File size: 24,040 Bytes
428ca43 |
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 |
2023-10-18 16:48:35,521 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:35,521 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(32001, 128)
(position_embeddings): Embedding(512, 128)
(token_type_embeddings): Embedding(2, 128)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): BertEncoder(
(layer): ModuleList(
(0-1): 2 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=128, out_features=128, bias=True)
(key): Linear(in_features=128, out_features=128, bias=True)
(value): Linear(in_features=128, out_features=128, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=128, out_features=128, bias=True)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=128, out_features=512, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=512, out_features=128, bias=True)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=128, out_features=128, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=128, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-18 16:48:35,521 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:35,521 MultiCorpus: 966 train + 219 dev + 204 test sentences
- NER_HIPE_2022 Corpus: 966 train + 219 dev + 204 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/fr/with_doc_seperator
2023-10-18 16:48:35,521 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:35,521 Train: 966 sentences
2023-10-18 16:48:35,521 (train_with_dev=False, train_with_test=False)
2023-10-18 16:48:35,521 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:35,521 Training Params:
2023-10-18 16:48:35,521 - learning_rate: "3e-05"
2023-10-18 16:48:35,522 - mini_batch_size: "8"
2023-10-18 16:48:35,522 - max_epochs: "10"
2023-10-18 16:48:35,522 - shuffle: "True"
2023-10-18 16:48:35,522 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:35,522 Plugins:
2023-10-18 16:48:35,522 - TensorboardLogger
2023-10-18 16:48:35,522 - LinearScheduler | warmup_fraction: '0.1'
2023-10-18 16:48:35,522 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:35,522 Final evaluation on model from best epoch (best-model.pt)
2023-10-18 16:48:35,522 - metric: "('micro avg', 'f1-score')"
2023-10-18 16:48:35,522 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:35,522 Computation:
2023-10-18 16:48:35,522 - compute on device: cuda:0
2023-10-18 16:48:35,522 - embedding storage: none
2023-10-18 16:48:35,522 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:35,522 Model training base path: "hmbench-ajmc/fr-dbmdz/bert-tiny-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5"
2023-10-18 16:48:35,522 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:35,522 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:35,522 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-18 16:48:35,781 epoch 1 - iter 12/121 - loss 3.73135666 - time (sec): 0.26 - samples/sec: 8989.30 - lr: 0.000003 - momentum: 0.000000
2023-10-18 16:48:36,037 epoch 1 - iter 24/121 - loss 3.70230323 - time (sec): 0.51 - samples/sec: 8365.49 - lr: 0.000006 - momentum: 0.000000
2023-10-18 16:48:36,312 epoch 1 - iter 36/121 - loss 3.64697934 - time (sec): 0.79 - samples/sec: 9056.15 - lr: 0.000009 - momentum: 0.000000
2023-10-18 16:48:36,585 epoch 1 - iter 48/121 - loss 3.64897593 - time (sec): 1.06 - samples/sec: 8930.69 - lr: 0.000012 - momentum: 0.000000
2023-10-18 16:48:36,853 epoch 1 - iter 60/121 - loss 3.61059523 - time (sec): 1.33 - samples/sec: 8967.33 - lr: 0.000015 - momentum: 0.000000
2023-10-18 16:48:37,110 epoch 1 - iter 72/121 - loss 3.52905406 - time (sec): 1.59 - samples/sec: 8849.71 - lr: 0.000018 - momentum: 0.000000
2023-10-18 16:48:37,383 epoch 1 - iter 84/121 - loss 3.40711096 - time (sec): 1.86 - samples/sec: 9028.03 - lr: 0.000021 - momentum: 0.000000
2023-10-18 16:48:37,654 epoch 1 - iter 96/121 - loss 3.27300372 - time (sec): 2.13 - samples/sec: 9245.22 - lr: 0.000024 - momentum: 0.000000
2023-10-18 16:48:37,931 epoch 1 - iter 108/121 - loss 3.11948802 - time (sec): 2.41 - samples/sec: 9258.34 - lr: 0.000027 - momentum: 0.000000
2023-10-18 16:48:38,186 epoch 1 - iter 120/121 - loss 2.98945042 - time (sec): 2.66 - samples/sec: 9256.13 - lr: 0.000030 - momentum: 0.000000
2023-10-18 16:48:38,202 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:38,202 EPOCH 1 done: loss 2.9857 - lr: 0.000030
2023-10-18 16:48:38,475 DEV : loss 0.888546884059906 - f1-score (micro avg) 0.0
2023-10-18 16:48:38,480 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:38,743 epoch 2 - iter 12/121 - loss 1.34267957 - time (sec): 0.26 - samples/sec: 8726.53 - lr: 0.000030 - momentum: 0.000000
2023-10-18 16:48:39,014 epoch 2 - iter 24/121 - loss 1.20006107 - time (sec): 0.53 - samples/sec: 8960.14 - lr: 0.000029 - momentum: 0.000000
2023-10-18 16:48:39,281 epoch 2 - iter 36/121 - loss 1.07692122 - time (sec): 0.80 - samples/sec: 9001.51 - lr: 0.000029 - momentum: 0.000000
2023-10-18 16:48:39,559 epoch 2 - iter 48/121 - loss 1.02313150 - time (sec): 1.08 - samples/sec: 9151.99 - lr: 0.000029 - momentum: 0.000000
2023-10-18 16:48:39,830 epoch 2 - iter 60/121 - loss 0.98698774 - time (sec): 1.35 - samples/sec: 8953.45 - lr: 0.000028 - momentum: 0.000000
2023-10-18 16:48:40,110 epoch 2 - iter 72/121 - loss 0.95003231 - time (sec): 1.63 - samples/sec: 9022.75 - lr: 0.000028 - momentum: 0.000000
2023-10-18 16:48:40,342 epoch 2 - iter 84/121 - loss 0.89889328 - time (sec): 1.86 - samples/sec: 9294.91 - lr: 0.000028 - momentum: 0.000000
2023-10-18 16:48:40,744 epoch 2 - iter 96/121 - loss 0.86386461 - time (sec): 2.26 - samples/sec: 8747.32 - lr: 0.000027 - momentum: 0.000000
2023-10-18 16:48:41,018 epoch 2 - iter 108/121 - loss 0.86218533 - time (sec): 2.54 - samples/sec: 8706.77 - lr: 0.000027 - momentum: 0.000000
2023-10-18 16:48:41,283 epoch 2 - iter 120/121 - loss 0.84499695 - time (sec): 2.80 - samples/sec: 8744.14 - lr: 0.000027 - momentum: 0.000000
2023-10-18 16:48:41,305 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:41,305 EPOCH 2 done: loss 0.8462 - lr: 0.000027
2023-10-18 16:48:41,719 DEV : loss 0.6577260494232178 - f1-score (micro avg) 0.0
2023-10-18 16:48:41,724 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:42,009 epoch 3 - iter 12/121 - loss 0.72147088 - time (sec): 0.28 - samples/sec: 8779.43 - lr: 0.000026 - momentum: 0.000000
2023-10-18 16:48:42,279 epoch 3 - iter 24/121 - loss 0.76590949 - time (sec): 0.56 - samples/sec: 8647.80 - lr: 0.000026 - momentum: 0.000000
2023-10-18 16:48:42,457 epoch 3 - iter 36/121 - loss 0.74445104 - time (sec): 0.73 - samples/sec: 9633.18 - lr: 0.000026 - momentum: 0.000000
2023-10-18 16:48:42,640 epoch 3 - iter 48/121 - loss 0.73264098 - time (sec): 0.92 - samples/sec: 10434.44 - lr: 0.000025 - momentum: 0.000000
2023-10-18 16:48:42,827 epoch 3 - iter 60/121 - loss 0.71998033 - time (sec): 1.10 - samples/sec: 10910.00 - lr: 0.000025 - momentum: 0.000000
2023-10-18 16:48:43,005 epoch 3 - iter 72/121 - loss 0.70956971 - time (sec): 1.28 - samples/sec: 11157.50 - lr: 0.000025 - momentum: 0.000000
2023-10-18 16:48:43,192 epoch 3 - iter 84/121 - loss 0.69546580 - time (sec): 1.47 - samples/sec: 11494.33 - lr: 0.000024 - momentum: 0.000000
2023-10-18 16:48:43,392 epoch 3 - iter 96/121 - loss 0.68506052 - time (sec): 1.67 - samples/sec: 11815.01 - lr: 0.000024 - momentum: 0.000000
2023-10-18 16:48:43,616 epoch 3 - iter 108/121 - loss 0.67453678 - time (sec): 1.89 - samples/sec: 11736.39 - lr: 0.000024 - momentum: 0.000000
2023-10-18 16:48:43,840 epoch 3 - iter 120/121 - loss 0.67556474 - time (sec): 2.12 - samples/sec: 11614.32 - lr: 0.000023 - momentum: 0.000000
2023-10-18 16:48:43,855 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:43,855 EPOCH 3 done: loss 0.6758 - lr: 0.000023
2023-10-18 16:48:44,269 DEV : loss 0.5368312001228333 - f1-score (micro avg) 0.0
2023-10-18 16:48:44,273 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:44,540 epoch 4 - iter 12/121 - loss 0.67001378 - time (sec): 0.27 - samples/sec: 7861.72 - lr: 0.000023 - momentum: 0.000000
2023-10-18 16:48:44,819 epoch 4 - iter 24/121 - loss 0.64732484 - time (sec): 0.55 - samples/sec: 8204.05 - lr: 0.000023 - momentum: 0.000000
2023-10-18 16:48:45,098 epoch 4 - iter 36/121 - loss 0.61911750 - time (sec): 0.82 - samples/sec: 8823.53 - lr: 0.000022 - momentum: 0.000000
2023-10-18 16:48:45,356 epoch 4 - iter 48/121 - loss 0.61389990 - time (sec): 1.08 - samples/sec: 8897.35 - lr: 0.000022 - momentum: 0.000000
2023-10-18 16:48:45,623 epoch 4 - iter 60/121 - loss 0.61282173 - time (sec): 1.35 - samples/sec: 8954.13 - lr: 0.000022 - momentum: 0.000000
2023-10-18 16:48:45,889 epoch 4 - iter 72/121 - loss 0.59776583 - time (sec): 1.61 - samples/sec: 9038.88 - lr: 0.000021 - momentum: 0.000000
2023-10-18 16:48:46,159 epoch 4 - iter 84/121 - loss 0.59023191 - time (sec): 1.89 - samples/sec: 9178.13 - lr: 0.000021 - momentum: 0.000000
2023-10-18 16:48:46,434 epoch 4 - iter 96/121 - loss 0.58243842 - time (sec): 2.16 - samples/sec: 9158.22 - lr: 0.000021 - momentum: 0.000000
2023-10-18 16:48:46,721 epoch 4 - iter 108/121 - loss 0.58528172 - time (sec): 2.45 - samples/sec: 9086.95 - lr: 0.000020 - momentum: 0.000000
2023-10-18 16:48:46,991 epoch 4 - iter 120/121 - loss 0.57638690 - time (sec): 2.72 - samples/sec: 9077.62 - lr: 0.000020 - momentum: 0.000000
2023-10-18 16:48:47,008 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:47,008 EPOCH 4 done: loss 0.5767 - lr: 0.000020
2023-10-18 16:48:47,426 DEV : loss 0.43159720301628113 - f1-score (micro avg) 0.0952
2023-10-18 16:48:47,430 saving best model
2023-10-18 16:48:47,463 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:47,734 epoch 5 - iter 12/121 - loss 0.54535688 - time (sec): 0.27 - samples/sec: 9126.35 - lr: 0.000020 - momentum: 0.000000
2023-10-18 16:48:48,003 epoch 5 - iter 24/121 - loss 0.53077766 - time (sec): 0.54 - samples/sec: 9335.20 - lr: 0.000019 - momentum: 0.000000
2023-10-18 16:48:48,272 epoch 5 - iter 36/121 - loss 0.51623647 - time (sec): 0.81 - samples/sec: 9313.01 - lr: 0.000019 - momentum: 0.000000
2023-10-18 16:48:48,551 epoch 5 - iter 48/121 - loss 0.52815520 - time (sec): 1.09 - samples/sec: 9360.77 - lr: 0.000019 - momentum: 0.000000
2023-10-18 16:48:48,739 epoch 5 - iter 60/121 - loss 0.52803554 - time (sec): 1.27 - samples/sec: 9878.01 - lr: 0.000018 - momentum: 0.000000
2023-10-18 16:48:48,937 epoch 5 - iter 72/121 - loss 0.52301950 - time (sec): 1.47 - samples/sec: 10188.78 - lr: 0.000018 - momentum: 0.000000
2023-10-18 16:48:49,159 epoch 5 - iter 84/121 - loss 0.51809378 - time (sec): 1.69 - samples/sec: 10286.49 - lr: 0.000018 - momentum: 0.000000
2023-10-18 16:48:49,354 epoch 5 - iter 96/121 - loss 0.51630126 - time (sec): 1.89 - samples/sec: 10552.00 - lr: 0.000017 - momentum: 0.000000
2023-10-18 16:48:49,577 epoch 5 - iter 108/121 - loss 0.50390050 - time (sec): 2.11 - samples/sec: 10559.03 - lr: 0.000017 - momentum: 0.000000
2023-10-18 16:48:49,829 epoch 5 - iter 120/121 - loss 0.50184025 - time (sec): 2.36 - samples/sec: 10415.44 - lr: 0.000017 - momentum: 0.000000
2023-10-18 16:48:49,847 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:49,847 EPOCH 5 done: loss 0.5019 - lr: 0.000017
2023-10-18 16:48:50,277 DEV : loss 0.38968807458877563 - f1-score (micro avg) 0.236
2023-10-18 16:48:50,283 saving best model
2023-10-18 16:48:50,320 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:50,611 epoch 6 - iter 12/121 - loss 0.48606714 - time (sec): 0.29 - samples/sec: 8841.20 - lr: 0.000016 - momentum: 0.000000
2023-10-18 16:48:50,929 epoch 6 - iter 24/121 - loss 0.47768069 - time (sec): 0.61 - samples/sec: 8232.48 - lr: 0.000016 - momentum: 0.000000
2023-10-18 16:48:51,229 epoch 6 - iter 36/121 - loss 0.49199054 - time (sec): 0.91 - samples/sec: 8306.29 - lr: 0.000016 - momentum: 0.000000
2023-10-18 16:48:51,507 epoch 6 - iter 48/121 - loss 0.45279491 - time (sec): 1.19 - samples/sec: 8222.37 - lr: 0.000015 - momentum: 0.000000
2023-10-18 16:48:51,781 epoch 6 - iter 60/121 - loss 0.45754327 - time (sec): 1.46 - samples/sec: 8416.10 - lr: 0.000015 - momentum: 0.000000
2023-10-18 16:48:52,049 epoch 6 - iter 72/121 - loss 0.46451499 - time (sec): 1.73 - samples/sec: 8553.99 - lr: 0.000015 - momentum: 0.000000
2023-10-18 16:48:52,313 epoch 6 - iter 84/121 - loss 0.46666705 - time (sec): 1.99 - samples/sec: 8596.65 - lr: 0.000014 - momentum: 0.000000
2023-10-18 16:48:52,598 epoch 6 - iter 96/121 - loss 0.46486521 - time (sec): 2.28 - samples/sec: 8641.31 - lr: 0.000014 - momentum: 0.000000
2023-10-18 16:48:52,856 epoch 6 - iter 108/121 - loss 0.46672906 - time (sec): 2.54 - samples/sec: 8654.28 - lr: 0.000014 - momentum: 0.000000
2023-10-18 16:48:53,087 epoch 6 - iter 120/121 - loss 0.46874155 - time (sec): 2.77 - samples/sec: 8879.28 - lr: 0.000013 - momentum: 0.000000
2023-10-18 16:48:53,104 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:53,104 EPOCH 6 done: loss 0.4705 - lr: 0.000013
2023-10-18 16:48:53,519 DEV : loss 0.36645838618278503 - f1-score (micro avg) 0.4151
2023-10-18 16:48:53,524 saving best model
2023-10-18 16:48:53,558 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:53,838 epoch 7 - iter 12/121 - loss 0.50523558 - time (sec): 0.28 - samples/sec: 9255.92 - lr: 0.000013 - momentum: 0.000000
2023-10-18 16:48:54,101 epoch 7 - iter 24/121 - loss 0.52339164 - time (sec): 0.54 - samples/sec: 8657.21 - lr: 0.000013 - momentum: 0.000000
2023-10-18 16:48:54,392 epoch 7 - iter 36/121 - loss 0.48059441 - time (sec): 0.83 - samples/sec: 8492.88 - lr: 0.000012 - momentum: 0.000000
2023-10-18 16:48:54,658 epoch 7 - iter 48/121 - loss 0.46494498 - time (sec): 1.10 - samples/sec: 8560.68 - lr: 0.000012 - momentum: 0.000000
2023-10-18 16:48:54,908 epoch 7 - iter 60/121 - loss 0.45264593 - time (sec): 1.35 - samples/sec: 8831.21 - lr: 0.000012 - momentum: 0.000000
2023-10-18 16:48:55,174 epoch 7 - iter 72/121 - loss 0.43933276 - time (sec): 1.62 - samples/sec: 8974.31 - lr: 0.000011 - momentum: 0.000000
2023-10-18 16:48:55,445 epoch 7 - iter 84/121 - loss 0.43595279 - time (sec): 1.89 - samples/sec: 8984.70 - lr: 0.000011 - momentum: 0.000000
2023-10-18 16:48:55,726 epoch 7 - iter 96/121 - loss 0.43515044 - time (sec): 2.17 - samples/sec: 9174.93 - lr: 0.000011 - momentum: 0.000000
2023-10-18 16:48:56,000 epoch 7 - iter 108/121 - loss 0.43071109 - time (sec): 2.44 - samples/sec: 9085.00 - lr: 0.000010 - momentum: 0.000000
2023-10-18 16:48:56,270 epoch 7 - iter 120/121 - loss 0.43660508 - time (sec): 2.71 - samples/sec: 9052.66 - lr: 0.000010 - momentum: 0.000000
2023-10-18 16:48:56,291 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:56,291 EPOCH 7 done: loss 0.4377 - lr: 0.000010
2023-10-18 16:48:56,717 DEV : loss 0.34206482768058777 - f1-score (micro avg) 0.4824
2023-10-18 16:48:56,722 saving best model
2023-10-18 16:48:56,754 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:57,038 epoch 8 - iter 12/121 - loss 0.39657285 - time (sec): 0.28 - samples/sec: 9431.46 - lr: 0.000010 - momentum: 0.000000
2023-10-18 16:48:57,301 epoch 8 - iter 24/121 - loss 0.41035317 - time (sec): 0.55 - samples/sec: 9594.04 - lr: 0.000009 - momentum: 0.000000
2023-10-18 16:48:57,576 epoch 8 - iter 36/121 - loss 0.43193841 - time (sec): 0.82 - samples/sec: 9142.53 - lr: 0.000009 - momentum: 0.000000
2023-10-18 16:48:57,855 epoch 8 - iter 48/121 - loss 0.42462507 - time (sec): 1.10 - samples/sec: 8958.80 - lr: 0.000009 - momentum: 0.000000
2023-10-18 16:48:58,128 epoch 8 - iter 60/121 - loss 0.42070240 - time (sec): 1.37 - samples/sec: 9029.39 - lr: 0.000008 - momentum: 0.000000
2023-10-18 16:48:58,400 epoch 8 - iter 72/121 - loss 0.43213186 - time (sec): 1.64 - samples/sec: 9184.13 - lr: 0.000008 - momentum: 0.000000
2023-10-18 16:48:58,659 epoch 8 - iter 84/121 - loss 0.43272228 - time (sec): 1.90 - samples/sec: 9123.10 - lr: 0.000008 - momentum: 0.000000
2023-10-18 16:48:58,943 epoch 8 - iter 96/121 - loss 0.42418418 - time (sec): 2.19 - samples/sec: 9023.54 - lr: 0.000008 - momentum: 0.000000
2023-10-18 16:48:59,233 epoch 8 - iter 108/121 - loss 0.42133625 - time (sec): 2.48 - samples/sec: 8978.89 - lr: 0.000007 - momentum: 0.000000
2023-10-18 16:48:59,501 epoch 8 - iter 120/121 - loss 0.42313856 - time (sec): 2.75 - samples/sec: 8956.45 - lr: 0.000007 - momentum: 0.000000
2023-10-18 16:48:59,521 ----------------------------------------------------------------------------------------------------
2023-10-18 16:48:59,521 EPOCH 8 done: loss 0.4254 - lr: 0.000007
2023-10-18 16:48:59,953 DEV : loss 0.33342665433883667 - f1-score (micro avg) 0.4978
2023-10-18 16:48:59,958 saving best model
2023-10-18 16:48:59,989 ----------------------------------------------------------------------------------------------------
2023-10-18 16:49:00,250 epoch 9 - iter 12/121 - loss 0.38063288 - time (sec): 0.26 - samples/sec: 7829.16 - lr: 0.000006 - momentum: 0.000000
2023-10-18 16:49:00,517 epoch 9 - iter 24/121 - loss 0.39518238 - time (sec): 0.53 - samples/sec: 8546.75 - lr: 0.000006 - momentum: 0.000000
2023-10-18 16:49:00,801 epoch 9 - iter 36/121 - loss 0.41125975 - time (sec): 0.81 - samples/sec: 8801.67 - lr: 0.000006 - momentum: 0.000000
2023-10-18 16:49:01,072 epoch 9 - iter 48/121 - loss 0.42758374 - time (sec): 1.08 - samples/sec: 8857.82 - lr: 0.000006 - momentum: 0.000000
2023-10-18 16:49:01,308 epoch 9 - iter 60/121 - loss 0.43010899 - time (sec): 1.32 - samples/sec: 9144.43 - lr: 0.000005 - momentum: 0.000000
2023-10-18 16:49:01,487 epoch 9 - iter 72/121 - loss 0.41322277 - time (sec): 1.50 - samples/sec: 9543.12 - lr: 0.000005 - momentum: 0.000000
2023-10-18 16:49:01,756 epoch 9 - iter 84/121 - loss 0.41753163 - time (sec): 1.77 - samples/sec: 9339.69 - lr: 0.000005 - momentum: 0.000000
2023-10-18 16:49:02,050 epoch 9 - iter 96/121 - loss 0.41759424 - time (sec): 2.06 - samples/sec: 9336.66 - lr: 0.000004 - momentum: 0.000000
2023-10-18 16:49:02,327 epoch 9 - iter 108/121 - loss 0.41214860 - time (sec): 2.34 - samples/sec: 9410.62 - lr: 0.000004 - momentum: 0.000000
2023-10-18 16:49:02,609 epoch 9 - iter 120/121 - loss 0.40768587 - time (sec): 2.62 - samples/sec: 9430.79 - lr: 0.000004 - momentum: 0.000000
2023-10-18 16:49:02,630 ----------------------------------------------------------------------------------------------------
2023-10-18 16:49:02,630 EPOCH 9 done: loss 0.4079 - lr: 0.000004
2023-10-18 16:49:03,057 DEV : loss 0.3280840814113617 - f1-score (micro avg) 0.5015
2023-10-18 16:49:03,061 saving best model
2023-10-18 16:49:03,093 ----------------------------------------------------------------------------------------------------
2023-10-18 16:49:03,354 epoch 10 - iter 12/121 - loss 0.36093503 - time (sec): 0.26 - samples/sec: 8626.32 - lr: 0.000003 - momentum: 0.000000
2023-10-18 16:49:03,619 epoch 10 - iter 24/121 - loss 0.36817826 - time (sec): 0.53 - samples/sec: 8930.16 - lr: 0.000003 - momentum: 0.000000
2023-10-18 16:49:03,896 epoch 10 - iter 36/121 - loss 0.38006792 - time (sec): 0.80 - samples/sec: 8872.17 - lr: 0.000003 - momentum: 0.000000
2023-10-18 16:49:04,162 epoch 10 - iter 48/121 - loss 0.37238255 - time (sec): 1.07 - samples/sec: 9074.65 - lr: 0.000002 - momentum: 0.000000
2023-10-18 16:49:04,428 epoch 10 - iter 60/121 - loss 0.39144881 - time (sec): 1.33 - samples/sec: 9013.20 - lr: 0.000002 - momentum: 0.000000
2023-10-18 16:49:04,698 epoch 10 - iter 72/121 - loss 0.40218852 - time (sec): 1.60 - samples/sec: 9206.61 - lr: 0.000002 - momentum: 0.000000
2023-10-18 16:49:04,990 epoch 10 - iter 84/121 - loss 0.38920723 - time (sec): 1.90 - samples/sec: 9100.69 - lr: 0.000001 - momentum: 0.000000
2023-10-18 16:49:05,255 epoch 10 - iter 96/121 - loss 0.39328880 - time (sec): 2.16 - samples/sec: 9091.63 - lr: 0.000001 - momentum: 0.000000
2023-10-18 16:49:05,531 epoch 10 - iter 108/121 - loss 0.39338392 - time (sec): 2.44 - samples/sec: 9085.86 - lr: 0.000001 - momentum: 0.000000
2023-10-18 16:49:05,795 epoch 10 - iter 120/121 - loss 0.40198011 - time (sec): 2.70 - samples/sec: 9092.62 - lr: 0.000000 - momentum: 0.000000
2023-10-18 16:49:05,814 ----------------------------------------------------------------------------------------------------
2023-10-18 16:49:05,814 EPOCH 10 done: loss 0.4003 - lr: 0.000000
2023-10-18 16:49:06,241 DEV : loss 0.32836541533470154 - f1-score (micro avg) 0.4978
2023-10-18 16:49:06,276 ----------------------------------------------------------------------------------------------------
2023-10-18 16:49:06,276 Loading model from best epoch ...
2023-10-18 16:49:06,355 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
2023-10-18 16:49:06,782
Results:
- F-score (micro) 0.4535
- F-score (macro) 0.2114
- Accuracy 0.3051
By class:
precision recall f1-score support
pers 0.6099 0.6187 0.6143 139
scope 0.4361 0.4496 0.4427 129
work 0.0000 0.0000 0.0000 80
loc 0.0000 0.0000 0.0000 9
date 0.0000 0.0000 0.0000 3
micro avg 0.5236 0.4000 0.4535 360
macro avg 0.2092 0.2137 0.2114 360
weighted avg 0.3918 0.4000 0.3958 360
2023-10-18 16:49:06,782 ----------------------------------------------------------------------------------------------------
|