File size: 24,018 Bytes
f2bb399 |
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 |
2023-10-25 17:00:22,362 ----------------------------------------------------------------------------------------------------
2023-10-25 17:00:22,363 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(64001, 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=17, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-25 17:00:22,363 ----------------------------------------------------------------------------------------------------
2023-10-25 17:00:22,364 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
2023-10-25 17:00:22,364 ----------------------------------------------------------------------------------------------------
2023-10-25 17:00:22,364 Train: 7142 sentences
2023-10-25 17:00:22,364 (train_with_dev=False, train_with_test=False)
2023-10-25 17:00:22,364 ----------------------------------------------------------------------------------------------------
2023-10-25 17:00:22,364 Training Params:
2023-10-25 17:00:22,364 - learning_rate: "3e-05"
2023-10-25 17:00:22,364 - mini_batch_size: "8"
2023-10-25 17:00:22,364 - max_epochs: "10"
2023-10-25 17:00:22,364 - shuffle: "True"
2023-10-25 17:00:22,364 ----------------------------------------------------------------------------------------------------
2023-10-25 17:00:22,364 Plugins:
2023-10-25 17:00:22,364 - TensorboardLogger
2023-10-25 17:00:22,364 - LinearScheduler | warmup_fraction: '0.1'
2023-10-25 17:00:22,364 ----------------------------------------------------------------------------------------------------
2023-10-25 17:00:22,364 Final evaluation on model from best epoch (best-model.pt)
2023-10-25 17:00:22,364 - metric: "('micro avg', 'f1-score')"
2023-10-25 17:00:22,364 ----------------------------------------------------------------------------------------------------
2023-10-25 17:00:22,364 Computation:
2023-10-25 17:00:22,364 - compute on device: cuda:0
2023-10-25 17:00:22,364 - embedding storage: none
2023-10-25 17:00:22,364 ----------------------------------------------------------------------------------------------------
2023-10-25 17:00:22,364 Model training base path: "hmbench-newseye/fr-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-25 17:00:22,364 ----------------------------------------------------------------------------------------------------
2023-10-25 17:00:22,364 ----------------------------------------------------------------------------------------------------
2023-10-25 17:00:22,365 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-25 17:00:28,672 epoch 1 - iter 89/893 - loss 2.19742480 - time (sec): 6.31 - samples/sec: 4004.07 - lr: 0.000003 - momentum: 0.000000
2023-10-25 17:00:35,048 epoch 1 - iter 178/893 - loss 1.39693997 - time (sec): 12.68 - samples/sec: 4002.88 - lr: 0.000006 - momentum: 0.000000
2023-10-25 17:00:41,309 epoch 1 - iter 267/893 - loss 1.07633123 - time (sec): 18.94 - samples/sec: 3947.26 - lr: 0.000009 - momentum: 0.000000
2023-10-25 17:00:47,306 epoch 1 - iter 356/893 - loss 0.87429157 - time (sec): 24.94 - samples/sec: 3993.37 - lr: 0.000012 - momentum: 0.000000
2023-10-25 17:00:53,202 epoch 1 - iter 445/893 - loss 0.74799460 - time (sec): 30.84 - samples/sec: 4002.33 - lr: 0.000015 - momentum: 0.000000
2023-10-25 17:00:59,211 epoch 1 - iter 534/893 - loss 0.65718250 - time (sec): 36.85 - samples/sec: 4018.73 - lr: 0.000018 - momentum: 0.000000
2023-10-25 17:01:05,222 epoch 1 - iter 623/893 - loss 0.58410501 - time (sec): 42.86 - samples/sec: 4035.30 - lr: 0.000021 - momentum: 0.000000
2023-10-25 17:01:11,176 epoch 1 - iter 712/893 - loss 0.52802697 - time (sec): 48.81 - samples/sec: 4073.71 - lr: 0.000024 - momentum: 0.000000
2023-10-25 17:01:17,243 epoch 1 - iter 801/893 - loss 0.48658916 - time (sec): 54.88 - samples/sec: 4080.04 - lr: 0.000027 - momentum: 0.000000
2023-10-25 17:01:23,202 epoch 1 - iter 890/893 - loss 0.45333206 - time (sec): 60.84 - samples/sec: 4071.24 - lr: 0.000030 - momentum: 0.000000
2023-10-25 17:01:23,417 ----------------------------------------------------------------------------------------------------
2023-10-25 17:01:23,417 EPOCH 1 done: loss 0.4518 - lr: 0.000030
2023-10-25 17:01:27,249 DEV : loss 0.0998985692858696 - f1-score (micro avg) 0.7288
2023-10-25 17:01:27,270 saving best model
2023-10-25 17:01:27,743 ----------------------------------------------------------------------------------------------------
2023-10-25 17:01:33,963 epoch 2 - iter 89/893 - loss 0.11010133 - time (sec): 6.22 - samples/sec: 3972.66 - lr: 0.000030 - momentum: 0.000000
2023-10-25 17:01:40,098 epoch 2 - iter 178/893 - loss 0.10005667 - time (sec): 12.35 - samples/sec: 3975.15 - lr: 0.000029 - momentum: 0.000000
2023-10-25 17:01:46,248 epoch 2 - iter 267/893 - loss 0.10041275 - time (sec): 18.50 - samples/sec: 4054.32 - lr: 0.000029 - momentum: 0.000000
2023-10-25 17:01:52,378 epoch 2 - iter 356/893 - loss 0.10308505 - time (sec): 24.63 - samples/sec: 4108.35 - lr: 0.000029 - momentum: 0.000000
2023-10-25 17:01:58,565 epoch 2 - iter 445/893 - loss 0.10163209 - time (sec): 30.82 - samples/sec: 4120.74 - lr: 0.000028 - momentum: 0.000000
2023-10-25 17:02:04,485 epoch 2 - iter 534/893 - loss 0.10147740 - time (sec): 36.74 - samples/sec: 4101.34 - lr: 0.000028 - momentum: 0.000000
2023-10-25 17:02:10,425 epoch 2 - iter 623/893 - loss 0.10276131 - time (sec): 42.68 - samples/sec: 4107.64 - lr: 0.000028 - momentum: 0.000000
2023-10-25 17:02:16,395 epoch 2 - iter 712/893 - loss 0.10195994 - time (sec): 48.65 - samples/sec: 4123.54 - lr: 0.000027 - momentum: 0.000000
2023-10-25 17:02:22,164 epoch 2 - iter 801/893 - loss 0.10209269 - time (sec): 54.42 - samples/sec: 4096.53 - lr: 0.000027 - momentum: 0.000000
2023-10-25 17:02:28,238 epoch 2 - iter 890/893 - loss 0.10100025 - time (sec): 60.49 - samples/sec: 4102.24 - lr: 0.000027 - momentum: 0.000000
2023-10-25 17:02:28,441 ----------------------------------------------------------------------------------------------------
2023-10-25 17:02:28,441 EPOCH 2 done: loss 0.1009 - lr: 0.000027
2023-10-25 17:02:33,319 DEV : loss 0.09367502480745316 - f1-score (micro avg) 0.7629
2023-10-25 17:02:33,342 saving best model
2023-10-25 17:02:34,008 ----------------------------------------------------------------------------------------------------
2023-10-25 17:02:39,971 epoch 3 - iter 89/893 - loss 0.06332527 - time (sec): 5.96 - samples/sec: 3937.76 - lr: 0.000026 - momentum: 0.000000
2023-10-25 17:02:46,251 epoch 3 - iter 178/893 - loss 0.06194394 - time (sec): 12.24 - samples/sec: 4036.39 - lr: 0.000026 - momentum: 0.000000
2023-10-25 17:02:52,083 epoch 3 - iter 267/893 - loss 0.06017200 - time (sec): 18.07 - samples/sec: 4109.10 - lr: 0.000026 - momentum: 0.000000
2023-10-25 17:02:58,255 epoch 3 - iter 356/893 - loss 0.06164862 - time (sec): 24.24 - samples/sec: 4086.99 - lr: 0.000025 - momentum: 0.000000
2023-10-25 17:03:04,363 epoch 3 - iter 445/893 - loss 0.06151607 - time (sec): 30.35 - samples/sec: 4106.56 - lr: 0.000025 - momentum: 0.000000
2023-10-25 17:03:10,344 epoch 3 - iter 534/893 - loss 0.06067296 - time (sec): 36.33 - samples/sec: 4120.16 - lr: 0.000025 - momentum: 0.000000
2023-10-25 17:03:16,234 epoch 3 - iter 623/893 - loss 0.06086561 - time (sec): 42.22 - samples/sec: 4129.45 - lr: 0.000024 - momentum: 0.000000
2023-10-25 17:03:22,147 epoch 3 - iter 712/893 - loss 0.06099317 - time (sec): 48.13 - samples/sec: 4094.37 - lr: 0.000024 - momentum: 0.000000
2023-10-25 17:03:28,280 epoch 3 - iter 801/893 - loss 0.06051995 - time (sec): 54.27 - samples/sec: 4119.00 - lr: 0.000024 - momentum: 0.000000
2023-10-25 17:03:34,251 epoch 3 - iter 890/893 - loss 0.06220034 - time (sec): 60.24 - samples/sec: 4117.78 - lr: 0.000023 - momentum: 0.000000
2023-10-25 17:03:34,451 ----------------------------------------------------------------------------------------------------
2023-10-25 17:03:34,451 EPOCH 3 done: loss 0.0624 - lr: 0.000023
2023-10-25 17:03:39,555 DEV : loss 0.10349678248167038 - f1-score (micro avg) 0.7851
2023-10-25 17:03:39,573 saving best model
2023-10-25 17:03:40,237 ----------------------------------------------------------------------------------------------------
2023-10-25 17:03:46,372 epoch 4 - iter 89/893 - loss 0.03754159 - time (sec): 6.13 - samples/sec: 4230.43 - lr: 0.000023 - momentum: 0.000000
2023-10-25 17:03:52,281 epoch 4 - iter 178/893 - loss 0.04483007 - time (sec): 12.04 - samples/sec: 4282.44 - lr: 0.000023 - momentum: 0.000000
2023-10-25 17:03:57,977 epoch 4 - iter 267/893 - loss 0.04464268 - time (sec): 17.74 - samples/sec: 4228.06 - lr: 0.000022 - momentum: 0.000000
2023-10-25 17:04:04,177 epoch 4 - iter 356/893 - loss 0.04410251 - time (sec): 23.94 - samples/sec: 4134.91 - lr: 0.000022 - momentum: 0.000000
2023-10-25 17:04:10,450 epoch 4 - iter 445/893 - loss 0.04290747 - time (sec): 30.21 - samples/sec: 4118.64 - lr: 0.000022 - momentum: 0.000000
2023-10-25 17:04:16,490 epoch 4 - iter 534/893 - loss 0.04337588 - time (sec): 36.25 - samples/sec: 4135.74 - lr: 0.000021 - momentum: 0.000000
2023-10-25 17:04:22,526 epoch 4 - iter 623/893 - loss 0.04450695 - time (sec): 42.29 - samples/sec: 4100.04 - lr: 0.000021 - momentum: 0.000000
2023-10-25 17:04:28,608 epoch 4 - iter 712/893 - loss 0.04474457 - time (sec): 48.37 - samples/sec: 4104.66 - lr: 0.000021 - momentum: 0.000000
2023-10-25 17:04:34,686 epoch 4 - iter 801/893 - loss 0.04576971 - time (sec): 54.45 - samples/sec: 4112.57 - lr: 0.000020 - momentum: 0.000000
2023-10-25 17:04:40,591 epoch 4 - iter 890/893 - loss 0.04490676 - time (sec): 60.35 - samples/sec: 4097.87 - lr: 0.000020 - momentum: 0.000000
2023-10-25 17:04:40,899 ----------------------------------------------------------------------------------------------------
2023-10-25 17:04:40,904 EPOCH 4 done: loss 0.0447 - lr: 0.000020
2023-10-25 17:04:45,230 DEV : loss 0.14620383083820343 - f1-score (micro avg) 0.8037
2023-10-25 17:04:45,256 saving best model
2023-10-25 17:04:46,044 ----------------------------------------------------------------------------------------------------
2023-10-25 17:04:52,073 epoch 5 - iter 89/893 - loss 0.03530497 - time (sec): 6.03 - samples/sec: 3848.91 - lr: 0.000020 - momentum: 0.000000
2023-10-25 17:04:58,050 epoch 5 - iter 178/893 - loss 0.03402744 - time (sec): 12.00 - samples/sec: 4001.34 - lr: 0.000019 - momentum: 0.000000
2023-10-25 17:05:04,150 epoch 5 - iter 267/893 - loss 0.03369793 - time (sec): 18.10 - samples/sec: 4023.43 - lr: 0.000019 - momentum: 0.000000
2023-10-25 17:05:10,344 epoch 5 - iter 356/893 - loss 0.03388932 - time (sec): 24.30 - samples/sec: 4021.69 - lr: 0.000019 - momentum: 0.000000
2023-10-25 17:05:16,425 epoch 5 - iter 445/893 - loss 0.03377847 - time (sec): 30.38 - samples/sec: 4048.47 - lr: 0.000018 - momentum: 0.000000
2023-10-25 17:05:22,583 epoch 5 - iter 534/893 - loss 0.03360074 - time (sec): 36.53 - samples/sec: 4053.71 - lr: 0.000018 - momentum: 0.000000
2023-10-25 17:05:28,654 epoch 5 - iter 623/893 - loss 0.03242307 - time (sec): 42.61 - samples/sec: 4046.48 - lr: 0.000018 - momentum: 0.000000
2023-10-25 17:05:34,820 epoch 5 - iter 712/893 - loss 0.03229538 - time (sec): 48.77 - samples/sec: 4034.35 - lr: 0.000017 - momentum: 0.000000
2023-10-25 17:05:40,937 epoch 5 - iter 801/893 - loss 0.03238963 - time (sec): 54.89 - samples/sec: 4065.34 - lr: 0.000017 - momentum: 0.000000
2023-10-25 17:05:47,041 epoch 5 - iter 890/893 - loss 0.03261197 - time (sec): 60.99 - samples/sec: 4063.33 - lr: 0.000017 - momentum: 0.000000
2023-10-25 17:05:47,251 ----------------------------------------------------------------------------------------------------
2023-10-25 17:05:47,251 EPOCH 5 done: loss 0.0325 - lr: 0.000017
2023-10-25 17:05:52,885 DEV : loss 0.1633528769016266 - f1-score (micro avg) 0.797
2023-10-25 17:05:52,915 ----------------------------------------------------------------------------------------------------
2023-10-25 17:05:59,081 epoch 6 - iter 89/893 - loss 0.03029989 - time (sec): 6.16 - samples/sec: 3842.51 - lr: 0.000016 - momentum: 0.000000
2023-10-25 17:06:05,172 epoch 6 - iter 178/893 - loss 0.02564591 - time (sec): 12.26 - samples/sec: 3799.19 - lr: 0.000016 - momentum: 0.000000
2023-10-25 17:06:11,310 epoch 6 - iter 267/893 - loss 0.02415048 - time (sec): 18.39 - samples/sec: 3923.45 - lr: 0.000016 - momentum: 0.000000
2023-10-25 17:06:17,327 epoch 6 - iter 356/893 - loss 0.02531047 - time (sec): 24.41 - samples/sec: 3979.94 - lr: 0.000015 - momentum: 0.000000
2023-10-25 17:06:23,353 epoch 6 - iter 445/893 - loss 0.02540534 - time (sec): 30.44 - samples/sec: 4031.53 - lr: 0.000015 - momentum: 0.000000
2023-10-25 17:06:29,489 epoch 6 - iter 534/893 - loss 0.02638207 - time (sec): 36.57 - samples/sec: 4054.77 - lr: 0.000015 - momentum: 0.000000
2023-10-25 17:06:35,690 epoch 6 - iter 623/893 - loss 0.02582057 - time (sec): 42.77 - samples/sec: 4044.98 - lr: 0.000014 - momentum: 0.000000
2023-10-25 17:06:41,917 epoch 6 - iter 712/893 - loss 0.02512173 - time (sec): 49.00 - samples/sec: 4050.05 - lr: 0.000014 - momentum: 0.000000
2023-10-25 17:06:48,057 epoch 6 - iter 801/893 - loss 0.02591665 - time (sec): 55.14 - samples/sec: 4038.91 - lr: 0.000014 - momentum: 0.000000
2023-10-25 17:06:54,250 epoch 6 - iter 890/893 - loss 0.02583713 - time (sec): 61.33 - samples/sec: 4048.55 - lr: 0.000013 - momentum: 0.000000
2023-10-25 17:06:54,451 ----------------------------------------------------------------------------------------------------
2023-10-25 17:06:54,452 EPOCH 6 done: loss 0.0259 - lr: 0.000013
2023-10-25 17:06:59,824 DEV : loss 0.18684536218643188 - f1-score (micro avg) 0.7976
2023-10-25 17:06:59,848 ----------------------------------------------------------------------------------------------------
2023-10-25 17:07:06,024 epoch 7 - iter 89/893 - loss 0.01485614 - time (sec): 6.17 - samples/sec: 3881.67 - lr: 0.000013 - momentum: 0.000000
2023-10-25 17:07:12,130 epoch 7 - iter 178/893 - loss 0.01598830 - time (sec): 12.28 - samples/sec: 3958.05 - lr: 0.000013 - momentum: 0.000000
2023-10-25 17:07:18,166 epoch 7 - iter 267/893 - loss 0.01783078 - time (sec): 18.32 - samples/sec: 4076.48 - lr: 0.000012 - momentum: 0.000000
2023-10-25 17:07:24,070 epoch 7 - iter 356/893 - loss 0.01936177 - time (sec): 24.22 - samples/sec: 4105.59 - lr: 0.000012 - momentum: 0.000000
2023-10-25 17:07:30,059 epoch 7 - iter 445/893 - loss 0.01988732 - time (sec): 30.21 - samples/sec: 4146.37 - lr: 0.000012 - momentum: 0.000000
2023-10-25 17:07:36,108 epoch 7 - iter 534/893 - loss 0.01928793 - time (sec): 36.26 - samples/sec: 4162.19 - lr: 0.000011 - momentum: 0.000000
2023-10-25 17:07:42,411 epoch 7 - iter 623/893 - loss 0.02039273 - time (sec): 42.56 - samples/sec: 4121.89 - lr: 0.000011 - momentum: 0.000000
2023-10-25 17:07:48,261 epoch 7 - iter 712/893 - loss 0.02001692 - time (sec): 48.41 - samples/sec: 4092.95 - lr: 0.000011 - momentum: 0.000000
2023-10-25 17:07:54,353 epoch 7 - iter 801/893 - loss 0.02017325 - time (sec): 54.50 - samples/sec: 4087.57 - lr: 0.000010 - momentum: 0.000000
2023-10-25 17:08:00,473 epoch 7 - iter 890/893 - loss 0.01991182 - time (sec): 60.62 - samples/sec: 4094.74 - lr: 0.000010 - momentum: 0.000000
2023-10-25 17:08:00,656 ----------------------------------------------------------------------------------------------------
2023-10-25 17:08:00,657 EPOCH 7 done: loss 0.0199 - lr: 0.000010
2023-10-25 17:08:05,217 DEV : loss 0.2105928510427475 - f1-score (micro avg) 0.8011
2023-10-25 17:08:05,237 ----------------------------------------------------------------------------------------------------
2023-10-25 17:08:11,326 epoch 8 - iter 89/893 - loss 0.01743845 - time (sec): 6.09 - samples/sec: 4235.38 - lr: 0.000010 - momentum: 0.000000
2023-10-25 17:08:17,447 epoch 8 - iter 178/893 - loss 0.01794746 - time (sec): 12.21 - samples/sec: 4130.72 - lr: 0.000009 - momentum: 0.000000
2023-10-25 17:08:23,444 epoch 8 - iter 267/893 - loss 0.01517857 - time (sec): 18.21 - samples/sec: 4109.13 - lr: 0.000009 - momentum: 0.000000
2023-10-25 17:08:29,466 epoch 8 - iter 356/893 - loss 0.01542169 - time (sec): 24.23 - samples/sec: 4050.01 - lr: 0.000009 - momentum: 0.000000
2023-10-25 17:08:35,548 epoch 8 - iter 445/893 - loss 0.01465711 - time (sec): 30.31 - samples/sec: 4032.48 - lr: 0.000008 - momentum: 0.000000
2023-10-25 17:08:41,861 epoch 8 - iter 534/893 - loss 0.01456755 - time (sec): 36.62 - samples/sec: 4029.83 - lr: 0.000008 - momentum: 0.000000
2023-10-25 17:08:47,636 epoch 8 - iter 623/893 - loss 0.01409835 - time (sec): 42.40 - samples/sec: 4055.86 - lr: 0.000008 - momentum: 0.000000
2023-10-25 17:08:53,646 epoch 8 - iter 712/893 - loss 0.01384014 - time (sec): 48.41 - samples/sec: 4051.49 - lr: 0.000007 - momentum: 0.000000
2023-10-25 17:08:59,620 epoch 8 - iter 801/893 - loss 0.01422010 - time (sec): 54.38 - samples/sec: 4078.14 - lr: 0.000007 - momentum: 0.000000
2023-10-25 17:09:05,944 epoch 8 - iter 890/893 - loss 0.01456695 - time (sec): 60.71 - samples/sec: 4085.39 - lr: 0.000007 - momentum: 0.000000
2023-10-25 17:09:06,139 ----------------------------------------------------------------------------------------------------
2023-10-25 17:09:06,140 EPOCH 8 done: loss 0.0146 - lr: 0.000007
2023-10-25 17:09:11,159 DEV : loss 0.21266496181488037 - f1-score (micro avg) 0.7947
2023-10-25 17:09:11,180 ----------------------------------------------------------------------------------------------------
2023-10-25 17:09:17,250 epoch 9 - iter 89/893 - loss 0.00472214 - time (sec): 6.07 - samples/sec: 4171.11 - lr: 0.000006 - momentum: 0.000000
2023-10-25 17:09:23,216 epoch 9 - iter 178/893 - loss 0.00879912 - time (sec): 12.03 - samples/sec: 4177.79 - lr: 0.000006 - momentum: 0.000000
2023-10-25 17:09:29,325 epoch 9 - iter 267/893 - loss 0.01001564 - time (sec): 18.14 - samples/sec: 4086.48 - lr: 0.000006 - momentum: 0.000000
2023-10-25 17:09:35,358 epoch 9 - iter 356/893 - loss 0.01086924 - time (sec): 24.18 - samples/sec: 4140.04 - lr: 0.000005 - momentum: 0.000000
2023-10-25 17:09:41,382 epoch 9 - iter 445/893 - loss 0.01063271 - time (sec): 30.20 - samples/sec: 4151.32 - lr: 0.000005 - momentum: 0.000000
2023-10-25 17:09:47,352 epoch 9 - iter 534/893 - loss 0.01049232 - time (sec): 36.17 - samples/sec: 4114.54 - lr: 0.000005 - momentum: 0.000000
2023-10-25 17:09:53,532 epoch 9 - iter 623/893 - loss 0.01032612 - time (sec): 42.35 - samples/sec: 4133.27 - lr: 0.000004 - momentum: 0.000000
2023-10-25 17:09:59,456 epoch 9 - iter 712/893 - loss 0.01040970 - time (sec): 48.27 - samples/sec: 4105.70 - lr: 0.000004 - momentum: 0.000000
2023-10-25 17:10:05,526 epoch 9 - iter 801/893 - loss 0.01045359 - time (sec): 54.34 - samples/sec: 4091.22 - lr: 0.000004 - momentum: 0.000000
2023-10-25 17:10:11,623 epoch 9 - iter 890/893 - loss 0.01060696 - time (sec): 60.44 - samples/sec: 4100.12 - lr: 0.000003 - momentum: 0.000000
2023-10-25 17:10:11,820 ----------------------------------------------------------------------------------------------------
2023-10-25 17:10:11,820 EPOCH 9 done: loss 0.0106 - lr: 0.000003
2023-10-25 17:10:17,087 DEV : loss 0.2295289933681488 - f1-score (micro avg) 0.8011
2023-10-25 17:10:17,112 ----------------------------------------------------------------------------------------------------
2023-10-25 17:10:23,006 epoch 10 - iter 89/893 - loss 0.01002804 - time (sec): 5.89 - samples/sec: 4117.60 - lr: 0.000003 - momentum: 0.000000
2023-10-25 17:10:29,006 epoch 10 - iter 178/893 - loss 0.01019071 - time (sec): 11.89 - samples/sec: 3983.06 - lr: 0.000003 - momentum: 0.000000
2023-10-25 17:10:35,268 epoch 10 - iter 267/893 - loss 0.00952875 - time (sec): 18.15 - samples/sec: 4042.88 - lr: 0.000002 - momentum: 0.000000
2023-10-25 17:10:41,474 epoch 10 - iter 356/893 - loss 0.00888864 - time (sec): 24.36 - samples/sec: 4047.46 - lr: 0.000002 - momentum: 0.000000
2023-10-25 17:10:47,407 epoch 10 - iter 445/893 - loss 0.00924368 - time (sec): 30.29 - samples/sec: 4025.62 - lr: 0.000002 - momentum: 0.000000
2023-10-25 17:10:53,592 epoch 10 - iter 534/893 - loss 0.00922564 - time (sec): 36.48 - samples/sec: 4056.96 - lr: 0.000001 - momentum: 0.000000
2023-10-25 17:10:59,611 epoch 10 - iter 623/893 - loss 0.00891649 - time (sec): 42.50 - samples/sec: 4071.51 - lr: 0.000001 - momentum: 0.000000
2023-10-25 17:11:05,620 epoch 10 - iter 712/893 - loss 0.00840213 - time (sec): 48.51 - samples/sec: 4045.77 - lr: 0.000001 - momentum: 0.000000
2023-10-25 17:11:11,776 epoch 10 - iter 801/893 - loss 0.00803404 - time (sec): 54.66 - samples/sec: 4059.14 - lr: 0.000000 - momentum: 0.000000
2023-10-25 17:11:18,055 epoch 10 - iter 890/893 - loss 0.00778595 - time (sec): 60.94 - samples/sec: 4068.83 - lr: 0.000000 - momentum: 0.000000
2023-10-25 17:11:18,253 ----------------------------------------------------------------------------------------------------
2023-10-25 17:11:18,253 EPOCH 10 done: loss 0.0078 - lr: 0.000000
2023-10-25 17:11:22,853 DEV : loss 0.23914724588394165 - f1-score (micro avg) 0.7997
2023-10-25 17:11:23,516 ----------------------------------------------------------------------------------------------------
2023-10-25 17:11:23,517 Loading model from best epoch ...
2023-10-25 17:11:25,628 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
2023-10-25 17:11:37,463
Results:
- F-score (micro) 0.6825
- F-score (macro) 0.5925
- Accuracy 0.5411
By class:
precision recall f1-score support
LOC 0.7044 0.6813 0.6927 1095
PER 0.7967 0.7628 0.7794 1012
ORG 0.3908 0.5966 0.4723 357
HumanProd 0.3279 0.6061 0.4255 33
micro avg 0.6648 0.7012 0.6825 2497
macro avg 0.5550 0.6617 0.5925 2497
weighted avg 0.6920 0.7012 0.6928 2497
2023-10-25 17:11:37,464 ----------------------------------------------------------------------------------------------------
|