File size: 23,824 Bytes
8cb5fa8 |
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 |
2023-10-25 20:47:46,272 ----------------------------------------------------------------------------------------------------
2023-10-25 20:47:46,273 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 20:47:46,273 ----------------------------------------------------------------------------------------------------
2023-10-25 20:47:46,273 MultiCorpus: 1166 train + 165 dev + 415 test sentences
- NER_HIPE_2022 Corpus: 1166 train + 165 dev + 415 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fi/with_doc_seperator
2023-10-25 20:47:46,273 ----------------------------------------------------------------------------------------------------
2023-10-25 20:47:46,273 Train: 1166 sentences
2023-10-25 20:47:46,273 (train_with_dev=False, train_with_test=False)
2023-10-25 20:47:46,273 ----------------------------------------------------------------------------------------------------
2023-10-25 20:47:46,273 Training Params:
2023-10-25 20:47:46,273 - learning_rate: "5e-05"
2023-10-25 20:47:46,273 - mini_batch_size: "8"
2023-10-25 20:47:46,273 - max_epochs: "10"
2023-10-25 20:47:46,273 - shuffle: "True"
2023-10-25 20:47:46,273 ----------------------------------------------------------------------------------------------------
2023-10-25 20:47:46,273 Plugins:
2023-10-25 20:47:46,273 - TensorboardLogger
2023-10-25 20:47:46,273 - LinearScheduler | warmup_fraction: '0.1'
2023-10-25 20:47:46,274 ----------------------------------------------------------------------------------------------------
2023-10-25 20:47:46,274 Final evaluation on model from best epoch (best-model.pt)
2023-10-25 20:47:46,274 - metric: "('micro avg', 'f1-score')"
2023-10-25 20:47:46,274 ----------------------------------------------------------------------------------------------------
2023-10-25 20:47:46,274 Computation:
2023-10-25 20:47:46,274 - compute on device: cuda:0
2023-10-25 20:47:46,274 - embedding storage: none
2023-10-25 20:47:46,274 ----------------------------------------------------------------------------------------------------
2023-10-25 20:47:46,274 Model training base path: "hmbench-newseye/fi-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-25 20:47:46,274 ----------------------------------------------------------------------------------------------------
2023-10-25 20:47:46,274 ----------------------------------------------------------------------------------------------------
2023-10-25 20:47:46,274 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-25 20:47:47,228 epoch 1 - iter 14/146 - loss 3.32619155 - time (sec): 0.95 - samples/sec: 4905.02 - lr: 0.000004 - momentum: 0.000000
2023-10-25 20:47:48,090 epoch 1 - iter 28/146 - loss 2.77463951 - time (sec): 1.81 - samples/sec: 4662.84 - lr: 0.000009 - momentum: 0.000000
2023-10-25 20:47:49,025 epoch 1 - iter 42/146 - loss 2.06209058 - time (sec): 2.75 - samples/sec: 4645.68 - lr: 0.000014 - momentum: 0.000000
2023-10-25 20:47:49,928 epoch 1 - iter 56/146 - loss 1.66590766 - time (sec): 3.65 - samples/sec: 4744.10 - lr: 0.000019 - momentum: 0.000000
2023-10-25 20:47:50,870 epoch 1 - iter 70/146 - loss 1.41744737 - time (sec): 4.60 - samples/sec: 4763.62 - lr: 0.000024 - momentum: 0.000000
2023-10-25 20:47:51,793 epoch 1 - iter 84/146 - loss 1.23294191 - time (sec): 5.52 - samples/sec: 4814.17 - lr: 0.000028 - momentum: 0.000000
2023-10-25 20:47:52,708 epoch 1 - iter 98/146 - loss 1.10453887 - time (sec): 6.43 - samples/sec: 4784.49 - lr: 0.000033 - momentum: 0.000000
2023-10-25 20:47:53,444 epoch 1 - iter 112/146 - loss 1.01746390 - time (sec): 7.17 - samples/sec: 4779.66 - lr: 0.000038 - momentum: 0.000000
2023-10-25 20:47:54,435 epoch 1 - iter 126/146 - loss 0.91742517 - time (sec): 8.16 - samples/sec: 4783.24 - lr: 0.000043 - momentum: 0.000000
2023-10-25 20:47:55,261 epoch 1 - iter 140/146 - loss 0.85738814 - time (sec): 8.99 - samples/sec: 4759.96 - lr: 0.000048 - momentum: 0.000000
2023-10-25 20:47:55,630 ----------------------------------------------------------------------------------------------------
2023-10-25 20:47:55,630 EPOCH 1 done: loss 0.8327 - lr: 0.000048
2023-10-25 20:47:56,280 DEV : loss 0.17366763949394226 - f1-score (micro avg) 0.4951
2023-10-25 20:47:56,285 saving best model
2023-10-25 20:47:56,703 ----------------------------------------------------------------------------------------------------
2023-10-25 20:47:57,513 epoch 2 - iter 14/146 - loss 0.18263706 - time (sec): 0.81 - samples/sec: 4908.96 - lr: 0.000050 - momentum: 0.000000
2023-10-25 20:47:58,350 epoch 2 - iter 28/146 - loss 0.18295105 - time (sec): 1.65 - samples/sec: 5022.09 - lr: 0.000049 - momentum: 0.000000
2023-10-25 20:47:59,282 epoch 2 - iter 42/146 - loss 0.19154440 - time (sec): 2.58 - samples/sec: 5039.20 - lr: 0.000048 - momentum: 0.000000
2023-10-25 20:48:00,078 epoch 2 - iter 56/146 - loss 0.18243784 - time (sec): 3.37 - samples/sec: 5084.29 - lr: 0.000048 - momentum: 0.000000
2023-10-25 20:48:00,853 epoch 2 - iter 70/146 - loss 0.18313209 - time (sec): 4.15 - samples/sec: 5020.62 - lr: 0.000047 - momentum: 0.000000
2023-10-25 20:48:01,828 epoch 2 - iter 84/146 - loss 0.17440435 - time (sec): 5.12 - samples/sec: 4918.17 - lr: 0.000047 - momentum: 0.000000
2023-10-25 20:48:02,875 epoch 2 - iter 98/146 - loss 0.16903145 - time (sec): 6.17 - samples/sec: 4880.88 - lr: 0.000046 - momentum: 0.000000
2023-10-25 20:48:03,747 epoch 2 - iter 112/146 - loss 0.16353225 - time (sec): 7.04 - samples/sec: 4891.08 - lr: 0.000046 - momentum: 0.000000
2023-10-25 20:48:04,623 epoch 2 - iter 126/146 - loss 0.16247055 - time (sec): 7.92 - samples/sec: 4827.66 - lr: 0.000045 - momentum: 0.000000
2023-10-25 20:48:05,426 epoch 2 - iter 140/146 - loss 0.16260366 - time (sec): 8.72 - samples/sec: 4806.35 - lr: 0.000045 - momentum: 0.000000
2023-10-25 20:48:06,000 ----------------------------------------------------------------------------------------------------
2023-10-25 20:48:06,000 EPOCH 2 done: loss 0.1626 - lr: 0.000045
2023-10-25 20:48:06,908 DEV : loss 0.10925206542015076 - f1-score (micro avg) 0.683
2023-10-25 20:48:06,913 saving best model
2023-10-25 20:48:07,451 ----------------------------------------------------------------------------------------------------
2023-10-25 20:48:08,430 epoch 3 - iter 14/146 - loss 0.08329505 - time (sec): 0.98 - samples/sec: 5121.07 - lr: 0.000044 - momentum: 0.000000
2023-10-25 20:48:09,436 epoch 3 - iter 28/146 - loss 0.07646522 - time (sec): 1.98 - samples/sec: 5046.33 - lr: 0.000043 - momentum: 0.000000
2023-10-25 20:48:10,199 epoch 3 - iter 42/146 - loss 0.07857219 - time (sec): 2.75 - samples/sec: 4897.73 - lr: 0.000043 - momentum: 0.000000
2023-10-25 20:48:11,036 epoch 3 - iter 56/146 - loss 0.08146631 - time (sec): 3.58 - samples/sec: 4909.24 - lr: 0.000042 - momentum: 0.000000
2023-10-25 20:48:11,890 epoch 3 - iter 70/146 - loss 0.08586188 - time (sec): 4.44 - samples/sec: 4941.40 - lr: 0.000042 - momentum: 0.000000
2023-10-25 20:48:12,633 epoch 3 - iter 84/146 - loss 0.08659125 - time (sec): 5.18 - samples/sec: 4866.47 - lr: 0.000041 - momentum: 0.000000
2023-10-25 20:48:13,560 epoch 3 - iter 98/146 - loss 0.08365767 - time (sec): 6.11 - samples/sec: 4921.59 - lr: 0.000041 - momentum: 0.000000
2023-10-25 20:48:14,480 epoch 3 - iter 112/146 - loss 0.08564022 - time (sec): 7.03 - samples/sec: 4903.22 - lr: 0.000040 - momentum: 0.000000
2023-10-25 20:48:15,417 epoch 3 - iter 126/146 - loss 0.08934483 - time (sec): 7.96 - samples/sec: 4888.54 - lr: 0.000040 - momentum: 0.000000
2023-10-25 20:48:16,286 epoch 3 - iter 140/146 - loss 0.08642318 - time (sec): 8.83 - samples/sec: 4847.70 - lr: 0.000039 - momentum: 0.000000
2023-10-25 20:48:16,655 ----------------------------------------------------------------------------------------------------
2023-10-25 20:48:16,655 EPOCH 3 done: loss 0.0862 - lr: 0.000039
2023-10-25 20:48:17,571 DEV : loss 0.09392837435007095 - f1-score (micro avg) 0.7484
2023-10-25 20:48:17,576 saving best model
2023-10-25 20:48:18,398 ----------------------------------------------------------------------------------------------------
2023-10-25 20:48:19,170 epoch 4 - iter 14/146 - loss 0.03589195 - time (sec): 0.77 - samples/sec: 4505.55 - lr: 0.000038 - momentum: 0.000000
2023-10-25 20:48:20,000 epoch 4 - iter 28/146 - loss 0.04570279 - time (sec): 1.60 - samples/sec: 4779.15 - lr: 0.000038 - momentum: 0.000000
2023-10-25 20:48:21,050 epoch 4 - iter 42/146 - loss 0.04554306 - time (sec): 2.65 - samples/sec: 4530.48 - lr: 0.000037 - momentum: 0.000000
2023-10-25 20:48:22,028 epoch 4 - iter 56/146 - loss 0.04561147 - time (sec): 3.62 - samples/sec: 4602.70 - lr: 0.000037 - momentum: 0.000000
2023-10-25 20:48:23,042 epoch 4 - iter 70/146 - loss 0.04012849 - time (sec): 4.64 - samples/sec: 4799.02 - lr: 0.000036 - momentum: 0.000000
2023-10-25 20:48:23,878 epoch 4 - iter 84/146 - loss 0.04312134 - time (sec): 5.47 - samples/sec: 4736.42 - lr: 0.000036 - momentum: 0.000000
2023-10-25 20:48:24,937 epoch 4 - iter 98/146 - loss 0.04547308 - time (sec): 6.53 - samples/sec: 4683.90 - lr: 0.000035 - momentum: 0.000000
2023-10-25 20:48:25,734 epoch 4 - iter 112/146 - loss 0.04785940 - time (sec): 7.33 - samples/sec: 4702.37 - lr: 0.000035 - momentum: 0.000000
2023-10-25 20:48:26,619 epoch 4 - iter 126/146 - loss 0.04800961 - time (sec): 8.22 - samples/sec: 4673.38 - lr: 0.000034 - momentum: 0.000000
2023-10-25 20:48:27,443 epoch 4 - iter 140/146 - loss 0.04853540 - time (sec): 9.04 - samples/sec: 4727.50 - lr: 0.000034 - momentum: 0.000000
2023-10-25 20:48:27,795 ----------------------------------------------------------------------------------------------------
2023-10-25 20:48:27,795 EPOCH 4 done: loss 0.0492 - lr: 0.000034
2023-10-25 20:48:28,710 DEV : loss 0.14798803627490997 - f1-score (micro avg) 0.7111
2023-10-25 20:48:28,715 ----------------------------------------------------------------------------------------------------
2023-10-25 20:48:29,538 epoch 5 - iter 14/146 - loss 0.02945627 - time (sec): 0.82 - samples/sec: 4595.35 - lr: 0.000033 - momentum: 0.000000
2023-10-25 20:48:30,394 epoch 5 - iter 28/146 - loss 0.03208317 - time (sec): 1.68 - samples/sec: 4783.75 - lr: 0.000032 - momentum: 0.000000
2023-10-25 20:48:31,343 epoch 5 - iter 42/146 - loss 0.03529632 - time (sec): 2.63 - samples/sec: 4876.90 - lr: 0.000032 - momentum: 0.000000
2023-10-25 20:48:32,177 epoch 5 - iter 56/146 - loss 0.03385465 - time (sec): 3.46 - samples/sec: 4834.24 - lr: 0.000031 - momentum: 0.000000
2023-10-25 20:48:33,089 epoch 5 - iter 70/146 - loss 0.03310305 - time (sec): 4.37 - samples/sec: 4832.85 - lr: 0.000031 - momentum: 0.000000
2023-10-25 20:48:33,897 epoch 5 - iter 84/146 - loss 0.03279721 - time (sec): 5.18 - samples/sec: 4806.60 - lr: 0.000030 - momentum: 0.000000
2023-10-25 20:48:34,879 epoch 5 - iter 98/146 - loss 0.03501715 - time (sec): 6.16 - samples/sec: 4786.64 - lr: 0.000030 - momentum: 0.000000
2023-10-25 20:48:35,798 epoch 5 - iter 112/146 - loss 0.03321133 - time (sec): 7.08 - samples/sec: 4785.39 - lr: 0.000029 - momentum: 0.000000
2023-10-25 20:48:36,745 epoch 5 - iter 126/146 - loss 0.03182421 - time (sec): 8.03 - samples/sec: 4743.23 - lr: 0.000029 - momentum: 0.000000
2023-10-25 20:48:37,639 epoch 5 - iter 140/146 - loss 0.02979870 - time (sec): 8.92 - samples/sec: 4755.87 - lr: 0.000028 - momentum: 0.000000
2023-10-25 20:48:38,075 ----------------------------------------------------------------------------------------------------
2023-10-25 20:48:38,076 EPOCH 5 done: loss 0.0316 - lr: 0.000028
2023-10-25 20:48:38,990 DEV : loss 0.14484462141990662 - f1-score (micro avg) 0.7101
2023-10-25 20:48:38,995 ----------------------------------------------------------------------------------------------------
2023-10-25 20:48:39,886 epoch 6 - iter 14/146 - loss 0.03930537 - time (sec): 0.89 - samples/sec: 4602.61 - lr: 0.000027 - momentum: 0.000000
2023-10-25 20:48:40,777 epoch 6 - iter 28/146 - loss 0.02555020 - time (sec): 1.78 - samples/sec: 4689.85 - lr: 0.000027 - momentum: 0.000000
2023-10-25 20:48:41,846 epoch 6 - iter 42/146 - loss 0.02053346 - time (sec): 2.85 - samples/sec: 4560.50 - lr: 0.000026 - momentum: 0.000000
2023-10-25 20:48:42,830 epoch 6 - iter 56/146 - loss 0.02171275 - time (sec): 3.83 - samples/sec: 4695.94 - lr: 0.000026 - momentum: 0.000000
2023-10-25 20:48:43,736 epoch 6 - iter 70/146 - loss 0.02182934 - time (sec): 4.74 - samples/sec: 4617.08 - lr: 0.000025 - momentum: 0.000000
2023-10-25 20:48:44,587 epoch 6 - iter 84/146 - loss 0.02098662 - time (sec): 5.59 - samples/sec: 4613.56 - lr: 0.000025 - momentum: 0.000000
2023-10-25 20:48:45,499 epoch 6 - iter 98/146 - loss 0.02256168 - time (sec): 6.50 - samples/sec: 4659.28 - lr: 0.000024 - momentum: 0.000000
2023-10-25 20:48:46,401 epoch 6 - iter 112/146 - loss 0.02105759 - time (sec): 7.41 - samples/sec: 4694.00 - lr: 0.000024 - momentum: 0.000000
2023-10-25 20:48:47,177 epoch 6 - iter 126/146 - loss 0.02169497 - time (sec): 8.18 - samples/sec: 4703.57 - lr: 0.000023 - momentum: 0.000000
2023-10-25 20:48:47,967 epoch 6 - iter 140/146 - loss 0.02216839 - time (sec): 8.97 - samples/sec: 4782.72 - lr: 0.000023 - momentum: 0.000000
2023-10-25 20:48:48,373 ----------------------------------------------------------------------------------------------------
2023-10-25 20:48:48,373 EPOCH 6 done: loss 0.0229 - lr: 0.000023
2023-10-25 20:48:49,285 DEV : loss 0.1573459953069687 - f1-score (micro avg) 0.7339
2023-10-25 20:48:49,290 ----------------------------------------------------------------------------------------------------
2023-10-25 20:48:50,319 epoch 7 - iter 14/146 - loss 0.02023463 - time (sec): 1.03 - samples/sec: 3920.05 - lr: 0.000022 - momentum: 0.000000
2023-10-25 20:48:51,207 epoch 7 - iter 28/146 - loss 0.01645952 - time (sec): 1.92 - samples/sec: 4177.15 - lr: 0.000021 - momentum: 0.000000
2023-10-25 20:48:52,071 epoch 7 - iter 42/146 - loss 0.01621473 - time (sec): 2.78 - samples/sec: 4372.48 - lr: 0.000021 - momentum: 0.000000
2023-10-25 20:48:52,893 epoch 7 - iter 56/146 - loss 0.01498802 - time (sec): 3.60 - samples/sec: 4437.54 - lr: 0.000020 - momentum: 0.000000
2023-10-25 20:48:54,004 epoch 7 - iter 70/146 - loss 0.01718484 - time (sec): 4.71 - samples/sec: 4498.40 - lr: 0.000020 - momentum: 0.000000
2023-10-25 20:48:55,044 epoch 7 - iter 84/146 - loss 0.01586359 - time (sec): 5.75 - samples/sec: 4509.60 - lr: 0.000019 - momentum: 0.000000
2023-10-25 20:48:55,972 epoch 7 - iter 98/146 - loss 0.01684853 - time (sec): 6.68 - samples/sec: 4510.54 - lr: 0.000019 - momentum: 0.000000
2023-10-25 20:48:56,896 epoch 7 - iter 112/146 - loss 0.01618138 - time (sec): 7.61 - samples/sec: 4548.68 - lr: 0.000018 - momentum: 0.000000
2023-10-25 20:48:57,791 epoch 7 - iter 126/146 - loss 0.01618698 - time (sec): 8.50 - samples/sec: 4505.04 - lr: 0.000018 - momentum: 0.000000
2023-10-25 20:48:58,621 epoch 7 - iter 140/146 - loss 0.01703696 - time (sec): 9.33 - samples/sec: 4586.23 - lr: 0.000017 - momentum: 0.000000
2023-10-25 20:48:58,968 ----------------------------------------------------------------------------------------------------
2023-10-25 20:48:58,968 EPOCH 7 done: loss 0.0171 - lr: 0.000017
2023-10-25 20:48:59,889 DEV : loss 0.16221855580806732 - f1-score (micro avg) 0.7226
2023-10-25 20:48:59,893 ----------------------------------------------------------------------------------------------------
2023-10-25 20:49:00,821 epoch 8 - iter 14/146 - loss 0.00624075 - time (sec): 0.93 - samples/sec: 4750.92 - lr: 0.000016 - momentum: 0.000000
2023-10-25 20:49:01,689 epoch 8 - iter 28/146 - loss 0.00947817 - time (sec): 1.79 - samples/sec: 5225.63 - lr: 0.000016 - momentum: 0.000000
2023-10-25 20:49:02,544 epoch 8 - iter 42/146 - loss 0.00932840 - time (sec): 2.65 - samples/sec: 5019.16 - lr: 0.000015 - momentum: 0.000000
2023-10-25 20:49:03,436 epoch 8 - iter 56/146 - loss 0.00899173 - time (sec): 3.54 - samples/sec: 4972.49 - lr: 0.000015 - momentum: 0.000000
2023-10-25 20:49:04,307 epoch 8 - iter 70/146 - loss 0.00929944 - time (sec): 4.41 - samples/sec: 4876.07 - lr: 0.000014 - momentum: 0.000000
2023-10-25 20:49:05,239 epoch 8 - iter 84/146 - loss 0.01037046 - time (sec): 5.34 - samples/sec: 4857.64 - lr: 0.000014 - momentum: 0.000000
2023-10-25 20:49:06,385 epoch 8 - iter 98/146 - loss 0.01144120 - time (sec): 6.49 - samples/sec: 4718.27 - lr: 0.000013 - momentum: 0.000000
2023-10-25 20:49:07,293 epoch 8 - iter 112/146 - loss 0.01108653 - time (sec): 7.40 - samples/sec: 4721.18 - lr: 0.000013 - momentum: 0.000000
2023-10-25 20:49:08,200 epoch 8 - iter 126/146 - loss 0.01315505 - time (sec): 8.31 - samples/sec: 4639.70 - lr: 0.000012 - momentum: 0.000000
2023-10-25 20:49:09,092 epoch 8 - iter 140/146 - loss 0.01296013 - time (sec): 9.20 - samples/sec: 4659.37 - lr: 0.000012 - momentum: 0.000000
2023-10-25 20:49:09,419 ----------------------------------------------------------------------------------------------------
2023-10-25 20:49:09,419 EPOCH 8 done: loss 0.0131 - lr: 0.000012
2023-10-25 20:49:10,337 DEV : loss 0.17299844324588776 - f1-score (micro avg) 0.7257
2023-10-25 20:49:10,342 ----------------------------------------------------------------------------------------------------
2023-10-25 20:49:11,325 epoch 9 - iter 14/146 - loss 0.01830680 - time (sec): 0.98 - samples/sec: 4488.91 - lr: 0.000011 - momentum: 0.000000
2023-10-25 20:49:12,187 epoch 9 - iter 28/146 - loss 0.01109420 - time (sec): 1.84 - samples/sec: 4547.76 - lr: 0.000010 - momentum: 0.000000
2023-10-25 20:49:12,956 epoch 9 - iter 42/146 - loss 0.01001611 - time (sec): 2.61 - samples/sec: 4550.41 - lr: 0.000010 - momentum: 0.000000
2023-10-25 20:49:13,857 epoch 9 - iter 56/146 - loss 0.00902443 - time (sec): 3.51 - samples/sec: 4624.04 - lr: 0.000009 - momentum: 0.000000
2023-10-25 20:49:14,691 epoch 9 - iter 70/146 - loss 0.00932559 - time (sec): 4.35 - samples/sec: 4630.51 - lr: 0.000009 - momentum: 0.000000
2023-10-25 20:49:15,596 epoch 9 - iter 84/146 - loss 0.00988126 - time (sec): 5.25 - samples/sec: 4582.74 - lr: 0.000008 - momentum: 0.000000
2023-10-25 20:49:16,492 epoch 9 - iter 98/146 - loss 0.00934970 - time (sec): 6.15 - samples/sec: 4703.10 - lr: 0.000008 - momentum: 0.000000
2023-10-25 20:49:17,401 epoch 9 - iter 112/146 - loss 0.00834335 - time (sec): 7.06 - samples/sec: 4723.11 - lr: 0.000007 - momentum: 0.000000
2023-10-25 20:49:18,388 epoch 9 - iter 126/146 - loss 0.00786094 - time (sec): 8.04 - samples/sec: 4757.90 - lr: 0.000007 - momentum: 0.000000
2023-10-25 20:49:19,410 epoch 9 - iter 140/146 - loss 0.00807518 - time (sec): 9.07 - samples/sec: 4739.93 - lr: 0.000006 - momentum: 0.000000
2023-10-25 20:49:19,762 ----------------------------------------------------------------------------------------------------
2023-10-25 20:49:19,762 EPOCH 9 done: loss 0.0082 - lr: 0.000006
2023-10-25 20:49:20,679 DEV : loss 0.19021421670913696 - f1-score (micro avg) 0.7032
2023-10-25 20:49:20,684 ----------------------------------------------------------------------------------------------------
2023-10-25 20:49:21,526 epoch 10 - iter 14/146 - loss 0.00264529 - time (sec): 0.84 - samples/sec: 5002.29 - lr: 0.000005 - momentum: 0.000000
2023-10-25 20:49:22,516 epoch 10 - iter 28/146 - loss 0.00715687 - time (sec): 1.83 - samples/sec: 4562.72 - lr: 0.000005 - momentum: 0.000000
2023-10-25 20:49:23,371 epoch 10 - iter 42/146 - loss 0.00756824 - time (sec): 2.69 - samples/sec: 4594.76 - lr: 0.000004 - momentum: 0.000000
2023-10-25 20:49:24,269 epoch 10 - iter 56/146 - loss 0.00627773 - time (sec): 3.58 - samples/sec: 4685.83 - lr: 0.000004 - momentum: 0.000000
2023-10-25 20:49:25,223 epoch 10 - iter 70/146 - loss 0.00727587 - time (sec): 4.54 - samples/sec: 4629.20 - lr: 0.000003 - momentum: 0.000000
2023-10-25 20:49:26,052 epoch 10 - iter 84/146 - loss 0.00623229 - time (sec): 5.37 - samples/sec: 4579.56 - lr: 0.000003 - momentum: 0.000000
2023-10-25 20:49:27,230 epoch 10 - iter 98/146 - loss 0.00645332 - time (sec): 6.54 - samples/sec: 4523.53 - lr: 0.000002 - momentum: 0.000000
2023-10-25 20:49:28,185 epoch 10 - iter 112/146 - loss 0.00637249 - time (sec): 7.50 - samples/sec: 4586.34 - lr: 0.000002 - momentum: 0.000000
2023-10-25 20:49:29,154 epoch 10 - iter 126/146 - loss 0.00621133 - time (sec): 8.47 - samples/sec: 4552.42 - lr: 0.000001 - momentum: 0.000000
2023-10-25 20:49:29,992 epoch 10 - iter 140/146 - loss 0.00573965 - time (sec): 9.31 - samples/sec: 4551.44 - lr: 0.000000 - momentum: 0.000000
2023-10-25 20:49:30,393 ----------------------------------------------------------------------------------------------------
2023-10-25 20:49:30,393 EPOCH 10 done: loss 0.0055 - lr: 0.000000
2023-10-25 20:49:31,304 DEV : loss 0.19148127734661102 - f1-score (micro avg) 0.6975
2023-10-25 20:49:31,833 ----------------------------------------------------------------------------------------------------
2023-10-25 20:49:31,834 Loading model from best epoch ...
2023-10-25 20:49:33,617 SequenceTagger predicts: Dictionary with 17 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
2023-10-25 20:49:35,165
Results:
- F-score (micro) 0.754
- F-score (macro) 0.67
- Accuracy 0.6246
By class:
precision recall f1-score support
PER 0.7983 0.8305 0.8141 348
LOC 0.6616 0.8391 0.7399 261
ORG 0.4565 0.4038 0.4286 52
HumanProd 0.7143 0.6818 0.6977 22
micro avg 0.7158 0.7965 0.7540 683
macro avg 0.6577 0.6888 0.6700 683
weighted avg 0.7174 0.7965 0.7526 683
2023-10-25 20:49:35,166 ----------------------------------------------------------------------------------------------------
|