File size: 23,978 Bytes
8994d0d |
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 |
2023-09-04 11:16:25,341 ----------------------------------------------------------------------------------------------------
2023-09-04 11:16:25,342 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()
)"
2023-09-04 11:16:25,342 ----------------------------------------------------------------------------------------------------
2023-09-04 11:16:25,342 MultiCorpus: 5901 train + 1287 dev + 1505 test sentences
- NER_HIPE_2022 Corpus: 5901 train + 1287 dev + 1505 test sentences - /app/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/fr/with_doc_seperator
2023-09-04 11:16:25,342 ----------------------------------------------------------------------------------------------------
2023-09-04 11:16:25,342 Train: 5901 sentences
2023-09-04 11:16:25,342 (train_with_dev=False, train_with_test=False)
2023-09-04 11:16:25,342 ----------------------------------------------------------------------------------------------------
2023-09-04 11:16:25,342 Training Params:
2023-09-04 11:16:25,342 - learning_rate: "3e-05"
2023-09-04 11:16:25,343 - mini_batch_size: "8"
2023-09-04 11:16:25,343 - max_epochs: "10"
2023-09-04 11:16:25,343 - shuffle: "True"
2023-09-04 11:16:25,343 ----------------------------------------------------------------------------------------------------
2023-09-04 11:16:25,343 Plugins:
2023-09-04 11:16:25,343 - LinearScheduler | warmup_fraction: '0.1'
2023-09-04 11:16:25,343 ----------------------------------------------------------------------------------------------------
2023-09-04 11:16:25,343 Final evaluation on model from best epoch (best-model.pt)
2023-09-04 11:16:25,343 - metric: "('micro avg', 'f1-score')"
2023-09-04 11:16:25,343 ----------------------------------------------------------------------------------------------------
2023-09-04 11:16:25,343 Computation:
2023-09-04 11:16:25,343 - compute on device: cuda:0
2023-09-04 11:16:25,343 - embedding storage: none
2023-09-04 11:16:25,343 ----------------------------------------------------------------------------------------------------
2023-09-04 11:16:25,343 Model training base path: "hmbench-hipe2020/fr-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2"
2023-09-04 11:16:25,343 ----------------------------------------------------------------------------------------------------
2023-09-04 11:16:25,343 ----------------------------------------------------------------------------------------------------
2023-09-04 11:16:40,182 epoch 1 - iter 73/738 - loss 3.06648327 - time (sec): 14.84 - samples/sec: 1185.94 - lr: 0.000003 - momentum: 0.000000
2023-09-04 11:16:54,521 epoch 1 - iter 146/738 - loss 2.03380649 - time (sec): 29.18 - samples/sec: 1226.77 - lr: 0.000006 - momentum: 0.000000
2023-09-04 11:17:08,295 epoch 1 - iter 219/738 - loss 1.55275994 - time (sec): 42.95 - samples/sec: 1203.95 - lr: 0.000009 - momentum: 0.000000
2023-09-04 11:17:22,200 epoch 1 - iter 292/738 - loss 1.26157976 - time (sec): 56.86 - samples/sec: 1201.71 - lr: 0.000012 - momentum: 0.000000
2023-09-04 11:17:36,211 epoch 1 - iter 365/738 - loss 1.08469915 - time (sec): 70.87 - samples/sec: 1199.52 - lr: 0.000015 - momentum: 0.000000
2023-09-04 11:17:49,770 epoch 1 - iter 438/738 - loss 0.95415774 - time (sec): 84.43 - samples/sec: 1204.60 - lr: 0.000018 - momentum: 0.000000
2023-09-04 11:18:03,731 epoch 1 - iter 511/738 - loss 0.85774269 - time (sec): 98.39 - samples/sec: 1197.64 - lr: 0.000021 - momentum: 0.000000
2023-09-04 11:18:15,906 epoch 1 - iter 584/738 - loss 0.79054942 - time (sec): 110.56 - samples/sec: 1197.35 - lr: 0.000024 - momentum: 0.000000
2023-09-04 11:18:29,496 epoch 1 - iter 657/738 - loss 0.72897502 - time (sec): 124.15 - samples/sec: 1196.74 - lr: 0.000027 - momentum: 0.000000
2023-09-04 11:18:42,826 epoch 1 - iter 730/738 - loss 0.67545291 - time (sec): 137.48 - samples/sec: 1199.69 - lr: 0.000030 - momentum: 0.000000
2023-09-04 11:18:44,111 ----------------------------------------------------------------------------------------------------
2023-09-04 11:18:44,111 EPOCH 1 done: loss 0.6710 - lr: 0.000030
2023-09-04 11:18:57,888 DEV : loss 0.1332985907793045 - f1-score (micro avg) 0.711
2023-09-04 11:18:57,916 saving best model
2023-09-04 11:18:58,390 ----------------------------------------------------------------------------------------------------
2023-09-04 11:19:09,963 epoch 2 - iter 73/738 - loss 0.14414587 - time (sec): 11.57 - samples/sec: 1306.62 - lr: 0.000030 - momentum: 0.000000
2023-09-04 11:19:23,202 epoch 2 - iter 146/738 - loss 0.14416364 - time (sec): 24.81 - samples/sec: 1262.22 - lr: 0.000029 - momentum: 0.000000
2023-09-04 11:19:36,336 epoch 2 - iter 219/738 - loss 0.14447359 - time (sec): 37.94 - samples/sec: 1250.81 - lr: 0.000029 - momentum: 0.000000
2023-09-04 11:19:49,994 epoch 2 - iter 292/738 - loss 0.13943446 - time (sec): 51.60 - samples/sec: 1223.27 - lr: 0.000029 - momentum: 0.000000
2023-09-04 11:20:02,942 epoch 2 - iter 365/738 - loss 0.14063830 - time (sec): 64.55 - samples/sec: 1217.88 - lr: 0.000028 - momentum: 0.000000
2023-09-04 11:20:17,121 epoch 2 - iter 438/738 - loss 0.13642158 - time (sec): 78.73 - samples/sec: 1212.44 - lr: 0.000028 - momentum: 0.000000
2023-09-04 11:20:32,838 epoch 2 - iter 511/738 - loss 0.13356226 - time (sec): 94.45 - samples/sec: 1205.26 - lr: 0.000028 - momentum: 0.000000
2023-09-04 11:20:46,437 epoch 2 - iter 584/738 - loss 0.12802162 - time (sec): 108.05 - samples/sec: 1204.42 - lr: 0.000027 - momentum: 0.000000
2023-09-04 11:21:00,585 epoch 2 - iter 657/738 - loss 0.12895343 - time (sec): 122.19 - samples/sec: 1205.41 - lr: 0.000027 - momentum: 0.000000
2023-09-04 11:21:15,845 epoch 2 - iter 730/738 - loss 0.12767766 - time (sec): 137.45 - samples/sec: 1198.38 - lr: 0.000027 - momentum: 0.000000
2023-09-04 11:21:17,240 ----------------------------------------------------------------------------------------------------
2023-09-04 11:21:17,241 EPOCH 2 done: loss 0.1275 - lr: 0.000027
2023-09-04 11:21:35,278 DEV : loss 0.10566549748182297 - f1-score (micro avg) 0.7649
2023-09-04 11:21:35,307 saving best model
2023-09-04 11:21:37,107 ----------------------------------------------------------------------------------------------------
2023-09-04 11:21:50,327 epoch 3 - iter 73/738 - loss 0.06542645 - time (sec): 13.22 - samples/sec: 1167.53 - lr: 0.000026 - momentum: 0.000000
2023-09-04 11:22:03,772 epoch 3 - iter 146/738 - loss 0.07399537 - time (sec): 26.66 - samples/sec: 1202.78 - lr: 0.000026 - momentum: 0.000000
2023-09-04 11:22:17,525 epoch 3 - iter 219/738 - loss 0.07697646 - time (sec): 40.42 - samples/sec: 1196.14 - lr: 0.000026 - momentum: 0.000000
2023-09-04 11:22:29,378 epoch 3 - iter 292/738 - loss 0.07623798 - time (sec): 52.27 - samples/sec: 1210.85 - lr: 0.000025 - momentum: 0.000000
2023-09-04 11:22:45,122 epoch 3 - iter 365/738 - loss 0.07410337 - time (sec): 68.01 - samples/sec: 1190.47 - lr: 0.000025 - momentum: 0.000000
2023-09-04 11:23:00,135 epoch 3 - iter 438/738 - loss 0.07256508 - time (sec): 83.03 - samples/sec: 1197.06 - lr: 0.000025 - momentum: 0.000000
2023-09-04 11:23:13,662 epoch 3 - iter 511/738 - loss 0.07021164 - time (sec): 96.55 - samples/sec: 1195.05 - lr: 0.000024 - momentum: 0.000000
2023-09-04 11:23:27,456 epoch 3 - iter 584/738 - loss 0.07127441 - time (sec): 110.35 - samples/sec: 1198.87 - lr: 0.000024 - momentum: 0.000000
2023-09-04 11:23:41,892 epoch 3 - iter 657/738 - loss 0.07075954 - time (sec): 124.78 - samples/sec: 1197.07 - lr: 0.000024 - momentum: 0.000000
2023-09-04 11:23:54,828 epoch 3 - iter 730/738 - loss 0.07175536 - time (sec): 137.72 - samples/sec: 1196.62 - lr: 0.000023 - momentum: 0.000000
2023-09-04 11:23:56,063 ----------------------------------------------------------------------------------------------------
2023-09-04 11:23:56,063 EPOCH 3 done: loss 0.0717 - lr: 0.000023
2023-09-04 11:24:13,715 DEV : loss 0.10343769192695618 - f1-score (micro avg) 0.8241
2023-09-04 11:24:13,745 saving best model
2023-09-04 11:24:15,083 ----------------------------------------------------------------------------------------------------
2023-09-04 11:24:28,030 epoch 4 - iter 73/738 - loss 0.03831422 - time (sec): 12.94 - samples/sec: 1169.21 - lr: 0.000023 - momentum: 0.000000
2023-09-04 11:24:40,926 epoch 4 - iter 146/738 - loss 0.04482622 - time (sec): 25.84 - samples/sec: 1192.96 - lr: 0.000023 - momentum: 0.000000
2023-09-04 11:24:54,494 epoch 4 - iter 219/738 - loss 0.04752270 - time (sec): 39.41 - samples/sec: 1201.44 - lr: 0.000022 - momentum: 0.000000
2023-09-04 11:25:07,213 epoch 4 - iter 292/738 - loss 0.04665692 - time (sec): 52.13 - samples/sec: 1204.30 - lr: 0.000022 - momentum: 0.000000
2023-09-04 11:25:21,351 epoch 4 - iter 365/738 - loss 0.04665212 - time (sec): 66.27 - samples/sec: 1199.76 - lr: 0.000022 - momentum: 0.000000
2023-09-04 11:25:36,569 epoch 4 - iter 438/738 - loss 0.04711492 - time (sec): 81.48 - samples/sec: 1188.35 - lr: 0.000021 - momentum: 0.000000
2023-09-04 11:25:52,845 epoch 4 - iter 511/738 - loss 0.04622890 - time (sec): 97.76 - samples/sec: 1181.64 - lr: 0.000021 - momentum: 0.000000
2023-09-04 11:26:05,967 epoch 4 - iter 584/738 - loss 0.04565772 - time (sec): 110.88 - samples/sec: 1190.01 - lr: 0.000021 - momentum: 0.000000
2023-09-04 11:26:20,323 epoch 4 - iter 657/738 - loss 0.04774170 - time (sec): 125.24 - samples/sec: 1187.85 - lr: 0.000020 - momentum: 0.000000
2023-09-04 11:26:33,220 epoch 4 - iter 730/738 - loss 0.04754700 - time (sec): 138.14 - samples/sec: 1193.00 - lr: 0.000020 - momentum: 0.000000
2023-09-04 11:26:34,562 ----------------------------------------------------------------------------------------------------
2023-09-04 11:26:34,562 EPOCH 4 done: loss 0.0476 - lr: 0.000020
2023-09-04 11:26:52,250 DEV : loss 0.1501348614692688 - f1-score (micro avg) 0.8197
2023-09-04 11:26:52,279 ----------------------------------------------------------------------------------------------------
2023-09-04 11:27:05,999 epoch 5 - iter 73/738 - loss 0.04082482 - time (sec): 13.72 - samples/sec: 1199.10 - lr: 0.000020 - momentum: 0.000000
2023-09-04 11:27:20,024 epoch 5 - iter 146/738 - loss 0.03463403 - time (sec): 27.74 - samples/sec: 1190.19 - lr: 0.000019 - momentum: 0.000000
2023-09-04 11:27:33,408 epoch 5 - iter 219/738 - loss 0.03390961 - time (sec): 41.13 - samples/sec: 1214.22 - lr: 0.000019 - momentum: 0.000000
2023-09-04 11:27:46,798 epoch 5 - iter 292/738 - loss 0.02989132 - time (sec): 54.52 - samples/sec: 1211.07 - lr: 0.000019 - momentum: 0.000000
2023-09-04 11:28:00,272 epoch 5 - iter 365/738 - loss 0.03275215 - time (sec): 67.99 - samples/sec: 1207.28 - lr: 0.000018 - momentum: 0.000000
2023-09-04 11:28:13,250 epoch 5 - iter 438/738 - loss 0.03347779 - time (sec): 80.97 - samples/sec: 1204.45 - lr: 0.000018 - momentum: 0.000000
2023-09-04 11:28:27,107 epoch 5 - iter 511/738 - loss 0.03299783 - time (sec): 94.83 - samples/sec: 1199.22 - lr: 0.000018 - momentum: 0.000000
2023-09-04 11:28:42,630 epoch 5 - iter 584/738 - loss 0.03455838 - time (sec): 110.35 - samples/sec: 1192.27 - lr: 0.000017 - momentum: 0.000000
2023-09-04 11:28:58,406 epoch 5 - iter 657/738 - loss 0.03460168 - time (sec): 126.13 - samples/sec: 1187.75 - lr: 0.000017 - momentum: 0.000000
2023-09-04 11:29:10,450 epoch 5 - iter 730/738 - loss 0.03584338 - time (sec): 138.17 - samples/sec: 1191.65 - lr: 0.000017 - momentum: 0.000000
2023-09-04 11:29:12,019 ----------------------------------------------------------------------------------------------------
2023-09-04 11:29:12,019 EPOCH 5 done: loss 0.0358 - lr: 0.000017
2023-09-04 11:29:29,908 DEV : loss 0.16683056950569153 - f1-score (micro avg) 0.8109
2023-09-04 11:29:29,937 ----------------------------------------------------------------------------------------------------
2023-09-04 11:29:42,285 epoch 6 - iter 73/738 - loss 0.02593474 - time (sec): 12.35 - samples/sec: 1201.98 - lr: 0.000016 - momentum: 0.000000
2023-09-04 11:29:58,445 epoch 6 - iter 146/738 - loss 0.02883744 - time (sec): 28.51 - samples/sec: 1190.61 - lr: 0.000016 - momentum: 0.000000
2023-09-04 11:30:12,484 epoch 6 - iter 219/738 - loss 0.02750899 - time (sec): 42.55 - samples/sec: 1194.32 - lr: 0.000016 - momentum: 0.000000
2023-09-04 11:30:25,524 epoch 6 - iter 292/738 - loss 0.02938105 - time (sec): 55.59 - samples/sec: 1192.04 - lr: 0.000015 - momentum: 0.000000
2023-09-04 11:30:41,579 epoch 6 - iter 365/738 - loss 0.02857489 - time (sec): 71.64 - samples/sec: 1178.72 - lr: 0.000015 - momentum: 0.000000
2023-09-04 11:30:55,316 epoch 6 - iter 438/738 - loss 0.02981127 - time (sec): 85.38 - samples/sec: 1186.79 - lr: 0.000015 - momentum: 0.000000
2023-09-04 11:31:07,585 epoch 6 - iter 511/738 - loss 0.02817810 - time (sec): 97.65 - samples/sec: 1194.75 - lr: 0.000014 - momentum: 0.000000
2023-09-04 11:31:21,427 epoch 6 - iter 584/738 - loss 0.02650218 - time (sec): 111.49 - samples/sec: 1194.80 - lr: 0.000014 - momentum: 0.000000
2023-09-04 11:31:34,788 epoch 6 - iter 657/738 - loss 0.02642219 - time (sec): 124.85 - samples/sec: 1192.33 - lr: 0.000014 - momentum: 0.000000
2023-09-04 11:31:48,464 epoch 6 - iter 730/738 - loss 0.02616732 - time (sec): 138.53 - samples/sec: 1189.72 - lr: 0.000013 - momentum: 0.000000
2023-09-04 11:31:49,692 ----------------------------------------------------------------------------------------------------
2023-09-04 11:31:49,693 EPOCH 6 done: loss 0.0262 - lr: 0.000013
2023-09-04 11:32:07,584 DEV : loss 0.18466810882091522 - f1-score (micro avg) 0.8091
2023-09-04 11:32:07,614 ----------------------------------------------------------------------------------------------------
2023-09-04 11:32:23,652 epoch 7 - iter 73/738 - loss 0.01307957 - time (sec): 16.04 - samples/sec: 1061.53 - lr: 0.000013 - momentum: 0.000000
2023-09-04 11:32:35,332 epoch 7 - iter 146/738 - loss 0.01247271 - time (sec): 27.72 - samples/sec: 1148.57 - lr: 0.000013 - momentum: 0.000000
2023-09-04 11:32:51,009 epoch 7 - iter 219/738 - loss 0.01585589 - time (sec): 43.39 - samples/sec: 1160.31 - lr: 0.000012 - momentum: 0.000000
2023-09-04 11:33:06,749 epoch 7 - iter 292/738 - loss 0.01732056 - time (sec): 59.13 - samples/sec: 1164.59 - lr: 0.000012 - momentum: 0.000000
2023-09-04 11:33:18,946 epoch 7 - iter 365/738 - loss 0.01721112 - time (sec): 71.33 - samples/sec: 1173.19 - lr: 0.000012 - momentum: 0.000000
2023-09-04 11:33:31,567 epoch 7 - iter 438/738 - loss 0.01753197 - time (sec): 83.95 - samples/sec: 1179.95 - lr: 0.000011 - momentum: 0.000000
2023-09-04 11:33:44,306 epoch 7 - iter 511/738 - loss 0.01824019 - time (sec): 96.69 - samples/sec: 1189.86 - lr: 0.000011 - momentum: 0.000000
2023-09-04 11:33:57,143 epoch 7 - iter 584/738 - loss 0.01854363 - time (sec): 109.53 - samples/sec: 1191.38 - lr: 0.000011 - momentum: 0.000000
2023-09-04 11:34:11,077 epoch 7 - iter 657/738 - loss 0.01811224 - time (sec): 123.46 - samples/sec: 1186.22 - lr: 0.000010 - momentum: 0.000000
2023-09-04 11:34:26,868 epoch 7 - iter 730/738 - loss 0.01862229 - time (sec): 139.25 - samples/sec: 1183.77 - lr: 0.000010 - momentum: 0.000000
2023-09-04 11:34:28,070 ----------------------------------------------------------------------------------------------------
2023-09-04 11:34:28,070 EPOCH 7 done: loss 0.0186 - lr: 0.000010
2023-09-04 11:34:45,827 DEV : loss 0.18642598390579224 - f1-score (micro avg) 0.8208
2023-09-04 11:34:45,857 ----------------------------------------------------------------------------------------------------
2023-09-04 11:34:59,191 epoch 8 - iter 73/738 - loss 0.01522610 - time (sec): 13.33 - samples/sec: 1251.51 - lr: 0.000010 - momentum: 0.000000
2023-09-04 11:35:14,095 epoch 8 - iter 146/738 - loss 0.01519085 - time (sec): 28.24 - samples/sec: 1191.42 - lr: 0.000009 - momentum: 0.000000
2023-09-04 11:35:30,259 epoch 8 - iter 219/738 - loss 0.01828244 - time (sec): 44.40 - samples/sec: 1189.71 - lr: 0.000009 - momentum: 0.000000
2023-09-04 11:35:43,080 epoch 8 - iter 292/738 - loss 0.01832505 - time (sec): 57.22 - samples/sec: 1182.77 - lr: 0.000009 - momentum: 0.000000
2023-09-04 11:35:55,402 epoch 8 - iter 365/738 - loss 0.01661591 - time (sec): 69.54 - samples/sec: 1184.72 - lr: 0.000008 - momentum: 0.000000
2023-09-04 11:36:08,970 epoch 8 - iter 438/738 - loss 0.01667432 - time (sec): 83.11 - samples/sec: 1185.80 - lr: 0.000008 - momentum: 0.000000
2023-09-04 11:36:22,551 epoch 8 - iter 511/738 - loss 0.01570984 - time (sec): 96.69 - samples/sec: 1186.58 - lr: 0.000008 - momentum: 0.000000
2023-09-04 11:36:34,441 epoch 8 - iter 584/738 - loss 0.01509001 - time (sec): 108.58 - samples/sec: 1191.30 - lr: 0.000007 - momentum: 0.000000
2023-09-04 11:36:47,899 epoch 8 - iter 657/738 - loss 0.01454722 - time (sec): 122.04 - samples/sec: 1190.17 - lr: 0.000007 - momentum: 0.000000
2023-09-04 11:37:03,823 epoch 8 - iter 730/738 - loss 0.01433091 - time (sec): 137.97 - samples/sec: 1193.89 - lr: 0.000007 - momentum: 0.000000
2023-09-04 11:37:05,182 ----------------------------------------------------------------------------------------------------
2023-09-04 11:37:05,183 EPOCH 8 done: loss 0.0142 - lr: 0.000007
2023-09-04 11:37:22,950 DEV : loss 0.18818210065364838 - f1-score (micro avg) 0.8326
2023-09-04 11:37:22,979 saving best model
2023-09-04 11:37:24,373 ----------------------------------------------------------------------------------------------------
2023-09-04 11:37:37,826 epoch 9 - iter 73/738 - loss 0.00383387 - time (sec): 13.45 - samples/sec: 1152.61 - lr: 0.000006 - momentum: 0.000000
2023-09-04 11:37:52,233 epoch 9 - iter 146/738 - loss 0.00624195 - time (sec): 27.86 - samples/sec: 1160.13 - lr: 0.000006 - momentum: 0.000000
2023-09-04 11:38:04,879 epoch 9 - iter 219/738 - loss 0.00642501 - time (sec): 40.50 - samples/sec: 1192.56 - lr: 0.000006 - momentum: 0.000000
2023-09-04 11:38:18,603 epoch 9 - iter 292/738 - loss 0.00735751 - time (sec): 54.23 - samples/sec: 1195.23 - lr: 0.000005 - momentum: 0.000000
2023-09-04 11:38:33,870 epoch 9 - iter 365/738 - loss 0.00866516 - time (sec): 69.49 - samples/sec: 1191.19 - lr: 0.000005 - momentum: 0.000000
2023-09-04 11:38:47,019 epoch 9 - iter 438/738 - loss 0.00809663 - time (sec): 82.64 - samples/sec: 1189.03 - lr: 0.000005 - momentum: 0.000000
2023-09-04 11:39:01,987 epoch 9 - iter 511/738 - loss 0.00809937 - time (sec): 97.61 - samples/sec: 1180.04 - lr: 0.000004 - momentum: 0.000000
2023-09-04 11:39:14,903 epoch 9 - iter 584/738 - loss 0.00805249 - time (sec): 110.53 - samples/sec: 1178.10 - lr: 0.000004 - momentum: 0.000000
2023-09-04 11:39:27,998 epoch 9 - iter 657/738 - loss 0.00768456 - time (sec): 123.62 - samples/sec: 1186.33 - lr: 0.000004 - momentum: 0.000000
2023-09-04 11:39:43,357 epoch 9 - iter 730/738 - loss 0.00940618 - time (sec): 138.98 - samples/sec: 1184.47 - lr: 0.000003 - momentum: 0.000000
2023-09-04 11:39:45,117 ----------------------------------------------------------------------------------------------------
2023-09-04 11:39:45,118 EPOCH 9 done: loss 0.0097 - lr: 0.000003
2023-09-04 11:40:02,942 DEV : loss 0.19703754782676697 - f1-score (micro avg) 0.8283
2023-09-04 11:40:02,971 ----------------------------------------------------------------------------------------------------
2023-09-04 11:40:17,925 epoch 10 - iter 73/738 - loss 0.00771900 - time (sec): 14.95 - samples/sec: 1177.95 - lr: 0.000003 - momentum: 0.000000
2023-09-04 11:40:30,928 epoch 10 - iter 146/738 - loss 0.00713738 - time (sec): 27.95 - samples/sec: 1199.04 - lr: 0.000003 - momentum: 0.000000
2023-09-04 11:40:42,538 epoch 10 - iter 219/738 - loss 0.00787082 - time (sec): 39.56 - samples/sec: 1237.61 - lr: 0.000002 - momentum: 0.000000
2023-09-04 11:40:56,818 epoch 10 - iter 292/738 - loss 0.00762140 - time (sec): 53.84 - samples/sec: 1212.75 - lr: 0.000002 - momentum: 0.000000
2023-09-04 11:41:10,798 epoch 10 - iter 365/738 - loss 0.00753159 - time (sec): 67.83 - samples/sec: 1199.26 - lr: 0.000002 - momentum: 0.000000
2023-09-04 11:41:26,672 epoch 10 - iter 438/738 - loss 0.00786862 - time (sec): 83.70 - samples/sec: 1195.58 - lr: 0.000001 - momentum: 0.000000
2023-09-04 11:41:39,151 epoch 10 - iter 511/738 - loss 0.00769068 - time (sec): 96.18 - samples/sec: 1193.13 - lr: 0.000001 - momentum: 0.000000
2023-09-04 11:41:54,301 epoch 10 - iter 584/738 - loss 0.00768488 - time (sec): 111.33 - samples/sec: 1183.58 - lr: 0.000001 - momentum: 0.000000
2023-09-04 11:42:08,346 epoch 10 - iter 657/738 - loss 0.00813962 - time (sec): 125.37 - samples/sec: 1182.65 - lr: 0.000000 - momentum: 0.000000
2023-09-04 11:42:22,788 epoch 10 - iter 730/738 - loss 0.00764454 - time (sec): 139.82 - samples/sec: 1179.55 - lr: 0.000000 - momentum: 0.000000
2023-09-04 11:42:23,896 ----------------------------------------------------------------------------------------------------
2023-09-04 11:42:23,897 EPOCH 10 done: loss 0.0076 - lr: 0.000000
2023-09-04 11:42:41,733 DEV : loss 0.20217673480510712 - f1-score (micro avg) 0.8313
2023-09-04 11:42:42,259 ----------------------------------------------------------------------------------------------------
2023-09-04 11:42:42,260 Loading model from best epoch ...
2023-09-04 11:42:44,189 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-time, B-time, E-time, I-time, S-prod, B-prod, E-prod, I-prod
2023-09-04 11:42:58,808
Results:
- F-score (micro) 0.7992
- F-score (macro) 0.6961
- Accuracy 0.6896
By class:
precision recall f1-score support
loc 0.8835 0.8753 0.8794 858
pers 0.7526 0.8045 0.7777 537
org 0.4934 0.5682 0.5282 132
time 0.5303 0.6481 0.5833 54
prod 0.7368 0.6885 0.7119 61
micro avg 0.7858 0.8130 0.7992 1642
macro avg 0.6793 0.7169 0.6961 1642
weighted avg 0.7923 0.8130 0.8019 1642
2023-09-04 11:42:58,808 ----------------------------------------------------------------------------------------------------
|