File size: 24,099 Bytes
a84b49a |
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 11:51:21,482 ----------------------------------------------------------------------------------------------------
2023-10-13 11:51:21,483 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 11:51:21,483 ----------------------------------------------------------------------------------------------------
2023-10-13 11:51:21,484 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 11:51:21,484 ----------------------------------------------------------------------------------------------------
2023-10-13 11:51:21,484 Train: 3575 sentences
2023-10-13 11:51:21,484 (train_with_dev=False, train_with_test=False)
2023-10-13 11:51:21,484 ----------------------------------------------------------------------------------------------------
2023-10-13 11:51:21,484 Training Params:
2023-10-13 11:51:21,484 - learning_rate: "3e-05"
2023-10-13 11:51:21,484 - mini_batch_size: "4"
2023-10-13 11:51:21,484 - max_epochs: "10"
2023-10-13 11:51:21,484 - shuffle: "True"
2023-10-13 11:51:21,484 ----------------------------------------------------------------------------------------------------
2023-10-13 11:51:21,484 Plugins:
2023-10-13 11:51:21,484 - LinearScheduler | warmup_fraction: '0.1'
2023-10-13 11:51:21,484 ----------------------------------------------------------------------------------------------------
2023-10-13 11:51:21,484 Final evaluation on model from best epoch (best-model.pt)
2023-10-13 11:51:21,484 - metric: "('micro avg', 'f1-score')"
2023-10-13 11:51:21,484 ----------------------------------------------------------------------------------------------------
2023-10-13 11:51:21,484 Computation:
2023-10-13 11:51:21,484 - compute on device: cuda:0
2023-10-13 11:51:21,484 - embedding storage: none
2023-10-13 11:51:21,484 ----------------------------------------------------------------------------------------------------
2023-10-13 11:51:21,484 Model training base path: "hmbench-hipe2020/de-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-13 11:51:21,484 ----------------------------------------------------------------------------------------------------
2023-10-13 11:51:21,485 ----------------------------------------------------------------------------------------------------
2023-10-13 11:51:25,904 epoch 1 - iter 89/894 - loss 2.95964861 - time (sec): 4.42 - samples/sec: 2164.33 - lr: 0.000003 - momentum: 0.000000
2023-10-13 11:51:29,956 epoch 1 - iter 178/894 - loss 2.03271239 - time (sec): 8.47 - samples/sec: 2071.90 - lr: 0.000006 - momentum: 0.000000
2023-10-13 11:51:33,863 epoch 1 - iter 267/894 - loss 1.55405415 - time (sec): 12.38 - samples/sec: 2043.74 - lr: 0.000009 - momentum: 0.000000
2023-10-13 11:51:38,226 epoch 1 - iter 356/894 - loss 1.26339446 - time (sec): 16.74 - samples/sec: 2022.31 - lr: 0.000012 - momentum: 0.000000
2023-10-13 11:51:43,014 epoch 1 - iter 445/894 - loss 1.08674410 - time (sec): 21.53 - samples/sec: 1951.65 - lr: 0.000015 - momentum: 0.000000
2023-10-13 11:51:47,265 epoch 1 - iter 534/894 - loss 0.95226748 - time (sec): 25.78 - samples/sec: 1976.18 - lr: 0.000018 - momentum: 0.000000
2023-10-13 11:51:51,515 epoch 1 - iter 623/894 - loss 0.85745994 - time (sec): 30.03 - samples/sec: 1982.65 - lr: 0.000021 - momentum: 0.000000
2023-10-13 11:51:56,037 epoch 1 - iter 712/894 - loss 0.77246236 - time (sec): 34.55 - samples/sec: 2000.83 - lr: 0.000024 - momentum: 0.000000
2023-10-13 11:52:00,247 epoch 1 - iter 801/894 - loss 0.71803691 - time (sec): 38.76 - samples/sec: 1989.55 - lr: 0.000027 - momentum: 0.000000
2023-10-13 11:52:04,660 epoch 1 - iter 890/894 - loss 0.67103102 - time (sec): 43.17 - samples/sec: 1994.21 - lr: 0.000030 - momentum: 0.000000
2023-10-13 11:52:04,875 ----------------------------------------------------------------------------------------------------
2023-10-13 11:52:04,875 EPOCH 1 done: loss 0.6686 - lr: 0.000030
2023-10-13 11:52:09,723 DEV : loss 0.19651389122009277 - f1-score (micro avg) 0.5309
2023-10-13 11:52:09,750 saving best model
2023-10-13 11:52:10,106 ----------------------------------------------------------------------------------------------------
2023-10-13 11:52:14,483 epoch 2 - iter 89/894 - loss 0.17874743 - time (sec): 4.38 - samples/sec: 1975.04 - lr: 0.000030 - momentum: 0.000000
2023-10-13 11:52:18,914 epoch 2 - iter 178/894 - loss 0.19245308 - time (sec): 8.81 - samples/sec: 1960.17 - lr: 0.000029 - momentum: 0.000000
2023-10-13 11:52:23,050 epoch 2 - iter 267/894 - loss 0.19114956 - time (sec): 12.94 - samples/sec: 2005.26 - lr: 0.000029 - momentum: 0.000000
2023-10-13 11:52:27,558 epoch 2 - iter 356/894 - loss 0.18340530 - time (sec): 17.45 - samples/sec: 1969.09 - lr: 0.000029 - momentum: 0.000000
2023-10-13 11:52:31,687 epoch 2 - iter 445/894 - loss 0.17962511 - time (sec): 21.58 - samples/sec: 1964.32 - lr: 0.000028 - momentum: 0.000000
2023-10-13 11:52:35,844 epoch 2 - iter 534/894 - loss 0.16974435 - time (sec): 25.74 - samples/sec: 1987.80 - lr: 0.000028 - momentum: 0.000000
2023-10-13 11:52:39,913 epoch 2 - iter 623/894 - loss 0.16649731 - time (sec): 29.81 - samples/sec: 2025.68 - lr: 0.000028 - momentum: 0.000000
2023-10-13 11:52:44,091 epoch 2 - iter 712/894 - loss 0.16541807 - time (sec): 33.98 - samples/sec: 2020.71 - lr: 0.000027 - momentum: 0.000000
2023-10-13 11:52:48,107 epoch 2 - iter 801/894 - loss 0.16505264 - time (sec): 38.00 - samples/sec: 2021.06 - lr: 0.000027 - momentum: 0.000000
2023-10-13 11:52:52,575 epoch 2 - iter 890/894 - loss 0.16086628 - time (sec): 42.47 - samples/sec: 2030.90 - lr: 0.000027 - momentum: 0.000000
2023-10-13 11:52:52,761 ----------------------------------------------------------------------------------------------------
2023-10-13 11:52:52,761 EPOCH 2 done: loss 0.1606 - lr: 0.000027
2023-10-13 11:53:01,183 DEV : loss 0.14430980384349823 - f1-score (micro avg) 0.7107
2023-10-13 11:53:01,211 saving best model
2023-10-13 11:53:01,665 ----------------------------------------------------------------------------------------------------
2023-10-13 11:53:05,704 epoch 3 - iter 89/894 - loss 0.09107293 - time (sec): 4.04 - samples/sec: 1935.11 - lr: 0.000026 - momentum: 0.000000
2023-10-13 11:53:09,734 epoch 3 - iter 178/894 - loss 0.08954741 - time (sec): 8.07 - samples/sec: 2005.89 - lr: 0.000026 - momentum: 0.000000
2023-10-13 11:53:14,016 epoch 3 - iter 267/894 - loss 0.09709904 - time (sec): 12.35 - samples/sec: 1972.74 - lr: 0.000026 - momentum: 0.000000
2023-10-13 11:53:18,466 epoch 3 - iter 356/894 - loss 0.09035845 - time (sec): 16.80 - samples/sec: 1978.54 - lr: 0.000025 - momentum: 0.000000
2023-10-13 11:53:22,769 epoch 3 - iter 445/894 - loss 0.09195587 - time (sec): 21.10 - samples/sec: 2011.97 - lr: 0.000025 - momentum: 0.000000
2023-10-13 11:53:26,791 epoch 3 - iter 534/894 - loss 0.09116096 - time (sec): 25.12 - samples/sec: 2044.47 - lr: 0.000025 - momentum: 0.000000
2023-10-13 11:53:30,992 epoch 3 - iter 623/894 - loss 0.09292099 - time (sec): 29.32 - samples/sec: 2045.35 - lr: 0.000024 - momentum: 0.000000
2023-10-13 11:53:35,224 epoch 3 - iter 712/894 - loss 0.09454837 - time (sec): 33.56 - samples/sec: 2041.67 - lr: 0.000024 - momentum: 0.000000
2023-10-13 11:53:39,463 epoch 3 - iter 801/894 - loss 0.09328361 - time (sec): 37.80 - samples/sec: 2036.11 - lr: 0.000024 - momentum: 0.000000
2023-10-13 11:53:43,927 epoch 3 - iter 890/894 - loss 0.09323024 - time (sec): 42.26 - samples/sec: 2040.91 - lr: 0.000023 - momentum: 0.000000
2023-10-13 11:53:44,111 ----------------------------------------------------------------------------------------------------
2023-10-13 11:53:44,111 EPOCH 3 done: loss 0.0930 - lr: 0.000023
2023-10-13 11:53:52,652 DEV : loss 0.15164530277252197 - f1-score (micro avg) 0.7297
2023-10-13 11:53:52,681 saving best model
2023-10-13 11:53:53,170 ----------------------------------------------------------------------------------------------------
2023-10-13 11:53:57,485 epoch 4 - iter 89/894 - loss 0.06575092 - time (sec): 4.31 - samples/sec: 2092.37 - lr: 0.000023 - momentum: 0.000000
2023-10-13 11:54:01,695 epoch 4 - iter 178/894 - loss 0.05914120 - time (sec): 8.52 - samples/sec: 2014.16 - lr: 0.000023 - momentum: 0.000000
2023-10-13 11:54:06,062 epoch 4 - iter 267/894 - loss 0.05800368 - time (sec): 12.89 - samples/sec: 2038.84 - lr: 0.000022 - momentum: 0.000000
2023-10-13 11:54:10,598 epoch 4 - iter 356/894 - loss 0.05785617 - time (sec): 17.42 - samples/sec: 2074.47 - lr: 0.000022 - momentum: 0.000000
2023-10-13 11:54:14,848 epoch 4 - iter 445/894 - loss 0.05338580 - time (sec): 21.67 - samples/sec: 2068.37 - lr: 0.000022 - momentum: 0.000000
2023-10-13 11:54:19,056 epoch 4 - iter 534/894 - loss 0.05470941 - time (sec): 25.88 - samples/sec: 2069.24 - lr: 0.000021 - momentum: 0.000000
2023-10-13 11:54:23,203 epoch 4 - iter 623/894 - loss 0.05504567 - time (sec): 30.03 - samples/sec: 2059.01 - lr: 0.000021 - momentum: 0.000000
2023-10-13 11:54:27,507 epoch 4 - iter 712/894 - loss 0.05336567 - time (sec): 34.33 - samples/sec: 2046.02 - lr: 0.000021 - momentum: 0.000000
2023-10-13 11:54:31,882 epoch 4 - iter 801/894 - loss 0.05371459 - time (sec): 38.71 - samples/sec: 2005.64 - lr: 0.000020 - momentum: 0.000000
2023-10-13 11:54:36,420 epoch 4 - iter 890/894 - loss 0.05405224 - time (sec): 43.24 - samples/sec: 1993.65 - lr: 0.000020 - momentum: 0.000000
2023-10-13 11:54:36,623 ----------------------------------------------------------------------------------------------------
2023-10-13 11:54:36,623 EPOCH 4 done: loss 0.0540 - lr: 0.000020
2023-10-13 11:54:45,586 DEV : loss 0.16767126321792603 - f1-score (micro avg) 0.7435
2023-10-13 11:54:45,629 saving best model
2023-10-13 11:54:46,202 ----------------------------------------------------------------------------------------------------
2023-10-13 11:54:51,104 epoch 5 - iter 89/894 - loss 0.03833014 - time (sec): 4.90 - samples/sec: 1983.32 - lr: 0.000020 - momentum: 0.000000
2023-10-13 11:54:55,089 epoch 5 - iter 178/894 - loss 0.04134649 - time (sec): 8.89 - samples/sec: 1998.30 - lr: 0.000019 - momentum: 0.000000
2023-10-13 11:54:59,154 epoch 5 - iter 267/894 - loss 0.04407342 - time (sec): 12.95 - samples/sec: 2053.09 - lr: 0.000019 - momentum: 0.000000
2023-10-13 11:55:03,033 epoch 5 - iter 356/894 - loss 0.04265799 - time (sec): 16.83 - samples/sec: 2061.37 - lr: 0.000019 - momentum: 0.000000
2023-10-13 11:55:07,321 epoch 5 - iter 445/894 - loss 0.03938600 - time (sec): 21.12 - samples/sec: 2070.49 - lr: 0.000018 - momentum: 0.000000
2023-10-13 11:55:11,359 epoch 5 - iter 534/894 - loss 0.03873257 - time (sec): 25.16 - samples/sec: 2086.45 - lr: 0.000018 - momentum: 0.000000
2023-10-13 11:55:15,767 epoch 5 - iter 623/894 - loss 0.03834078 - time (sec): 29.56 - samples/sec: 2053.11 - lr: 0.000018 - momentum: 0.000000
2023-10-13 11:55:20,229 epoch 5 - iter 712/894 - loss 0.03793415 - time (sec): 34.03 - samples/sec: 2041.73 - lr: 0.000017 - momentum: 0.000000
2023-10-13 11:55:24,623 epoch 5 - iter 801/894 - loss 0.03750295 - time (sec): 38.42 - samples/sec: 2023.69 - lr: 0.000017 - momentum: 0.000000
2023-10-13 11:55:28,852 epoch 5 - iter 890/894 - loss 0.03928703 - time (sec): 42.65 - samples/sec: 2021.24 - lr: 0.000017 - momentum: 0.000000
2023-10-13 11:55:29,052 ----------------------------------------------------------------------------------------------------
2023-10-13 11:55:29,053 EPOCH 5 done: loss 0.0391 - lr: 0.000017
2023-10-13 11:55:37,512 DEV : loss 0.18989010155200958 - f1-score (micro avg) 0.7799
2023-10-13 11:55:37,539 saving best model
2023-10-13 11:55:38,022 ----------------------------------------------------------------------------------------------------
2023-10-13 11:55:42,488 epoch 6 - iter 89/894 - loss 0.01887461 - time (sec): 4.46 - samples/sec: 1941.41 - lr: 0.000016 - momentum: 0.000000
2023-10-13 11:55:46,559 epoch 6 - iter 178/894 - loss 0.02591767 - time (sec): 8.54 - samples/sec: 1919.28 - lr: 0.000016 - momentum: 0.000000
2023-10-13 11:55:51,042 epoch 6 - iter 267/894 - loss 0.02075761 - time (sec): 13.02 - samples/sec: 1962.56 - lr: 0.000016 - momentum: 0.000000
2023-10-13 11:55:55,321 epoch 6 - iter 356/894 - loss 0.01989266 - time (sec): 17.30 - samples/sec: 1991.06 - lr: 0.000015 - momentum: 0.000000
2023-10-13 11:55:59,294 epoch 6 - iter 445/894 - loss 0.01956611 - time (sec): 21.27 - samples/sec: 1975.13 - lr: 0.000015 - momentum: 0.000000
2023-10-13 11:56:03,300 epoch 6 - iter 534/894 - loss 0.02114777 - time (sec): 25.28 - samples/sec: 1989.27 - lr: 0.000015 - momentum: 0.000000
2023-10-13 11:56:07,317 epoch 6 - iter 623/894 - loss 0.02490253 - time (sec): 29.29 - samples/sec: 1985.00 - lr: 0.000014 - momentum: 0.000000
2023-10-13 11:56:11,703 epoch 6 - iter 712/894 - loss 0.02394661 - time (sec): 33.68 - samples/sec: 2029.61 - lr: 0.000014 - momentum: 0.000000
2023-10-13 11:56:15,930 epoch 6 - iter 801/894 - loss 0.02522246 - time (sec): 37.91 - samples/sec: 2028.41 - lr: 0.000014 - momentum: 0.000000
2023-10-13 11:56:20,141 epoch 6 - iter 890/894 - loss 0.02648157 - time (sec): 42.12 - samples/sec: 2044.57 - lr: 0.000013 - momentum: 0.000000
2023-10-13 11:56:20,322 ----------------------------------------------------------------------------------------------------
2023-10-13 11:56:20,322 EPOCH 6 done: loss 0.0267 - lr: 0.000013
2023-10-13 11:56:28,672 DEV : loss 0.20904241502285004 - f1-score (micro avg) 0.7844
2023-10-13 11:56:28,699 saving best model
2023-10-13 11:56:29,218 ----------------------------------------------------------------------------------------------------
2023-10-13 11:56:33,169 epoch 7 - iter 89/894 - loss 0.01644043 - time (sec): 3.95 - samples/sec: 2225.26 - lr: 0.000013 - momentum: 0.000000
2023-10-13 11:56:37,137 epoch 7 - iter 178/894 - loss 0.01652289 - time (sec): 7.92 - samples/sec: 2160.68 - lr: 0.000013 - momentum: 0.000000
2023-10-13 11:56:41,572 epoch 7 - iter 267/894 - loss 0.01444836 - time (sec): 12.35 - samples/sec: 2206.43 - lr: 0.000012 - momentum: 0.000000
2023-10-13 11:56:45,757 epoch 7 - iter 356/894 - loss 0.01344021 - time (sec): 16.54 - samples/sec: 2154.40 - lr: 0.000012 - momentum: 0.000000
2023-10-13 11:56:50,038 epoch 7 - iter 445/894 - loss 0.01503458 - time (sec): 20.82 - samples/sec: 2132.69 - lr: 0.000012 - momentum: 0.000000
2023-10-13 11:56:54,094 epoch 7 - iter 534/894 - loss 0.01470901 - time (sec): 24.87 - samples/sec: 2114.67 - lr: 0.000011 - momentum: 0.000000
2023-10-13 11:56:58,166 epoch 7 - iter 623/894 - loss 0.01436044 - time (sec): 28.95 - samples/sec: 2105.78 - lr: 0.000011 - momentum: 0.000000
2023-10-13 11:57:02,335 epoch 7 - iter 712/894 - loss 0.01450503 - time (sec): 33.12 - samples/sec: 2089.24 - lr: 0.000011 - momentum: 0.000000
2023-10-13 11:57:06,532 epoch 7 - iter 801/894 - loss 0.01496761 - time (sec): 37.31 - samples/sec: 2065.77 - lr: 0.000010 - momentum: 0.000000
2023-10-13 11:57:11,096 epoch 7 - iter 890/894 - loss 0.01642279 - time (sec): 41.88 - samples/sec: 2060.17 - lr: 0.000010 - momentum: 0.000000
2023-10-13 11:57:11,276 ----------------------------------------------------------------------------------------------------
2023-10-13 11:57:11,276 EPOCH 7 done: loss 0.0164 - lr: 0.000010
2023-10-13 11:57:19,774 DEV : loss 0.20529742538928986 - f1-score (micro avg) 0.7771
2023-10-13 11:57:19,803 ----------------------------------------------------------------------------------------------------
2023-10-13 11:57:24,304 epoch 8 - iter 89/894 - loss 0.00943737 - time (sec): 4.50 - samples/sec: 1926.13 - lr: 0.000010 - momentum: 0.000000
2023-10-13 11:57:28,703 epoch 8 - iter 178/894 - loss 0.00873977 - time (sec): 8.90 - samples/sec: 1997.92 - lr: 0.000009 - momentum: 0.000000
2023-10-13 11:57:32,928 epoch 8 - iter 267/894 - loss 0.00852614 - time (sec): 13.12 - samples/sec: 2048.20 - lr: 0.000009 - momentum: 0.000000
2023-10-13 11:57:37,165 epoch 8 - iter 356/894 - loss 0.00785977 - time (sec): 17.36 - samples/sec: 2099.61 - lr: 0.000009 - momentum: 0.000000
2023-10-13 11:57:41,318 epoch 8 - iter 445/894 - loss 0.01056320 - time (sec): 21.51 - samples/sec: 2052.03 - lr: 0.000008 - momentum: 0.000000
2023-10-13 11:57:45,668 epoch 8 - iter 534/894 - loss 0.01160587 - time (sec): 25.86 - samples/sec: 2034.90 - lr: 0.000008 - momentum: 0.000000
2023-10-13 11:57:49,941 epoch 8 - iter 623/894 - loss 0.01157603 - time (sec): 30.14 - samples/sec: 2046.21 - lr: 0.000008 - momentum: 0.000000
2023-10-13 11:57:54,347 epoch 8 - iter 712/894 - loss 0.01161656 - time (sec): 34.54 - samples/sec: 2024.50 - lr: 0.000007 - momentum: 0.000000
2023-10-13 11:57:58,702 epoch 8 - iter 801/894 - loss 0.01109661 - time (sec): 38.90 - samples/sec: 2007.73 - lr: 0.000007 - momentum: 0.000000
2023-10-13 11:58:03,248 epoch 8 - iter 890/894 - loss 0.01081817 - time (sec): 43.44 - samples/sec: 1983.66 - lr: 0.000007 - momentum: 0.000000
2023-10-13 11:58:03,431 ----------------------------------------------------------------------------------------------------
2023-10-13 11:58:03,431 EPOCH 8 done: loss 0.0108 - lr: 0.000007
2023-10-13 11:58:11,986 DEV : loss 0.22993013262748718 - f1-score (micro avg) 0.7863
2023-10-13 11:58:12,017 saving best model
2023-10-13 11:58:12,506 ----------------------------------------------------------------------------------------------------
2023-10-13 11:58:16,908 epoch 9 - iter 89/894 - loss 0.01082872 - time (sec): 4.40 - samples/sec: 1878.92 - lr: 0.000006 - momentum: 0.000000
2023-10-13 11:58:21,317 epoch 9 - iter 178/894 - loss 0.00801392 - time (sec): 8.81 - samples/sec: 1995.09 - lr: 0.000006 - momentum: 0.000000
2023-10-13 11:58:25,532 epoch 9 - iter 267/894 - loss 0.00706807 - time (sec): 13.02 - samples/sec: 1987.89 - lr: 0.000006 - momentum: 0.000000
2023-10-13 11:58:29,789 epoch 9 - iter 356/894 - loss 0.00604321 - time (sec): 17.28 - samples/sec: 2045.69 - lr: 0.000005 - momentum: 0.000000
2023-10-13 11:58:34,164 epoch 9 - iter 445/894 - loss 0.00629406 - time (sec): 21.66 - samples/sec: 2040.71 - lr: 0.000005 - momentum: 0.000000
2023-10-13 11:58:38,525 epoch 9 - iter 534/894 - loss 0.00647986 - time (sec): 26.02 - samples/sec: 2054.88 - lr: 0.000005 - momentum: 0.000000
2023-10-13 11:58:42,646 epoch 9 - iter 623/894 - loss 0.00640126 - time (sec): 30.14 - samples/sec: 2057.71 - lr: 0.000004 - momentum: 0.000000
2023-10-13 11:58:46,740 epoch 9 - iter 712/894 - loss 0.00619220 - time (sec): 34.23 - samples/sec: 2045.28 - lr: 0.000004 - momentum: 0.000000
2023-10-13 11:58:50,747 epoch 9 - iter 801/894 - loss 0.00657126 - time (sec): 38.24 - samples/sec: 2044.79 - lr: 0.000004 - momentum: 0.000000
2023-10-13 11:58:54,968 epoch 9 - iter 890/894 - loss 0.00688760 - time (sec): 42.46 - samples/sec: 2028.36 - lr: 0.000003 - momentum: 0.000000
2023-10-13 11:58:55,162 ----------------------------------------------------------------------------------------------------
2023-10-13 11:58:55,162 EPOCH 9 done: loss 0.0069 - lr: 0.000003
2023-10-13 11:59:03,625 DEV : loss 0.23477818071842194 - f1-score (micro avg) 0.7876
2023-10-13 11:59:03,654 saving best model
2023-10-13 11:59:04,113 ----------------------------------------------------------------------------------------------------
2023-10-13 11:59:08,444 epoch 10 - iter 89/894 - loss 0.01073514 - time (sec): 4.33 - samples/sec: 2029.84 - lr: 0.000003 - momentum: 0.000000
2023-10-13 11:59:12,635 epoch 10 - iter 178/894 - loss 0.00686412 - time (sec): 8.52 - samples/sec: 1962.81 - lr: 0.000003 - momentum: 0.000000
2023-10-13 11:59:16,883 epoch 10 - iter 267/894 - loss 0.00526322 - time (sec): 12.77 - samples/sec: 1980.29 - lr: 0.000002 - momentum: 0.000000
2023-10-13 11:59:20,948 epoch 10 - iter 356/894 - loss 0.00548186 - time (sec): 16.83 - samples/sec: 2024.96 - lr: 0.000002 - momentum: 0.000000
2023-10-13 11:59:25,163 epoch 10 - iter 445/894 - loss 0.00532709 - time (sec): 21.05 - samples/sec: 2055.07 - lr: 0.000002 - momentum: 0.000000
2023-10-13 11:59:29,369 epoch 10 - iter 534/894 - loss 0.00531083 - time (sec): 25.25 - samples/sec: 2067.81 - lr: 0.000001 - momentum: 0.000000
2023-10-13 11:59:33,663 epoch 10 - iter 623/894 - loss 0.00505531 - time (sec): 29.55 - samples/sec: 2062.69 - lr: 0.000001 - momentum: 0.000000
2023-10-13 11:59:37,654 epoch 10 - iter 712/894 - loss 0.00524732 - time (sec): 33.54 - samples/sec: 2063.87 - lr: 0.000001 - momentum: 0.000000
2023-10-13 11:59:41,725 epoch 10 - iter 801/894 - loss 0.00519244 - time (sec): 37.61 - samples/sec: 2050.62 - lr: 0.000000 - momentum: 0.000000
2023-10-13 11:59:45,901 epoch 10 - iter 890/894 - loss 0.00521658 - time (sec): 41.79 - samples/sec: 2062.61 - lr: 0.000000 - momentum: 0.000000
2023-10-13 11:59:46,077 ----------------------------------------------------------------------------------------------------
2023-10-13 11:59:46,077 EPOCH 10 done: loss 0.0052 - lr: 0.000000
2023-10-13 11:59:54,618 DEV : loss 0.2303038239479065 - f1-score (micro avg) 0.7829
2023-10-13 11:59:55,005 ----------------------------------------------------------------------------------------------------
2023-10-13 11:59:55,006 Loading model from best epoch ...
2023-10-13 11:59:56,539 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 12:00:00,808
Results:
- F-score (micro) 0.7548
- F-score (macro) 0.6753
- Accuracy 0.6264
By class:
precision recall f1-score support
loc 0.8098 0.8574 0.8329 596
pers 0.6917 0.7748 0.7309 333
org 0.5826 0.5076 0.5425 132
prod 0.6444 0.4394 0.5225 66
time 0.7400 0.7551 0.7475 49
micro avg 0.7430 0.7670 0.7548 1176
macro avg 0.6937 0.6668 0.6753 1176
weighted avg 0.7387 0.7670 0.7505 1176
2023-10-13 12:00:00,808 ----------------------------------------------------------------------------------------------------
|