File size: 23,935 Bytes
6d98ed7 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 |
2023-10-13 11:10:18,371 ----------------------------------------------------------------------------------------------------
2023-10-13 11:10:18,372 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 11:10:18,372 ----------------------------------------------------------------------------------------------------
2023-10-13 11:10:18,372 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
2023-10-13 11:10:18,372 ----------------------------------------------------------------------------------------------------
2023-10-13 11:10:18,372 Train: 966 sentences
2023-10-13 11:10:18,372 (train_with_dev=False, train_with_test=False)
2023-10-13 11:10:18,372 ----------------------------------------------------------------------------------------------------
2023-10-13 11:10:18,372 Training Params:
2023-10-13 11:10:18,372 - learning_rate: "3e-05"
2023-10-13 11:10:18,372 - mini_batch_size: "8"
2023-10-13 11:10:18,372 - max_epochs: "10"
2023-10-13 11:10:18,372 - shuffle: "True"
2023-10-13 11:10:18,372 ----------------------------------------------------------------------------------------------------
2023-10-13 11:10:18,373 Plugins:
2023-10-13 11:10:18,373 - LinearScheduler | warmup_fraction: '0.1'
2023-10-13 11:10:18,373 ----------------------------------------------------------------------------------------------------
2023-10-13 11:10:18,373 Final evaluation on model from best epoch (best-model.pt)
2023-10-13 11:10:18,373 - metric: "('micro avg', 'f1-score')"
2023-10-13 11:10:18,373 ----------------------------------------------------------------------------------------------------
2023-10-13 11:10:18,373 Computation:
2023-10-13 11:10:18,373 - compute on device: cuda:0
2023-10-13 11:10:18,373 - embedding storage: none
2023-10-13 11:10:18,373 ----------------------------------------------------------------------------------------------------
2023-10-13 11:10:18,373 Model training base path: "hmbench-ajmc/fr-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5"
2023-10-13 11:10:18,373 ----------------------------------------------------------------------------------------------------
2023-10-13 11:10:18,373 ----------------------------------------------------------------------------------------------------
2023-10-13 11:10:19,028 epoch 1 - iter 12/121 - loss 3.23686547 - time (sec): 0.65 - samples/sec: 3357.97 - lr: 0.000003 - momentum: 0.000000
2023-10-13 11:10:19,789 epoch 1 - iter 24/121 - loss 3.08356686 - time (sec): 1.42 - samples/sec: 3404.65 - lr: 0.000006 - momentum: 0.000000
2023-10-13 11:10:20,522 epoch 1 - iter 36/121 - loss 2.75565966 - time (sec): 2.15 - samples/sec: 3356.65 - lr: 0.000009 - momentum: 0.000000
2023-10-13 11:10:21,260 epoch 1 - iter 48/121 - loss 2.22033244 - time (sec): 2.89 - samples/sec: 3445.03 - lr: 0.000012 - momentum: 0.000000
2023-10-13 11:10:22,022 epoch 1 - iter 60/121 - loss 1.90687504 - time (sec): 3.65 - samples/sec: 3413.67 - lr: 0.000015 - momentum: 0.000000
2023-10-13 11:10:22,689 epoch 1 - iter 72/121 - loss 1.70068116 - time (sec): 4.31 - samples/sec: 3377.30 - lr: 0.000018 - momentum: 0.000000
2023-10-13 11:10:23,464 epoch 1 - iter 84/121 - loss 1.54601736 - time (sec): 5.09 - samples/sec: 3371.31 - lr: 0.000021 - momentum: 0.000000
2023-10-13 11:10:24,156 epoch 1 - iter 96/121 - loss 1.42977512 - time (sec): 5.78 - samples/sec: 3382.42 - lr: 0.000024 - momentum: 0.000000
2023-10-13 11:10:24,867 epoch 1 - iter 108/121 - loss 1.33870901 - time (sec): 6.49 - samples/sec: 3365.58 - lr: 0.000027 - momentum: 0.000000
2023-10-13 11:10:25,655 epoch 1 - iter 120/121 - loss 1.23639310 - time (sec): 7.28 - samples/sec: 3361.78 - lr: 0.000030 - momentum: 0.000000
2023-10-13 11:10:25,726 ----------------------------------------------------------------------------------------------------
2023-10-13 11:10:25,726 EPOCH 1 done: loss 1.2258 - lr: 0.000030
2023-10-13 11:10:26,644 DEV : loss 0.36207425594329834 - f1-score (micro avg) 0.3053
2023-10-13 11:10:26,649 saving best model
2023-10-13 11:10:27,006 ----------------------------------------------------------------------------------------------------
2023-10-13 11:10:27,700 epoch 2 - iter 12/121 - loss 0.39607246 - time (sec): 0.69 - samples/sec: 3726.90 - lr: 0.000030 - momentum: 0.000000
2023-10-13 11:10:28,463 epoch 2 - iter 24/121 - loss 0.36432721 - time (sec): 1.46 - samples/sec: 3571.66 - lr: 0.000029 - momentum: 0.000000
2023-10-13 11:10:29,135 epoch 2 - iter 36/121 - loss 0.36955311 - time (sec): 2.13 - samples/sec: 3459.37 - lr: 0.000029 - momentum: 0.000000
2023-10-13 11:10:29,904 epoch 2 - iter 48/121 - loss 0.34687311 - time (sec): 2.90 - samples/sec: 3426.17 - lr: 0.000029 - momentum: 0.000000
2023-10-13 11:10:30,617 epoch 2 - iter 60/121 - loss 0.32491719 - time (sec): 3.61 - samples/sec: 3393.44 - lr: 0.000028 - momentum: 0.000000
2023-10-13 11:10:31,369 epoch 2 - iter 72/121 - loss 0.30750715 - time (sec): 4.36 - samples/sec: 3381.87 - lr: 0.000028 - momentum: 0.000000
2023-10-13 11:10:32,148 epoch 2 - iter 84/121 - loss 0.29997426 - time (sec): 5.14 - samples/sec: 3362.05 - lr: 0.000028 - momentum: 0.000000
2023-10-13 11:10:32,881 epoch 2 - iter 96/121 - loss 0.29230878 - time (sec): 5.87 - samples/sec: 3367.92 - lr: 0.000027 - momentum: 0.000000
2023-10-13 11:10:33,606 epoch 2 - iter 108/121 - loss 0.27725347 - time (sec): 6.60 - samples/sec: 3375.32 - lr: 0.000027 - momentum: 0.000000
2023-10-13 11:10:34,301 epoch 2 - iter 120/121 - loss 0.26927054 - time (sec): 7.29 - samples/sec: 3378.13 - lr: 0.000027 - momentum: 0.000000
2023-10-13 11:10:34,350 ----------------------------------------------------------------------------------------------------
2023-10-13 11:10:34,350 EPOCH 2 done: loss 0.2695 - lr: 0.000027
2023-10-13 11:10:35,153 DEV : loss 0.16814236342906952 - f1-score (micro avg) 0.6962
2023-10-13 11:10:35,159 saving best model
2023-10-13 11:10:35,771 ----------------------------------------------------------------------------------------------------
2023-10-13 11:10:36,503 epoch 3 - iter 12/121 - loss 0.15819860 - time (sec): 0.73 - samples/sec: 3394.19 - lr: 0.000026 - momentum: 0.000000
2023-10-13 11:10:37,304 epoch 3 - iter 24/121 - loss 0.16309934 - time (sec): 1.53 - samples/sec: 3229.87 - lr: 0.000026 - momentum: 0.000000
2023-10-13 11:10:38,024 epoch 3 - iter 36/121 - loss 0.15355893 - time (sec): 2.25 - samples/sec: 3108.45 - lr: 0.000026 - momentum: 0.000000
2023-10-13 11:10:38,805 epoch 3 - iter 48/121 - loss 0.14476800 - time (sec): 3.03 - samples/sec: 3151.69 - lr: 0.000025 - momentum: 0.000000
2023-10-13 11:10:39,562 epoch 3 - iter 60/121 - loss 0.15856289 - time (sec): 3.79 - samples/sec: 3198.34 - lr: 0.000025 - momentum: 0.000000
2023-10-13 11:10:40,321 epoch 3 - iter 72/121 - loss 0.15967065 - time (sec): 4.55 - samples/sec: 3238.68 - lr: 0.000025 - momentum: 0.000000
2023-10-13 11:10:41,049 epoch 3 - iter 84/121 - loss 0.15280106 - time (sec): 5.27 - samples/sec: 3290.71 - lr: 0.000024 - momentum: 0.000000
2023-10-13 11:10:41,787 epoch 3 - iter 96/121 - loss 0.15638849 - time (sec): 6.01 - samples/sec: 3282.68 - lr: 0.000024 - momentum: 0.000000
2023-10-13 11:10:42,441 epoch 3 - iter 108/121 - loss 0.15090289 - time (sec): 6.66 - samples/sec: 3289.20 - lr: 0.000024 - momentum: 0.000000
2023-10-13 11:10:43,180 epoch 3 - iter 120/121 - loss 0.14357302 - time (sec): 7.40 - samples/sec: 3309.23 - lr: 0.000023 - momentum: 0.000000
2023-10-13 11:10:43,232 ----------------------------------------------------------------------------------------------------
2023-10-13 11:10:43,232 EPOCH 3 done: loss 0.1423 - lr: 0.000023
2023-10-13 11:10:44,149 DEV : loss 0.126093789935112 - f1-score (micro avg) 0.8055
2023-10-13 11:10:44,157 saving best model
2023-10-13 11:10:44,687 ----------------------------------------------------------------------------------------------------
2023-10-13 11:10:45,522 epoch 4 - iter 12/121 - loss 0.08458435 - time (sec): 0.83 - samples/sec: 2905.79 - lr: 0.000023 - momentum: 0.000000
2023-10-13 11:10:46,450 epoch 4 - iter 24/121 - loss 0.09228534 - time (sec): 1.76 - samples/sec: 2859.11 - lr: 0.000023 - momentum: 0.000000
2023-10-13 11:10:47,240 epoch 4 - iter 36/121 - loss 0.09097094 - time (sec): 2.55 - samples/sec: 2886.20 - lr: 0.000022 - momentum: 0.000000
2023-10-13 11:10:48,030 epoch 4 - iter 48/121 - loss 0.09573211 - time (sec): 3.34 - samples/sec: 2936.27 - lr: 0.000022 - momentum: 0.000000
2023-10-13 11:10:48,873 epoch 4 - iter 60/121 - loss 0.09409124 - time (sec): 4.18 - samples/sec: 2961.19 - lr: 0.000022 - momentum: 0.000000
2023-10-13 11:10:49,726 epoch 4 - iter 72/121 - loss 0.09645539 - time (sec): 5.04 - samples/sec: 2968.80 - lr: 0.000021 - momentum: 0.000000
2023-10-13 11:10:50,507 epoch 4 - iter 84/121 - loss 0.09356465 - time (sec): 5.82 - samples/sec: 2970.75 - lr: 0.000021 - momentum: 0.000000
2023-10-13 11:10:51,306 epoch 4 - iter 96/121 - loss 0.09865829 - time (sec): 6.62 - samples/sec: 2949.74 - lr: 0.000021 - momentum: 0.000000
2023-10-13 11:10:52,131 epoch 4 - iter 108/121 - loss 0.10045796 - time (sec): 7.44 - samples/sec: 2944.22 - lr: 0.000020 - momentum: 0.000000
2023-10-13 11:10:53,065 epoch 4 - iter 120/121 - loss 0.09623331 - time (sec): 8.38 - samples/sec: 2938.18 - lr: 0.000020 - momentum: 0.000000
2023-10-13 11:10:53,130 ----------------------------------------------------------------------------------------------------
2023-10-13 11:10:53,130 EPOCH 4 done: loss 0.0959 - lr: 0.000020
2023-10-13 11:10:53,958 DEV : loss 0.11419466882944107 - f1-score (micro avg) 0.814
2023-10-13 11:10:53,963 saving best model
2023-10-13 11:10:54,443 ----------------------------------------------------------------------------------------------------
2023-10-13 11:10:55,252 epoch 5 - iter 12/121 - loss 0.07157336 - time (sec): 0.80 - samples/sec: 3282.21 - lr: 0.000020 - momentum: 0.000000
2023-10-13 11:10:55,972 epoch 5 - iter 24/121 - loss 0.07611632 - time (sec): 1.52 - samples/sec: 3279.00 - lr: 0.000019 - momentum: 0.000000
2023-10-13 11:10:56,678 epoch 5 - iter 36/121 - loss 0.06427848 - time (sec): 2.23 - samples/sec: 3257.32 - lr: 0.000019 - momentum: 0.000000
2023-10-13 11:10:57,412 epoch 5 - iter 48/121 - loss 0.07273777 - time (sec): 2.96 - samples/sec: 3296.01 - lr: 0.000019 - momentum: 0.000000
2023-10-13 11:10:58,115 epoch 5 - iter 60/121 - loss 0.07213143 - time (sec): 3.67 - samples/sec: 3344.49 - lr: 0.000018 - momentum: 0.000000
2023-10-13 11:10:58,831 epoch 5 - iter 72/121 - loss 0.06897480 - time (sec): 4.38 - samples/sec: 3370.17 - lr: 0.000018 - momentum: 0.000000
2023-10-13 11:10:59,579 epoch 5 - iter 84/121 - loss 0.06830377 - time (sec): 5.13 - samples/sec: 3398.16 - lr: 0.000018 - momentum: 0.000000
2023-10-13 11:11:00,354 epoch 5 - iter 96/121 - loss 0.06639599 - time (sec): 5.91 - samples/sec: 3361.63 - lr: 0.000017 - momentum: 0.000000
2023-10-13 11:11:01,097 epoch 5 - iter 108/121 - loss 0.06645515 - time (sec): 6.65 - samples/sec: 3375.34 - lr: 0.000017 - momentum: 0.000000
2023-10-13 11:11:01,778 epoch 5 - iter 120/121 - loss 0.06438298 - time (sec): 7.33 - samples/sec: 3356.46 - lr: 0.000017 - momentum: 0.000000
2023-10-13 11:11:01,827 ----------------------------------------------------------------------------------------------------
2023-10-13 11:11:01,827 EPOCH 5 done: loss 0.0645 - lr: 0.000017
2023-10-13 11:11:02,675 DEV : loss 0.14454086124897003 - f1-score (micro avg) 0.8054
2023-10-13 11:11:02,681 ----------------------------------------------------------------------------------------------------
2023-10-13 11:11:03,442 epoch 6 - iter 12/121 - loss 0.05954669 - time (sec): 0.76 - samples/sec: 3410.58 - lr: 0.000016 - momentum: 0.000000
2023-10-13 11:11:04,204 epoch 6 - iter 24/121 - loss 0.05911514 - time (sec): 1.52 - samples/sec: 3345.39 - lr: 0.000016 - momentum: 0.000000
2023-10-13 11:11:04,960 epoch 6 - iter 36/121 - loss 0.05197972 - time (sec): 2.28 - samples/sec: 3380.01 - lr: 0.000016 - momentum: 0.000000
2023-10-13 11:11:05,619 epoch 6 - iter 48/121 - loss 0.04652385 - time (sec): 2.94 - samples/sec: 3340.33 - lr: 0.000015 - momentum: 0.000000
2023-10-13 11:11:06,487 epoch 6 - iter 60/121 - loss 0.04783989 - time (sec): 3.80 - samples/sec: 3284.27 - lr: 0.000015 - momentum: 0.000000
2023-10-13 11:11:07,249 epoch 6 - iter 72/121 - loss 0.04450554 - time (sec): 4.57 - samples/sec: 3267.21 - lr: 0.000015 - momentum: 0.000000
2023-10-13 11:11:07,951 epoch 6 - iter 84/121 - loss 0.04614645 - time (sec): 5.27 - samples/sec: 3238.91 - lr: 0.000014 - momentum: 0.000000
2023-10-13 11:11:08,678 epoch 6 - iter 96/121 - loss 0.04321518 - time (sec): 6.00 - samples/sec: 3235.98 - lr: 0.000014 - momentum: 0.000000
2023-10-13 11:11:09,424 epoch 6 - iter 108/121 - loss 0.04354754 - time (sec): 6.74 - samples/sec: 3240.25 - lr: 0.000014 - momentum: 0.000000
2023-10-13 11:11:10,213 epoch 6 - iter 120/121 - loss 0.04690223 - time (sec): 7.53 - samples/sec: 3267.09 - lr: 0.000013 - momentum: 0.000000
2023-10-13 11:11:10,263 ----------------------------------------------------------------------------------------------------
2023-10-13 11:11:10,263 EPOCH 6 done: loss 0.0480 - lr: 0.000013
2023-10-13 11:11:11,090 DEV : loss 0.15227191150188446 - f1-score (micro avg) 0.818
2023-10-13 11:11:11,095 saving best model
2023-10-13 11:11:11,576 ----------------------------------------------------------------------------------------------------
2023-10-13 11:11:12,332 epoch 7 - iter 12/121 - loss 0.03956377 - time (sec): 0.75 - samples/sec: 3394.79 - lr: 0.000013 - momentum: 0.000000
2023-10-13 11:11:13,065 epoch 7 - iter 24/121 - loss 0.03515716 - time (sec): 1.48 - samples/sec: 3401.61 - lr: 0.000013 - momentum: 0.000000
2023-10-13 11:11:13,747 epoch 7 - iter 36/121 - loss 0.03182407 - time (sec): 2.17 - samples/sec: 3397.80 - lr: 0.000012 - momentum: 0.000000
2023-10-13 11:11:14,454 epoch 7 - iter 48/121 - loss 0.03351245 - time (sec): 2.87 - samples/sec: 3392.40 - lr: 0.000012 - momentum: 0.000000
2023-10-13 11:11:15,166 epoch 7 - iter 60/121 - loss 0.03833087 - time (sec): 3.59 - samples/sec: 3410.52 - lr: 0.000012 - momentum: 0.000000
2023-10-13 11:11:15,953 epoch 7 - iter 72/121 - loss 0.03730157 - time (sec): 4.37 - samples/sec: 3389.92 - lr: 0.000011 - momentum: 0.000000
2023-10-13 11:11:16,675 epoch 7 - iter 84/121 - loss 0.03774702 - time (sec): 5.09 - samples/sec: 3400.29 - lr: 0.000011 - momentum: 0.000000
2023-10-13 11:11:17,403 epoch 7 - iter 96/121 - loss 0.03950572 - time (sec): 5.82 - samples/sec: 3361.81 - lr: 0.000011 - momentum: 0.000000
2023-10-13 11:11:18,156 epoch 7 - iter 108/121 - loss 0.03913401 - time (sec): 6.58 - samples/sec: 3360.69 - lr: 0.000010 - momentum: 0.000000
2023-10-13 11:11:18,941 epoch 7 - iter 120/121 - loss 0.03782988 - time (sec): 7.36 - samples/sec: 3333.14 - lr: 0.000010 - momentum: 0.000000
2023-10-13 11:11:18,999 ----------------------------------------------------------------------------------------------------
2023-10-13 11:11:18,999 EPOCH 7 done: loss 0.0375 - lr: 0.000010
2023-10-13 11:11:19,763 DEV : loss 0.1678510457277298 - f1-score (micro avg) 0.8238
2023-10-13 11:11:19,768 saving best model
2023-10-13 11:11:20,230 ----------------------------------------------------------------------------------------------------
2023-10-13 11:11:21,029 epoch 8 - iter 12/121 - loss 0.02579330 - time (sec): 0.79 - samples/sec: 3386.12 - lr: 0.000010 - momentum: 0.000000
2023-10-13 11:11:21,760 epoch 8 - iter 24/121 - loss 0.02880649 - time (sec): 1.53 - samples/sec: 3406.78 - lr: 0.000009 - momentum: 0.000000
2023-10-13 11:11:22,597 epoch 8 - iter 36/121 - loss 0.02842207 - time (sec): 2.36 - samples/sec: 3260.17 - lr: 0.000009 - momentum: 0.000000
2023-10-13 11:11:23,347 epoch 8 - iter 48/121 - loss 0.02926475 - time (sec): 3.11 - samples/sec: 3281.81 - lr: 0.000009 - momentum: 0.000000
2023-10-13 11:11:24,083 epoch 8 - iter 60/121 - loss 0.02641771 - time (sec): 3.85 - samples/sec: 3333.45 - lr: 0.000008 - momentum: 0.000000
2023-10-13 11:11:24,829 epoch 8 - iter 72/121 - loss 0.02717774 - time (sec): 4.59 - samples/sec: 3308.41 - lr: 0.000008 - momentum: 0.000000
2023-10-13 11:11:25,566 epoch 8 - iter 84/121 - loss 0.02823503 - time (sec): 5.33 - samples/sec: 3278.29 - lr: 0.000008 - momentum: 0.000000
2023-10-13 11:11:26,274 epoch 8 - iter 96/121 - loss 0.02818844 - time (sec): 6.04 - samples/sec: 3280.47 - lr: 0.000008 - momentum: 0.000000
2023-10-13 11:11:26,975 epoch 8 - iter 108/121 - loss 0.02864591 - time (sec): 6.74 - samples/sec: 3317.23 - lr: 0.000007 - momentum: 0.000000
2023-10-13 11:11:27,685 epoch 8 - iter 120/121 - loss 0.02846050 - time (sec): 7.45 - samples/sec: 3309.76 - lr: 0.000007 - momentum: 0.000000
2023-10-13 11:11:27,732 ----------------------------------------------------------------------------------------------------
2023-10-13 11:11:27,733 EPOCH 8 done: loss 0.0284 - lr: 0.000007
2023-10-13 11:11:28,644 DEV : loss 0.16501988470554352 - f1-score (micro avg) 0.8385
2023-10-13 11:11:28,649 saving best model
2023-10-13 11:11:29,116 ----------------------------------------------------------------------------------------------------
2023-10-13 11:11:29,884 epoch 9 - iter 12/121 - loss 0.02637404 - time (sec): 0.77 - samples/sec: 3175.23 - lr: 0.000006 - momentum: 0.000000
2023-10-13 11:11:30,600 epoch 9 - iter 24/121 - loss 0.03206035 - time (sec): 1.48 - samples/sec: 3432.23 - lr: 0.000006 - momentum: 0.000000
2023-10-13 11:11:31,365 epoch 9 - iter 36/121 - loss 0.03177588 - time (sec): 2.25 - samples/sec: 3456.53 - lr: 0.000006 - momentum: 0.000000
2023-10-13 11:11:32,041 epoch 9 - iter 48/121 - loss 0.02688614 - time (sec): 2.92 - samples/sec: 3368.96 - lr: 0.000006 - momentum: 0.000000
2023-10-13 11:11:32,715 epoch 9 - iter 60/121 - loss 0.02569866 - time (sec): 3.60 - samples/sec: 3340.67 - lr: 0.000005 - momentum: 0.000000
2023-10-13 11:11:33,423 epoch 9 - iter 72/121 - loss 0.02485426 - time (sec): 4.31 - samples/sec: 3368.02 - lr: 0.000005 - momentum: 0.000000
2023-10-13 11:11:34,161 epoch 9 - iter 84/121 - loss 0.02371548 - time (sec): 5.04 - samples/sec: 3413.07 - lr: 0.000005 - momentum: 0.000000
2023-10-13 11:11:34,953 epoch 9 - iter 96/121 - loss 0.02197400 - time (sec): 5.84 - samples/sec: 3361.81 - lr: 0.000004 - momentum: 0.000000
2023-10-13 11:11:35,638 epoch 9 - iter 108/121 - loss 0.02375103 - time (sec): 6.52 - samples/sec: 3341.96 - lr: 0.000004 - momentum: 0.000000
2023-10-13 11:11:36,449 epoch 9 - iter 120/121 - loss 0.02309117 - time (sec): 7.33 - samples/sec: 3355.78 - lr: 0.000004 - momentum: 0.000000
2023-10-13 11:11:36,497 ----------------------------------------------------------------------------------------------------
2023-10-13 11:11:36,497 EPOCH 9 done: loss 0.0232 - lr: 0.000004
2023-10-13 11:11:37,265 DEV : loss 0.17005358636379242 - f1-score (micro avg) 0.8306
2023-10-13 11:11:37,269 ----------------------------------------------------------------------------------------------------
2023-10-13 11:11:37,953 epoch 10 - iter 12/121 - loss 0.01774069 - time (sec): 0.68 - samples/sec: 3559.64 - lr: 0.000003 - momentum: 0.000000
2023-10-13 11:11:38,720 epoch 10 - iter 24/121 - loss 0.02065838 - time (sec): 1.45 - samples/sec: 3574.69 - lr: 0.000003 - momentum: 0.000000
2023-10-13 11:11:39,427 epoch 10 - iter 36/121 - loss 0.02086100 - time (sec): 2.16 - samples/sec: 3450.79 - lr: 0.000003 - momentum: 0.000000
2023-10-13 11:11:40,124 epoch 10 - iter 48/121 - loss 0.02031963 - time (sec): 2.85 - samples/sec: 3362.98 - lr: 0.000002 - momentum: 0.000000
2023-10-13 11:11:40,892 epoch 10 - iter 60/121 - loss 0.02096082 - time (sec): 3.62 - samples/sec: 3451.73 - lr: 0.000002 - momentum: 0.000000
2023-10-13 11:11:41,615 epoch 10 - iter 72/121 - loss 0.02132037 - time (sec): 4.34 - samples/sec: 3421.31 - lr: 0.000002 - momentum: 0.000000
2023-10-13 11:11:42,362 epoch 10 - iter 84/121 - loss 0.02130246 - time (sec): 5.09 - samples/sec: 3374.54 - lr: 0.000001 - momentum: 0.000000
2023-10-13 11:11:43,063 epoch 10 - iter 96/121 - loss 0.02273356 - time (sec): 5.79 - samples/sec: 3353.51 - lr: 0.000001 - momentum: 0.000000
2023-10-13 11:11:43,822 epoch 10 - iter 108/121 - loss 0.02075952 - time (sec): 6.55 - samples/sec: 3344.69 - lr: 0.000001 - momentum: 0.000000
2023-10-13 11:11:44,676 epoch 10 - iter 120/121 - loss 0.02007672 - time (sec): 7.41 - samples/sec: 3316.90 - lr: 0.000000 - momentum: 0.000000
2023-10-13 11:11:44,728 ----------------------------------------------------------------------------------------------------
2023-10-13 11:11:44,728 EPOCH 10 done: loss 0.0199 - lr: 0.000000
2023-10-13 11:11:45,491 DEV : loss 0.1679229587316513 - f1-score (micro avg) 0.8362
2023-10-13 11:11:45,859 ----------------------------------------------------------------------------------------------------
2023-10-13 11:11:45,860 Loading model from best epoch ...
2023-10-13 11:11:47,203 SequenceTagger predicts: Dictionary with 25 tags: O, S-scope, B-scope, E-scope, I-scope, S-pers, B-pers, E-pers, I-pers, S-work, B-work, E-work, I-work, S-loc, B-loc, E-loc, I-loc, S-object, B-object, E-object, I-object, S-date, B-date, E-date, I-date
2023-10-13 11:11:47,847
Results:
- F-score (micro) 0.8267
- F-score (macro) 0.5381
- Accuracy 0.7284
By class:
precision recall f1-score support
pers 0.8696 0.8633 0.8664 139
scope 0.8273 0.8915 0.8582 129
work 0.7053 0.8375 0.7657 80
loc 1.0000 0.1111 0.2000 9
date 0.0000 0.0000 0.0000 3
micro avg 0.8123 0.8417 0.8267 360
macro avg 0.6804 0.5407 0.5381 360
weighted avg 0.8139 0.8417 0.8172 360
2023-10-13 11:11:47,848 ----------------------------------------------------------------------------------------------------
|