: 0.000006 - momentum: 0.000000
2023-10-13 12:30:10,054 epoch 9 - iter 178/894 - loss 0.00542088 - time (sec): 8.66 - samples/sec: 1970.11 - lr: 0.000006 - momentum: 0.000000
2023-10-13 12:30:14,177 epoch 9 - iter 267/894 - loss 0.00600949 - time (sec): 12.78 - samples/sec: 1982.58 - lr: 0.000006 - momentum: 0.000000
2023-10-13 12:30:18,237 epoch 9 - iter 356/894 - loss 0.00585206 - time (sec): 16.84 - samples/sec: 2022.13 - lr: 0.000005 - momentum: 0.000000
2023-10-13 12:30:22,725 epoch 9 - iter 445/894 - loss 0.00653931 - time (sec): 21.33 - samples/sec: 2054.47 - lr: 0.000005 - momentum: 0.000000
2023-10-13 12:30:26,910 epoch 9 - iter 534/894 - loss 0.00584633 - time (sec): 25.51 - samples/sec: 2046.84 - lr: 0.000005 - momentum: 0.000000
2023-10-13 12:30:31,048 epoch 9 - iter 623/894 - loss 0.00599610 - time (sec): 29.65 - samples/sec: 2043.13 - lr: 0.000004 - momentum: 0.000000
2023-10-13 12:30:35,839 epoch 9 - iter 712/894 - loss 0.00549033 - time (sec): 34.44 - samples/sec: 2017.01 - lr: 0.000004 - momentum: 0.000000
2023-10-13 12:30:40,447 epoch 9 - iter 801/894 - loss 0.00560849 - time (sec): 39.05 - samples/sec: 1985.39 - lr: 0.000004 - momentum: 0.000000
2023-10-13 12:30:45,313 epoch 9 - iter 890/894 - loss 0.00613256 - time (sec): 43.92 - samples/sec: 1962.55 - lr: 0.000003 - momentum: 0.000000
2023-10-13 12:30:45,531 ----------------------------------------------------------------------------------------------------
2023-10-13 12:30:45,531 EPOCH 9 done: loss 0.0063 - lr: 0.000003
2023-10-13 12:30:54,311 DEV : loss 0.2354293167591095 - f1-score (micro avg) 0.7952
2023-10-13 12:30:54,341 ----------------------------------------------------------------------------------------------------
2023-10-13 12:30:58,510 epoch 10 - iter 89/894 - loss 0.00042234 - time (sec): 4.17 - samples/sec: 2217.34 - lr: 0.000003 - momentum: 0.000000
2023-10-13 12:31:02,778 epoch 10 - iter 178/894 - loss 0.00238956 - time (sec): 8.44 - samples/sec: 2054.95 - lr: 0.000003 - momentum: 0.000000
2023-10-13 12:31:06,822 epoch 10 - iter 267/894 - loss 0.00321158 - time (sec): 12.48 - samples/sec: 2063.83 - lr: 0.000002 - momentum: 0.000000
2023-10-13 12:31:11,256 epoch 10 - iter 356/894 - loss 0.00253060 - time (sec): 16.91 - samples/sec: 2101.06 - lr: 0.000002 - momentum: 0.000000
2023-10-13 12:31:15,638 epoch 10 - iter 445/894 - loss 0.00264548 - time (sec): 21.30 - samples/sec: 2061.39 - lr: 0.000002 - momentum: 0.000000
2023-10-13 12:31:20,218 epoch 10 - iter 534/894 - loss 0.00484149 - time (sec): 25.88 - samples/sec: 2025.23 - lr: 0.000001 - momentum: 0.000000
2023-10-13 12:31:24,200 epoch 10 - iter 623/894 - loss 0.00448904 - time (sec): 29.86 - samples/sec: 2013.87 - lr: 0.000001 - momentum: 0.000000
2023-10-13 12:31:28,368 epoch 10 - iter 712/894 - loss 0.00433090 - time (sec): 34.03 - samples/sec: 2021.05 - lr: 0.000001 - momentum: 0.000000
2023-10-13 12:31:32,402 epoch 10 - iter 801/894 - loss 0.00434765 - time (sec): 38.06 - samples/sec: 2021.53 - lr: 0.000000 - momentum: 0.000000
2023-10-13 12:31:36,864 epoch 10 - iter 890/894 - loss 0.00420725 - time (sec): 42.52 - samples/sec: 2028.32 - lr: 0.000000 - momentum: 0.000000
2023-10-13 12:31:37,058 ----------------------------------------------------------------------------------------------------
2023-10-13 12:31:37,059 EPOCH 10 done: loss 0.0042 - lr: 0.000000
2023-10-13 12:31:45,825 DEV : loss 0.22871056199073792 - f1-score (micro avg) 0.7924
2023-10-13 12:31:46,193 ----------------------------------------------------------------------------------------------------
2023-10-13 12:31:46,194 Loading model from best epoch ...
2023-10-13 12:31:47,848 SequenceTagger predicts: Dictionary with 21 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org, S-prod, B-prod, E-prod, I-prod, S-time, B-time, E-time, I-time
2023-10-13 12:31:52,388
Results:
- F-score (micro) 0.746
- F-score (macro) 0.6695
- Accuracy 0.6143
By class:
precision recall f1-score support
loc 0.8104 0.8607 0.8348 596
pers 0.6807 0.7297 0.7043 333
org 0.5285 0.4924 0.5098 132
prod 0.7368 0.4242 0.5385 66
time 0.7451 0.7755 0.7600 49
micro avg 0.7379 0.7543 0.7460 1176
macro avg 0.7003 0.6565 0.6695 1176
weighted avg 0.7352 0.7543 0.7416 1176
2023-10-13 12:31:52,388 ----------------------------------------------------------------------------------------------------
|