File size: 24,394 Bytes
41eb92f |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 |
2023-10-19 19:51:06,768 ----------------------------------------------------------------------------------------------------
2023-10-19 19:51:06,768 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=17, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-19 19:51:06,768 ----------------------------------------------------------------------------------------------------
2023-10-19 19:51:06,768 MultiCorpus: 7142 train + 698 dev + 2570 test sentences
- NER_HIPE_2022 Corpus: 7142 train + 698 dev + 2570 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fr/with_doc_seperator
2023-10-19 19:51:06,768 ----------------------------------------------------------------------------------------------------
2023-10-19 19:51:06,768 Train: 7142 sentences
2023-10-19 19:51:06,768 (train_with_dev=False, train_with_test=False)
2023-10-19 19:51:06,768 ----------------------------------------------------------------------------------------------------
2023-10-19 19:51:06,768 Training Params:
2023-10-19 19:51:06,768 - learning_rate: "3e-05"
2023-10-19 19:51:06,768 - mini_batch_size: "4"
2023-10-19 19:51:06,769 - max_epochs: "10"
2023-10-19 19:51:06,769 - shuffle: "True"
2023-10-19 19:51:06,769 ----------------------------------------------------------------------------------------------------
2023-10-19 19:51:06,769 Plugins:
2023-10-19 19:51:06,769 - TensorboardLogger
2023-10-19 19:51:06,769 - LinearScheduler | warmup_fraction: '0.1'
2023-10-19 19:51:06,769 ----------------------------------------------------------------------------------------------------
2023-10-19 19:51:06,769 Final evaluation on model from best epoch (best-model.pt)
2023-10-19 19:51:06,769 - metric: "('micro avg', 'f1-score')"
2023-10-19 19:51:06,769 ----------------------------------------------------------------------------------------------------
2023-10-19 19:51:06,769 Computation:
2023-10-19 19:51:06,769 - compute on device: cuda:0
2023-10-19 19:51:06,769 - embedding storage: none
2023-10-19 19:51:06,769 ----------------------------------------------------------------------------------------------------
2023-10-19 19:51:06,769 Model training base path: "hmbench-newseye/fr-dbmdz/bert-tiny-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-19 19:51:06,769 ----------------------------------------------------------------------------------------------------
2023-10-19 19:51:06,769 ----------------------------------------------------------------------------------------------------
2023-10-19 19:51:06,769 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-19 19:51:09,948 epoch 1 - iter 178/1786 - loss 2.79687367 - time (sec): 3.18 - samples/sec: 8405.70 - lr: 0.000003 - momentum: 0.000000
2023-10-19 19:51:13,196 epoch 1 - iter 356/1786 - loss 2.50644631 - time (sec): 6.43 - samples/sec: 7917.10 - lr: 0.000006 - momentum: 0.000000
2023-10-19 19:51:16,242 epoch 1 - iter 534/1786 - loss 2.11669177 - time (sec): 9.47 - samples/sec: 7847.71 - lr: 0.000009 - momentum: 0.000000
2023-10-19 19:51:19,447 epoch 1 - iter 712/1786 - loss 1.78468271 - time (sec): 12.68 - samples/sec: 7832.18 - lr: 0.000012 - momentum: 0.000000
2023-10-19 19:51:22,749 epoch 1 - iter 890/1786 - loss 1.58099334 - time (sec): 15.98 - samples/sec: 7696.48 - lr: 0.000015 - momentum: 0.000000
2023-10-19 19:51:25,906 epoch 1 - iter 1068/1786 - loss 1.44626933 - time (sec): 19.14 - samples/sec: 7688.35 - lr: 0.000018 - momentum: 0.000000
2023-10-19 19:51:29,188 epoch 1 - iter 1246/1786 - loss 1.32135288 - time (sec): 22.42 - samples/sec: 7711.92 - lr: 0.000021 - momentum: 0.000000
2023-10-19 19:51:32,297 epoch 1 - iter 1424/1786 - loss 1.22571550 - time (sec): 25.53 - samples/sec: 7701.62 - lr: 0.000024 - momentum: 0.000000
2023-10-19 19:51:35,449 epoch 1 - iter 1602/1786 - loss 1.14718597 - time (sec): 28.68 - samples/sec: 7738.29 - lr: 0.000027 - momentum: 0.000000
2023-10-19 19:51:38,565 epoch 1 - iter 1780/1786 - loss 1.08157944 - time (sec): 31.80 - samples/sec: 7802.23 - lr: 0.000030 - momentum: 0.000000
2023-10-19 19:51:38,668 ----------------------------------------------------------------------------------------------------
2023-10-19 19:51:38,668 EPOCH 1 done: loss 1.0801 - lr: 0.000030
2023-10-19 19:51:40,134 DEV : loss 0.3227035701274872 - f1-score (micro avg) 0.1395
2023-10-19 19:51:40,149 saving best model
2023-10-19 19:51:40,183 ----------------------------------------------------------------------------------------------------
2023-10-19 19:51:43,289 epoch 2 - iter 178/1786 - loss 0.50600833 - time (sec): 3.11 - samples/sec: 7616.92 - lr: 0.000030 - momentum: 0.000000
2023-10-19 19:51:46,333 epoch 2 - iter 356/1786 - loss 0.46557022 - time (sec): 6.15 - samples/sec: 7898.55 - lr: 0.000029 - momentum: 0.000000
2023-10-19 19:51:49,371 epoch 2 - iter 534/1786 - loss 0.46513364 - time (sec): 9.19 - samples/sec: 7766.67 - lr: 0.000029 - momentum: 0.000000
2023-10-19 19:51:52,594 epoch 2 - iter 712/1786 - loss 0.45220596 - time (sec): 12.41 - samples/sec: 7737.80 - lr: 0.000029 - momentum: 0.000000
2023-10-19 19:51:56,019 epoch 2 - iter 890/1786 - loss 0.45404700 - time (sec): 15.84 - samples/sec: 7667.36 - lr: 0.000028 - momentum: 0.000000
2023-10-19 19:51:59,181 epoch 2 - iter 1068/1786 - loss 0.44855842 - time (sec): 19.00 - samples/sec: 7742.35 - lr: 0.000028 - momentum: 0.000000
2023-10-19 19:52:02,417 epoch 2 - iter 1246/1786 - loss 0.44009549 - time (sec): 22.23 - samples/sec: 7836.30 - lr: 0.000028 - momentum: 0.000000
2023-10-19 19:52:05,463 epoch 2 - iter 1424/1786 - loss 0.43715972 - time (sec): 25.28 - samples/sec: 7889.50 - lr: 0.000027 - momentum: 0.000000
2023-10-19 19:52:08,578 epoch 2 - iter 1602/1786 - loss 0.43518582 - time (sec): 28.39 - samples/sec: 7891.60 - lr: 0.000027 - momentum: 0.000000
2023-10-19 19:52:11,635 epoch 2 - iter 1780/1786 - loss 0.43341001 - time (sec): 31.45 - samples/sec: 7891.97 - lr: 0.000027 - momentum: 0.000000
2023-10-19 19:52:11,727 ----------------------------------------------------------------------------------------------------
2023-10-19 19:52:11,727 EPOCH 2 done: loss 0.4333 - lr: 0.000027
2023-10-19 19:52:14,547 DEV : loss 0.2583433985710144 - f1-score (micro avg) 0.3512
2023-10-19 19:52:14,562 saving best model
2023-10-19 19:52:14,594 ----------------------------------------------------------------------------------------------------
2023-10-19 19:52:17,843 epoch 3 - iter 178/1786 - loss 0.36689807 - time (sec): 3.25 - samples/sec: 7327.59 - lr: 0.000026 - momentum: 0.000000
2023-10-19 19:52:21,112 epoch 3 - iter 356/1786 - loss 0.34773508 - time (sec): 6.52 - samples/sec: 7774.91 - lr: 0.000026 - momentum: 0.000000
2023-10-19 19:52:24,217 epoch 3 - iter 534/1786 - loss 0.34418896 - time (sec): 9.62 - samples/sec: 7853.08 - lr: 0.000026 - momentum: 0.000000
2023-10-19 19:52:27,388 epoch 3 - iter 712/1786 - loss 0.35168554 - time (sec): 12.79 - samples/sec: 7808.77 - lr: 0.000025 - momentum: 0.000000
2023-10-19 19:52:30,291 epoch 3 - iter 890/1786 - loss 0.35595638 - time (sec): 15.70 - samples/sec: 8022.78 - lr: 0.000025 - momentum: 0.000000
2023-10-19 19:52:33,085 epoch 3 - iter 1068/1786 - loss 0.35812865 - time (sec): 18.49 - samples/sec: 8195.12 - lr: 0.000025 - momentum: 0.000000
2023-10-19 19:52:36,081 epoch 3 - iter 1246/1786 - loss 0.35668010 - time (sec): 21.49 - samples/sec: 8128.47 - lr: 0.000024 - momentum: 0.000000
2023-10-19 19:52:39,100 epoch 3 - iter 1424/1786 - loss 0.35997953 - time (sec): 24.50 - samples/sec: 8132.30 - lr: 0.000024 - momentum: 0.000000
2023-10-19 19:52:42,129 epoch 3 - iter 1602/1786 - loss 0.36107656 - time (sec): 27.53 - samples/sec: 8152.51 - lr: 0.000024 - momentum: 0.000000
2023-10-19 19:52:45,104 epoch 3 - iter 1780/1786 - loss 0.35849403 - time (sec): 30.51 - samples/sec: 8127.99 - lr: 0.000023 - momentum: 0.000000
2023-10-19 19:52:45,209 ----------------------------------------------------------------------------------------------------
2023-10-19 19:52:45,209 EPOCH 3 done: loss 0.3587 - lr: 0.000023
2023-10-19 19:52:47,579 DEV : loss 0.2307889312505722 - f1-score (micro avg) 0.4343
2023-10-19 19:52:47,594 saving best model
2023-10-19 19:52:47,627 ----------------------------------------------------------------------------------------------------
2023-10-19 19:52:50,603 epoch 4 - iter 178/1786 - loss 0.32503973 - time (sec): 2.98 - samples/sec: 8709.72 - lr: 0.000023 - momentum: 0.000000
2023-10-19 19:52:53,661 epoch 4 - iter 356/1786 - loss 0.33876825 - time (sec): 6.03 - samples/sec: 8239.92 - lr: 0.000023 - momentum: 0.000000
2023-10-19 19:52:56,706 epoch 4 - iter 534/1786 - loss 0.34909168 - time (sec): 9.08 - samples/sec: 8139.72 - lr: 0.000022 - momentum: 0.000000
2023-10-19 19:52:59,800 epoch 4 - iter 712/1786 - loss 0.33361868 - time (sec): 12.17 - samples/sec: 8180.84 - lr: 0.000022 - momentum: 0.000000
2023-10-19 19:53:02,796 epoch 4 - iter 890/1786 - loss 0.32921053 - time (sec): 15.17 - samples/sec: 8202.07 - lr: 0.000022 - momentum: 0.000000
2023-10-19 19:53:05,850 epoch 4 - iter 1068/1786 - loss 0.32434636 - time (sec): 18.22 - samples/sec: 8142.54 - lr: 0.000021 - momentum: 0.000000
2023-10-19 19:53:08,954 epoch 4 - iter 1246/1786 - loss 0.32469591 - time (sec): 21.33 - samples/sec: 8072.21 - lr: 0.000021 - momentum: 0.000000
2023-10-19 19:53:11,970 epoch 4 - iter 1424/1786 - loss 0.32393392 - time (sec): 24.34 - samples/sec: 8076.36 - lr: 0.000021 - momentum: 0.000000
2023-10-19 19:53:15,084 epoch 4 - iter 1602/1786 - loss 0.32421030 - time (sec): 27.46 - samples/sec: 8099.84 - lr: 0.000020 - momentum: 0.000000
2023-10-19 19:53:18,196 epoch 4 - iter 1780/1786 - loss 0.32088289 - time (sec): 30.57 - samples/sec: 8119.54 - lr: 0.000020 - momentum: 0.000000
2023-10-19 19:53:18,287 ----------------------------------------------------------------------------------------------------
2023-10-19 19:53:18,287 EPOCH 4 done: loss 0.3211 - lr: 0.000020
2023-10-19 19:53:21,110 DEV : loss 0.2131872922182083 - f1-score (micro avg) 0.4691
2023-10-19 19:53:21,125 saving best model
2023-10-19 19:53:21,160 ----------------------------------------------------------------------------------------------------
2023-10-19 19:53:24,078 epoch 5 - iter 178/1786 - loss 0.31966126 - time (sec): 2.92 - samples/sec: 8571.95 - lr: 0.000020 - momentum: 0.000000
2023-10-19 19:53:27,158 epoch 5 - iter 356/1786 - loss 0.30979983 - time (sec): 6.00 - samples/sec: 8542.72 - lr: 0.000019 - momentum: 0.000000
2023-10-19 19:53:30,224 epoch 5 - iter 534/1786 - loss 0.30109050 - time (sec): 9.06 - samples/sec: 8366.97 - lr: 0.000019 - momentum: 0.000000
2023-10-19 19:53:33,352 epoch 5 - iter 712/1786 - loss 0.30298422 - time (sec): 12.19 - samples/sec: 8163.31 - lr: 0.000019 - momentum: 0.000000
2023-10-19 19:53:36,499 epoch 5 - iter 890/1786 - loss 0.30175508 - time (sec): 15.34 - samples/sec: 7974.28 - lr: 0.000018 - momentum: 0.000000
2023-10-19 19:53:39,560 epoch 5 - iter 1068/1786 - loss 0.29354489 - time (sec): 18.40 - samples/sec: 8027.67 - lr: 0.000018 - momentum: 0.000000
2023-10-19 19:53:42,547 epoch 5 - iter 1246/1786 - loss 0.29628320 - time (sec): 21.39 - samples/sec: 7996.02 - lr: 0.000018 - momentum: 0.000000
2023-10-19 19:53:45,679 epoch 5 - iter 1424/1786 - loss 0.29373074 - time (sec): 24.52 - samples/sec: 8009.08 - lr: 0.000017 - momentum: 0.000000
2023-10-19 19:53:48,821 epoch 5 - iter 1602/1786 - loss 0.29243459 - time (sec): 27.66 - samples/sec: 8049.63 - lr: 0.000017 - momentum: 0.000000
2023-10-19 19:53:51,974 epoch 5 - iter 1780/1786 - loss 0.29269761 - time (sec): 30.81 - samples/sec: 8046.10 - lr: 0.000017 - momentum: 0.000000
2023-10-19 19:53:52,092 ----------------------------------------------------------------------------------------------------
2023-10-19 19:53:52,092 EPOCH 5 done: loss 0.2925 - lr: 0.000017
2023-10-19 19:53:54,463 DEV : loss 0.20653365552425385 - f1-score (micro avg) 0.4815
2023-10-19 19:53:54,477 saving best model
2023-10-19 19:53:54,510 ----------------------------------------------------------------------------------------------------
2023-10-19 19:53:57,812 epoch 6 - iter 178/1786 - loss 0.26481799 - time (sec): 3.30 - samples/sec: 7606.85 - lr: 0.000016 - momentum: 0.000000
2023-10-19 19:54:00,948 epoch 6 - iter 356/1786 - loss 0.26935936 - time (sec): 6.44 - samples/sec: 7536.84 - lr: 0.000016 - momentum: 0.000000
2023-10-19 19:54:04,139 epoch 6 - iter 534/1786 - loss 0.27348573 - time (sec): 9.63 - samples/sec: 7506.83 - lr: 0.000016 - momentum: 0.000000
2023-10-19 19:54:07,207 epoch 6 - iter 712/1786 - loss 0.27134878 - time (sec): 12.70 - samples/sec: 7760.38 - lr: 0.000015 - momentum: 0.000000
2023-10-19 19:54:10,303 epoch 6 - iter 890/1786 - loss 0.26920475 - time (sec): 15.79 - samples/sec: 7915.32 - lr: 0.000015 - momentum: 0.000000
2023-10-19 19:54:13,384 epoch 6 - iter 1068/1786 - loss 0.27009143 - time (sec): 18.87 - samples/sec: 7904.76 - lr: 0.000015 - momentum: 0.000000
2023-10-19 19:54:16,439 epoch 6 - iter 1246/1786 - loss 0.27126902 - time (sec): 21.93 - samples/sec: 7886.87 - lr: 0.000014 - momentum: 0.000000
2023-10-19 19:54:19,503 epoch 6 - iter 1424/1786 - loss 0.27292987 - time (sec): 24.99 - samples/sec: 7902.13 - lr: 0.000014 - momentum: 0.000000
2023-10-19 19:54:22,691 epoch 6 - iter 1602/1786 - loss 0.27225979 - time (sec): 28.18 - samples/sec: 7926.75 - lr: 0.000014 - momentum: 0.000000
2023-10-19 19:54:25,821 epoch 6 - iter 1780/1786 - loss 0.27262868 - time (sec): 31.31 - samples/sec: 7923.30 - lr: 0.000013 - momentum: 0.000000
2023-10-19 19:54:25,918 ----------------------------------------------------------------------------------------------------
2023-10-19 19:54:25,918 EPOCH 6 done: loss 0.2727 - lr: 0.000013
2023-10-19 19:54:28,745 DEV : loss 0.2003042846918106 - f1-score (micro avg) 0.4843
2023-10-19 19:54:28,759 saving best model
2023-10-19 19:54:28,791 ----------------------------------------------------------------------------------------------------
2023-10-19 19:54:31,964 epoch 7 - iter 178/1786 - loss 0.24380929 - time (sec): 3.17 - samples/sec: 8323.54 - lr: 0.000013 - momentum: 0.000000
2023-10-19 19:54:35,071 epoch 7 - iter 356/1786 - loss 0.25726475 - time (sec): 6.28 - samples/sec: 8238.91 - lr: 0.000013 - momentum: 0.000000
2023-10-19 19:54:38,117 epoch 7 - iter 534/1786 - loss 0.25410518 - time (sec): 9.33 - samples/sec: 8058.86 - lr: 0.000012 - momentum: 0.000000
2023-10-19 19:54:41,153 epoch 7 - iter 712/1786 - loss 0.25769061 - time (sec): 12.36 - samples/sec: 7964.33 - lr: 0.000012 - momentum: 0.000000
2023-10-19 19:54:44,262 epoch 7 - iter 890/1786 - loss 0.25823402 - time (sec): 15.47 - samples/sec: 7972.49 - lr: 0.000012 - momentum: 0.000000
2023-10-19 19:54:47,394 epoch 7 - iter 1068/1786 - loss 0.25747742 - time (sec): 18.60 - samples/sec: 7959.44 - lr: 0.000011 - momentum: 0.000000
2023-10-19 19:54:50,595 epoch 7 - iter 1246/1786 - loss 0.25719904 - time (sec): 21.80 - samples/sec: 7962.56 - lr: 0.000011 - momentum: 0.000000
2023-10-19 19:54:53,694 epoch 7 - iter 1424/1786 - loss 0.25589988 - time (sec): 24.90 - samples/sec: 8050.60 - lr: 0.000011 - momentum: 0.000000
2023-10-19 19:54:56,745 epoch 7 - iter 1602/1786 - loss 0.25845771 - time (sec): 27.95 - samples/sec: 8021.59 - lr: 0.000010 - momentum: 0.000000
2023-10-19 19:54:59,724 epoch 7 - iter 1780/1786 - loss 0.25787745 - time (sec): 30.93 - samples/sec: 8027.94 - lr: 0.000010 - momentum: 0.000000
2023-10-19 19:54:59,817 ----------------------------------------------------------------------------------------------------
2023-10-19 19:54:59,818 EPOCH 7 done: loss 0.2580 - lr: 0.000010
2023-10-19 19:55:02,193 DEV : loss 0.2004682868719101 - f1-score (micro avg) 0.5019
2023-10-19 19:55:02,209 saving best model
2023-10-19 19:55:02,246 ----------------------------------------------------------------------------------------------------
2023-10-19 19:55:05,316 epoch 8 - iter 178/1786 - loss 0.24861152 - time (sec): 3.07 - samples/sec: 8175.43 - lr: 0.000010 - momentum: 0.000000
2023-10-19 19:55:08,419 epoch 8 - iter 356/1786 - loss 0.24086117 - time (sec): 6.17 - samples/sec: 8098.43 - lr: 0.000009 - momentum: 0.000000
2023-10-19 19:55:11,525 epoch 8 - iter 534/1786 - loss 0.24815113 - time (sec): 9.28 - samples/sec: 7956.56 - lr: 0.000009 - momentum: 0.000000
2023-10-19 19:55:14,603 epoch 8 - iter 712/1786 - loss 0.24997628 - time (sec): 12.36 - samples/sec: 7987.55 - lr: 0.000009 - momentum: 0.000000
2023-10-19 19:55:17,675 epoch 8 - iter 890/1786 - loss 0.24787170 - time (sec): 15.43 - samples/sec: 8019.54 - lr: 0.000008 - momentum: 0.000000
2023-10-19 19:55:20,753 epoch 8 - iter 1068/1786 - loss 0.24807736 - time (sec): 18.51 - samples/sec: 8065.54 - lr: 0.000008 - momentum: 0.000000
2023-10-19 19:55:23,762 epoch 8 - iter 1246/1786 - loss 0.24768224 - time (sec): 21.52 - samples/sec: 8087.21 - lr: 0.000008 - momentum: 0.000000
2023-10-19 19:55:26,840 epoch 8 - iter 1424/1786 - loss 0.24665818 - time (sec): 24.59 - samples/sec: 8095.57 - lr: 0.000007 - momentum: 0.000000
2023-10-19 19:55:29,829 epoch 8 - iter 1602/1786 - loss 0.24985874 - time (sec): 27.58 - samples/sec: 8098.97 - lr: 0.000007 - momentum: 0.000000
2023-10-19 19:55:32,995 epoch 8 - iter 1780/1786 - loss 0.25000798 - time (sec): 30.75 - samples/sec: 8064.42 - lr: 0.000007 - momentum: 0.000000
2023-10-19 19:55:33,104 ----------------------------------------------------------------------------------------------------
2023-10-19 19:55:33,104 EPOCH 8 done: loss 0.2499 - lr: 0.000007
2023-10-19 19:55:35,972 DEV : loss 0.19668100774288177 - f1-score (micro avg) 0.5031
2023-10-19 19:55:35,986 saving best model
2023-10-19 19:55:36,020 ----------------------------------------------------------------------------------------------------
2023-10-19 19:55:39,211 epoch 9 - iter 178/1786 - loss 0.23800143 - time (sec): 3.19 - samples/sec: 8302.38 - lr: 0.000006 - momentum: 0.000000
2023-10-19 19:55:42,260 epoch 9 - iter 356/1786 - loss 0.23583672 - time (sec): 6.24 - samples/sec: 8301.37 - lr: 0.000006 - momentum: 0.000000
2023-10-19 19:55:45,281 epoch 9 - iter 534/1786 - loss 0.23499003 - time (sec): 9.26 - samples/sec: 8227.23 - lr: 0.000006 - momentum: 0.000000
2023-10-19 19:55:48,244 epoch 9 - iter 712/1786 - loss 0.23888394 - time (sec): 12.22 - samples/sec: 8139.72 - lr: 0.000005 - momentum: 0.000000
2023-10-19 19:55:51,288 epoch 9 - iter 890/1786 - loss 0.24062823 - time (sec): 15.27 - samples/sec: 8068.23 - lr: 0.000005 - momentum: 0.000000
2023-10-19 19:55:54,409 epoch 9 - iter 1068/1786 - loss 0.24208841 - time (sec): 18.39 - samples/sec: 8095.06 - lr: 0.000005 - momentum: 0.000000
2023-10-19 19:55:57,156 epoch 9 - iter 1246/1786 - loss 0.24572741 - time (sec): 21.13 - samples/sec: 8235.25 - lr: 0.000004 - momentum: 0.000000
2023-10-19 19:56:00,158 epoch 9 - iter 1424/1786 - loss 0.24461181 - time (sec): 24.14 - samples/sec: 8218.35 - lr: 0.000004 - momentum: 0.000000
2023-10-19 19:56:03,242 epoch 9 - iter 1602/1786 - loss 0.24434313 - time (sec): 27.22 - samples/sec: 8220.44 - lr: 0.000004 - momentum: 0.000000
2023-10-19 19:56:06,397 epoch 9 - iter 1780/1786 - loss 0.24230508 - time (sec): 30.38 - samples/sec: 8165.70 - lr: 0.000003 - momentum: 0.000000
2023-10-19 19:56:06,497 ----------------------------------------------------------------------------------------------------
2023-10-19 19:56:06,497 EPOCH 9 done: loss 0.2419 - lr: 0.000003
2023-10-19 19:56:08,853 DEV : loss 0.1973438411951065 - f1-score (micro avg) 0.508
2023-10-19 19:56:08,867 saving best model
2023-10-19 19:56:08,900 ----------------------------------------------------------------------------------------------------
2023-10-19 19:56:11,976 epoch 10 - iter 178/1786 - loss 0.24379983 - time (sec): 3.08 - samples/sec: 7549.01 - lr: 0.000003 - momentum: 0.000000
2023-10-19 19:56:15,113 epoch 10 - iter 356/1786 - loss 0.25116343 - time (sec): 6.21 - samples/sec: 7700.35 - lr: 0.000003 - momentum: 0.000000
2023-10-19 19:56:18,116 epoch 10 - iter 534/1786 - loss 0.25060293 - time (sec): 9.22 - samples/sec: 7865.97 - lr: 0.000002 - momentum: 0.000000
2023-10-19 19:56:21,260 epoch 10 - iter 712/1786 - loss 0.25011516 - time (sec): 12.36 - samples/sec: 7898.81 - lr: 0.000002 - momentum: 0.000000
2023-10-19 19:56:24,503 epoch 10 - iter 890/1786 - loss 0.24555631 - time (sec): 15.60 - samples/sec: 7783.06 - lr: 0.000002 - momentum: 0.000000
2023-10-19 19:56:27,557 epoch 10 - iter 1068/1786 - loss 0.24118692 - time (sec): 18.66 - samples/sec: 7834.94 - lr: 0.000001 - momentum: 0.000000
2023-10-19 19:56:30,299 epoch 10 - iter 1246/1786 - loss 0.23655811 - time (sec): 21.40 - samples/sec: 8048.55 - lr: 0.000001 - momentum: 0.000000
2023-10-19 19:56:33,323 epoch 10 - iter 1424/1786 - loss 0.23172756 - time (sec): 24.42 - samples/sec: 8110.64 - lr: 0.000001 - momentum: 0.000000
2023-10-19 19:56:36,431 epoch 10 - iter 1602/1786 - loss 0.23408678 - time (sec): 27.53 - samples/sec: 8114.79 - lr: 0.000000 - momentum: 0.000000
2023-10-19 19:56:39,501 epoch 10 - iter 1780/1786 - loss 0.23658798 - time (sec): 30.60 - samples/sec: 8109.58 - lr: 0.000000 - momentum: 0.000000
2023-10-19 19:56:39,599 ----------------------------------------------------------------------------------------------------
2023-10-19 19:56:39,599 EPOCH 10 done: loss 0.2373 - lr: 0.000000
2023-10-19 19:56:42,416 DEV : loss 0.19595196843147278 - f1-score (micro avg) 0.5103
2023-10-19 19:56:42,430 saving best model
2023-10-19 19:56:42,489 ----------------------------------------------------------------------------------------------------
2023-10-19 19:56:42,489 Loading model from best epoch ...
2023-10-19 19:56:42,562 SequenceTagger predicts: Dictionary with 17 tags: O, S-PER, B-PER, E-PER, I-PER, S-LOC, B-LOC, E-LOC, I-LOC, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
2023-10-19 19:56:47,193
Results:
- F-score (micro) 0.414
- F-score (macro) 0.2562
- Accuracy 0.271
By class:
precision recall f1-score support
LOC 0.3964 0.5169 0.4487 1095
PER 0.4249 0.4951 0.4573 1012
ORG 0.1581 0.0952 0.1189 357
HumanProd 0.0000 0.0000 0.0000 33
micro avg 0.3901 0.4409 0.4140 2497
macro avg 0.2449 0.2768 0.2562 2497
weighted avg 0.3686 0.4409 0.3991 2497
2023-10-19 19:56:47,193 ----------------------------------------------------------------------------------------------------
|