File size: 24,030 Bytes
ee09f3a |
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 |
2023-10-16 21:44:09,762 ----------------------------------------------------------------------------------------------------
2023-10-16 21:44:09,763 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=13, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-16 21:44:09,763 ----------------------------------------------------------------------------------------------------
2023-10-16 21:44:09,764 MultiCorpus: 6183 train + 680 dev + 2113 test sentences
- NER_HIPE_2022 Corpus: 6183 train + 680 dev + 2113 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/topres19th/en/with_doc_seperator
2023-10-16 21:44:09,764 ----------------------------------------------------------------------------------------------------
2023-10-16 21:44:09,764 Train: 6183 sentences
2023-10-16 21:44:09,764 (train_with_dev=False, train_with_test=False)
2023-10-16 21:44:09,764 ----------------------------------------------------------------------------------------------------
2023-10-16 21:44:09,764 Training Params:
2023-10-16 21:44:09,764 - learning_rate: "5e-05"
2023-10-16 21:44:09,764 - mini_batch_size: "4"
2023-10-16 21:44:09,764 - max_epochs: "10"
2023-10-16 21:44:09,764 - shuffle: "True"
2023-10-16 21:44:09,764 ----------------------------------------------------------------------------------------------------
2023-10-16 21:44:09,764 Plugins:
2023-10-16 21:44:09,764 - LinearScheduler | warmup_fraction: '0.1'
2023-10-16 21:44:09,764 ----------------------------------------------------------------------------------------------------
2023-10-16 21:44:09,764 Final evaluation on model from best epoch (best-model.pt)
2023-10-16 21:44:09,764 - metric: "('micro avg', 'f1-score')"
2023-10-16 21:44:09,764 ----------------------------------------------------------------------------------------------------
2023-10-16 21:44:09,764 Computation:
2023-10-16 21:44:09,764 - compute on device: cuda:0
2023-10-16 21:44:09,764 - embedding storage: none
2023-10-16 21:44:09,764 ----------------------------------------------------------------------------------------------------
2023-10-16 21:44:09,764 Model training base path: "hmbench-topres19th/en-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-16 21:44:09,764 ----------------------------------------------------------------------------------------------------
2023-10-16 21:44:09,764 ----------------------------------------------------------------------------------------------------
2023-10-16 21:44:16,668 epoch 1 - iter 154/1546 - loss 1.65007029 - time (sec): 6.90 - samples/sec: 1715.81 - lr: 0.000005 - momentum: 0.000000
2023-10-16 21:44:23,611 epoch 1 - iter 308/1546 - loss 0.94129662 - time (sec): 13.85 - samples/sec: 1714.44 - lr: 0.000010 - momentum: 0.000000
2023-10-16 21:44:30,625 epoch 1 - iter 462/1546 - loss 0.64763640 - time (sec): 20.86 - samples/sec: 1777.67 - lr: 0.000015 - momentum: 0.000000
2023-10-16 21:44:37,494 epoch 1 - iter 616/1546 - loss 0.52517050 - time (sec): 27.73 - samples/sec: 1772.61 - lr: 0.000020 - momentum: 0.000000
2023-10-16 21:44:44,367 epoch 1 - iter 770/1546 - loss 0.44714948 - time (sec): 34.60 - samples/sec: 1761.60 - lr: 0.000025 - momentum: 0.000000
2023-10-16 21:44:51,242 epoch 1 - iter 924/1546 - loss 0.39140356 - time (sec): 41.48 - samples/sec: 1762.58 - lr: 0.000030 - momentum: 0.000000
2023-10-16 21:44:58,147 epoch 1 - iter 1078/1546 - loss 0.35053117 - time (sec): 48.38 - samples/sec: 1776.97 - lr: 0.000035 - momentum: 0.000000
2023-10-16 21:45:05,162 epoch 1 - iter 1232/1546 - loss 0.32273432 - time (sec): 55.40 - samples/sec: 1793.05 - lr: 0.000040 - momentum: 0.000000
2023-10-16 21:45:12,237 epoch 1 - iter 1386/1546 - loss 0.29966657 - time (sec): 62.47 - samples/sec: 1779.42 - lr: 0.000045 - momentum: 0.000000
2023-10-16 21:45:19,081 epoch 1 - iter 1540/1546 - loss 0.28039720 - time (sec): 69.32 - samples/sec: 1787.73 - lr: 0.000050 - momentum: 0.000000
2023-10-16 21:45:19,336 ----------------------------------------------------------------------------------------------------
2023-10-16 21:45:19,336 EPOCH 1 done: loss 0.2799 - lr: 0.000050
2023-10-16 21:45:21,072 DEV : loss 0.0895804762840271 - f1-score (micro avg) 0.637
2023-10-16 21:45:21,095 saving best model
2023-10-16 21:45:21,441 ----------------------------------------------------------------------------------------------------
2023-10-16 21:45:28,278 epoch 2 - iter 154/1546 - loss 0.09243588 - time (sec): 6.84 - samples/sec: 1772.05 - lr: 0.000049 - momentum: 0.000000
2023-10-16 21:45:35,087 epoch 2 - iter 308/1546 - loss 0.10563878 - time (sec): 13.65 - samples/sec: 1755.06 - lr: 0.000049 - momentum: 0.000000
2023-10-16 21:45:42,042 epoch 2 - iter 462/1546 - loss 0.10498928 - time (sec): 20.60 - samples/sec: 1786.65 - lr: 0.000048 - momentum: 0.000000
2023-10-16 21:45:48,905 epoch 2 - iter 616/1546 - loss 0.10253956 - time (sec): 27.46 - samples/sec: 1779.24 - lr: 0.000048 - momentum: 0.000000
2023-10-16 21:45:55,770 epoch 2 - iter 770/1546 - loss 0.10307917 - time (sec): 34.33 - samples/sec: 1793.99 - lr: 0.000047 - momentum: 0.000000
2023-10-16 21:46:02,644 epoch 2 - iter 924/1546 - loss 0.10366383 - time (sec): 41.20 - samples/sec: 1775.94 - lr: 0.000047 - momentum: 0.000000
2023-10-16 21:46:09,503 epoch 2 - iter 1078/1546 - loss 0.10275984 - time (sec): 48.06 - samples/sec: 1780.26 - lr: 0.000046 - momentum: 0.000000
2023-10-16 21:46:16,416 epoch 2 - iter 1232/1546 - loss 0.10145230 - time (sec): 54.97 - samples/sec: 1779.20 - lr: 0.000046 - momentum: 0.000000
2023-10-16 21:46:23,230 epoch 2 - iter 1386/1546 - loss 0.09850958 - time (sec): 61.79 - samples/sec: 1784.50 - lr: 0.000045 - momentum: 0.000000
2023-10-16 21:46:30,555 epoch 2 - iter 1540/1546 - loss 0.09639554 - time (sec): 69.11 - samples/sec: 1791.82 - lr: 0.000044 - momentum: 0.000000
2023-10-16 21:46:30,812 ----------------------------------------------------------------------------------------------------
2023-10-16 21:46:30,812 EPOCH 2 done: loss 0.0961 - lr: 0.000044
2023-10-16 21:46:32,814 DEV : loss 0.0685592070221901 - f1-score (micro avg) 0.7696
2023-10-16 21:46:32,827 saving best model
2023-10-16 21:46:33,306 ----------------------------------------------------------------------------------------------------
2023-10-16 21:46:40,223 epoch 3 - iter 154/1546 - loss 0.06577964 - time (sec): 6.91 - samples/sec: 1853.17 - lr: 0.000044 - momentum: 0.000000
2023-10-16 21:46:46,999 epoch 3 - iter 308/1546 - loss 0.06885278 - time (sec): 13.69 - samples/sec: 1832.42 - lr: 0.000043 - momentum: 0.000000
2023-10-16 21:46:53,930 epoch 3 - iter 462/1546 - loss 0.07613517 - time (sec): 20.62 - samples/sec: 1828.52 - lr: 0.000043 - momentum: 0.000000
2023-10-16 21:47:00,686 epoch 3 - iter 616/1546 - loss 0.07339670 - time (sec): 27.38 - samples/sec: 1822.81 - lr: 0.000042 - momentum: 0.000000
2023-10-16 21:47:07,536 epoch 3 - iter 770/1546 - loss 0.07279666 - time (sec): 34.23 - samples/sec: 1827.46 - lr: 0.000042 - momentum: 0.000000
2023-10-16 21:47:14,494 epoch 3 - iter 924/1546 - loss 0.07159335 - time (sec): 41.19 - samples/sec: 1818.28 - lr: 0.000041 - momentum: 0.000000
2023-10-16 21:47:21,360 epoch 3 - iter 1078/1546 - loss 0.06968918 - time (sec): 48.05 - samples/sec: 1814.87 - lr: 0.000041 - momentum: 0.000000
2023-10-16 21:47:28,153 epoch 3 - iter 1232/1546 - loss 0.06933368 - time (sec): 54.84 - samples/sec: 1802.51 - lr: 0.000040 - momentum: 0.000000
2023-10-16 21:47:35,020 epoch 3 - iter 1386/1546 - loss 0.06859729 - time (sec): 61.71 - samples/sec: 1812.58 - lr: 0.000039 - momentum: 0.000000
2023-10-16 21:47:41,777 epoch 3 - iter 1540/1546 - loss 0.06849329 - time (sec): 68.47 - samples/sec: 1810.30 - lr: 0.000039 - momentum: 0.000000
2023-10-16 21:47:42,035 ----------------------------------------------------------------------------------------------------
2023-10-16 21:47:42,035 EPOCH 3 done: loss 0.0685 - lr: 0.000039
2023-10-16 21:47:44,051 DEV : loss 0.08517798036336899 - f1-score (micro avg) 0.7447
2023-10-16 21:47:44,063 ----------------------------------------------------------------------------------------------------
2023-10-16 21:47:51,034 epoch 4 - iter 154/1546 - loss 0.05538366 - time (sec): 6.97 - samples/sec: 1785.10 - lr: 0.000038 - momentum: 0.000000
2023-10-16 21:47:57,870 epoch 4 - iter 308/1546 - loss 0.05887323 - time (sec): 13.81 - samples/sec: 1724.84 - lr: 0.000038 - momentum: 0.000000
2023-10-16 21:48:04,737 epoch 4 - iter 462/1546 - loss 0.05491630 - time (sec): 20.67 - samples/sec: 1741.61 - lr: 0.000037 - momentum: 0.000000
2023-10-16 21:48:11,492 epoch 4 - iter 616/1546 - loss 0.05243751 - time (sec): 27.43 - samples/sec: 1758.25 - lr: 0.000037 - momentum: 0.000000
2023-10-16 21:48:18,347 epoch 4 - iter 770/1546 - loss 0.05291705 - time (sec): 34.28 - samples/sec: 1778.29 - lr: 0.000036 - momentum: 0.000000
2023-10-16 21:48:25,115 epoch 4 - iter 924/1546 - loss 0.05316064 - time (sec): 41.05 - samples/sec: 1782.75 - lr: 0.000036 - momentum: 0.000000
2023-10-16 21:48:31,936 epoch 4 - iter 1078/1546 - loss 0.05226355 - time (sec): 47.87 - samples/sec: 1787.50 - lr: 0.000035 - momentum: 0.000000
2023-10-16 21:48:38,834 epoch 4 - iter 1232/1546 - loss 0.05147292 - time (sec): 54.77 - samples/sec: 1793.52 - lr: 0.000034 - momentum: 0.000000
2023-10-16 21:48:45,749 epoch 4 - iter 1386/1546 - loss 0.04971427 - time (sec): 61.68 - samples/sec: 1797.52 - lr: 0.000034 - momentum: 0.000000
2023-10-16 21:48:52,720 epoch 4 - iter 1540/1546 - loss 0.04975106 - time (sec): 68.66 - samples/sec: 1804.48 - lr: 0.000033 - momentum: 0.000000
2023-10-16 21:48:52,978 ----------------------------------------------------------------------------------------------------
2023-10-16 21:48:52,978 EPOCH 4 done: loss 0.0497 - lr: 0.000033
2023-10-16 21:48:55,028 DEV : loss 0.09497705101966858 - f1-score (micro avg) 0.7821
2023-10-16 21:48:55,040 saving best model
2023-10-16 21:48:55,509 ----------------------------------------------------------------------------------------------------
2023-10-16 21:49:02,349 epoch 5 - iter 154/1546 - loss 0.01734854 - time (sec): 6.83 - samples/sec: 1839.45 - lr: 0.000033 - momentum: 0.000000
2023-10-16 21:49:09,103 epoch 5 - iter 308/1546 - loss 0.02591889 - time (sec): 13.58 - samples/sec: 1813.31 - lr: 0.000032 - momentum: 0.000000
2023-10-16 21:49:15,969 epoch 5 - iter 462/1546 - loss 0.03215375 - time (sec): 20.45 - samples/sec: 1762.51 - lr: 0.000032 - momentum: 0.000000
2023-10-16 21:49:22,877 epoch 5 - iter 616/1546 - loss 0.03561342 - time (sec): 27.36 - samples/sec: 1798.27 - lr: 0.000031 - momentum: 0.000000
2023-10-16 21:49:29,717 epoch 5 - iter 770/1546 - loss 0.04014613 - time (sec): 34.20 - samples/sec: 1807.31 - lr: 0.000031 - momentum: 0.000000
2023-10-16 21:49:36,641 epoch 5 - iter 924/1546 - loss 0.04102622 - time (sec): 41.12 - samples/sec: 1818.94 - lr: 0.000030 - momentum: 0.000000
2023-10-16 21:49:43,561 epoch 5 - iter 1078/1546 - loss 0.03999871 - time (sec): 48.04 - samples/sec: 1820.62 - lr: 0.000029 - momentum: 0.000000
2023-10-16 21:49:50,438 epoch 5 - iter 1232/1546 - loss 0.04053831 - time (sec): 54.92 - samples/sec: 1816.84 - lr: 0.000029 - momentum: 0.000000
2023-10-16 21:49:57,266 epoch 5 - iter 1386/1546 - loss 0.04007914 - time (sec): 61.75 - samples/sec: 1811.89 - lr: 0.000028 - momentum: 0.000000
2023-10-16 21:50:04,082 epoch 5 - iter 1540/1546 - loss 0.03801735 - time (sec): 68.56 - samples/sec: 1807.74 - lr: 0.000028 - momentum: 0.000000
2023-10-16 21:50:04,338 ----------------------------------------------------------------------------------------------------
2023-10-16 21:50:04,338 EPOCH 5 done: loss 0.0380 - lr: 0.000028
2023-10-16 21:50:06,356 DEV : loss 0.09318046271800995 - f1-score (micro avg) 0.7782
2023-10-16 21:50:06,368 ----------------------------------------------------------------------------------------------------
2023-10-16 21:50:13,194 epoch 6 - iter 154/1546 - loss 0.02556469 - time (sec): 6.82 - samples/sec: 1809.18 - lr: 0.000027 - momentum: 0.000000
2023-10-16 21:50:20,094 epoch 6 - iter 308/1546 - loss 0.03097773 - time (sec): 13.72 - samples/sec: 1736.11 - lr: 0.000027 - momentum: 0.000000
2023-10-16 21:50:26,958 epoch 6 - iter 462/1546 - loss 0.02965652 - time (sec): 20.59 - samples/sec: 1746.62 - lr: 0.000026 - momentum: 0.000000
2023-10-16 21:50:33,830 epoch 6 - iter 616/1546 - loss 0.02790282 - time (sec): 27.46 - samples/sec: 1771.83 - lr: 0.000026 - momentum: 0.000000
2023-10-16 21:50:40,624 epoch 6 - iter 770/1546 - loss 0.02749592 - time (sec): 34.25 - samples/sec: 1771.92 - lr: 0.000025 - momentum: 0.000000
2023-10-16 21:50:47,433 epoch 6 - iter 924/1546 - loss 0.02610977 - time (sec): 41.06 - samples/sec: 1768.73 - lr: 0.000024 - momentum: 0.000000
2023-10-16 21:50:54,284 epoch 6 - iter 1078/1546 - loss 0.02639060 - time (sec): 47.91 - samples/sec: 1772.33 - lr: 0.000024 - momentum: 0.000000
2023-10-16 21:51:01,306 epoch 6 - iter 1232/1546 - loss 0.02605220 - time (sec): 54.94 - samples/sec: 1777.47 - lr: 0.000023 - momentum: 0.000000
2023-10-16 21:51:08,170 epoch 6 - iter 1386/1546 - loss 0.02522412 - time (sec): 61.80 - samples/sec: 1771.19 - lr: 0.000023 - momentum: 0.000000
2023-10-16 21:51:15,235 epoch 6 - iter 1540/1546 - loss 0.02642731 - time (sec): 68.87 - samples/sec: 1795.68 - lr: 0.000022 - momentum: 0.000000
2023-10-16 21:51:15,502 ----------------------------------------------------------------------------------------------------
2023-10-16 21:51:15,502 EPOCH 6 done: loss 0.0266 - lr: 0.000022
2023-10-16 21:51:17,895 DEV : loss 0.10446853190660477 - f1-score (micro avg) 0.7795
2023-10-16 21:51:17,908 ----------------------------------------------------------------------------------------------------
2023-10-16 21:51:24,779 epoch 7 - iter 154/1546 - loss 0.02404472 - time (sec): 6.87 - samples/sec: 1839.81 - lr: 0.000022 - momentum: 0.000000
2023-10-16 21:51:31,664 epoch 7 - iter 308/1546 - loss 0.02056482 - time (sec): 13.75 - samples/sec: 1861.72 - lr: 0.000021 - momentum: 0.000000
2023-10-16 21:51:38,496 epoch 7 - iter 462/1546 - loss 0.02046697 - time (sec): 20.59 - samples/sec: 1840.24 - lr: 0.000021 - momentum: 0.000000
2023-10-16 21:51:45,451 epoch 7 - iter 616/1546 - loss 0.02166477 - time (sec): 27.54 - samples/sec: 1809.86 - lr: 0.000020 - momentum: 0.000000
2023-10-16 21:51:52,463 epoch 7 - iter 770/1546 - loss 0.02146336 - time (sec): 34.55 - samples/sec: 1805.13 - lr: 0.000019 - momentum: 0.000000
2023-10-16 21:51:59,383 epoch 7 - iter 924/1546 - loss 0.02253466 - time (sec): 41.47 - samples/sec: 1791.51 - lr: 0.000019 - momentum: 0.000000
2023-10-16 21:52:06,243 epoch 7 - iter 1078/1546 - loss 0.02170121 - time (sec): 48.33 - samples/sec: 1787.33 - lr: 0.000018 - momentum: 0.000000
2023-10-16 21:52:13,045 epoch 7 - iter 1232/1546 - loss 0.02115814 - time (sec): 55.14 - samples/sec: 1788.46 - lr: 0.000018 - momentum: 0.000000
2023-10-16 21:52:19,960 epoch 7 - iter 1386/1546 - loss 0.02075815 - time (sec): 62.05 - samples/sec: 1795.57 - lr: 0.000017 - momentum: 0.000000
2023-10-16 21:52:26,876 epoch 7 - iter 1540/1546 - loss 0.02048075 - time (sec): 68.97 - samples/sec: 1796.06 - lr: 0.000017 - momentum: 0.000000
2023-10-16 21:52:27,154 ----------------------------------------------------------------------------------------------------
2023-10-16 21:52:27,154 EPOCH 7 done: loss 0.0204 - lr: 0.000017
2023-10-16 21:52:29,150 DEV : loss 0.11760783195495605 - f1-score (micro avg) 0.7724
2023-10-16 21:52:29,163 ----------------------------------------------------------------------------------------------------
2023-10-16 21:52:36,119 epoch 8 - iter 154/1546 - loss 0.00904914 - time (sec): 6.96 - samples/sec: 1799.83 - lr: 0.000016 - momentum: 0.000000
2023-10-16 21:52:43,087 epoch 8 - iter 308/1546 - loss 0.01120019 - time (sec): 13.92 - samples/sec: 1821.61 - lr: 0.000016 - momentum: 0.000000
2023-10-16 21:52:49,903 epoch 8 - iter 462/1546 - loss 0.01332359 - time (sec): 20.74 - samples/sec: 1822.53 - lr: 0.000015 - momentum: 0.000000
2023-10-16 21:52:56,846 epoch 8 - iter 616/1546 - loss 0.01192538 - time (sec): 27.68 - samples/sec: 1842.43 - lr: 0.000014 - momentum: 0.000000
2023-10-16 21:53:03,643 epoch 8 - iter 770/1546 - loss 0.01078426 - time (sec): 34.48 - samples/sec: 1815.73 - lr: 0.000014 - momentum: 0.000000
2023-10-16 21:53:10,438 epoch 8 - iter 924/1546 - loss 0.01161029 - time (sec): 41.27 - samples/sec: 1813.84 - lr: 0.000013 - momentum: 0.000000
2023-10-16 21:53:17,145 epoch 8 - iter 1078/1546 - loss 0.01243620 - time (sec): 47.98 - samples/sec: 1810.28 - lr: 0.000013 - momentum: 0.000000
2023-10-16 21:53:23,967 epoch 8 - iter 1232/1546 - loss 0.01233856 - time (sec): 54.80 - samples/sec: 1804.05 - lr: 0.000012 - momentum: 0.000000
2023-10-16 21:53:30,776 epoch 8 - iter 1386/1546 - loss 0.01272431 - time (sec): 61.61 - samples/sec: 1800.38 - lr: 0.000012 - momentum: 0.000000
2023-10-16 21:53:37,834 epoch 8 - iter 1540/1546 - loss 0.01300626 - time (sec): 68.67 - samples/sec: 1804.05 - lr: 0.000011 - momentum: 0.000000
2023-10-16 21:53:38,096 ----------------------------------------------------------------------------------------------------
2023-10-16 21:53:38,096 EPOCH 8 done: loss 0.0130 - lr: 0.000011
2023-10-16 21:53:40,162 DEV : loss 0.11045785248279572 - f1-score (micro avg) 0.7898
2023-10-16 21:53:40,175 saving best model
2023-10-16 21:53:40,649 ----------------------------------------------------------------------------------------------------
2023-10-16 21:53:47,486 epoch 9 - iter 154/1546 - loss 0.00775423 - time (sec): 6.83 - samples/sec: 1831.86 - lr: 0.000011 - momentum: 0.000000
2023-10-16 21:53:54,019 epoch 9 - iter 308/1546 - loss 0.00790082 - time (sec): 13.36 - samples/sec: 1827.72 - lr: 0.000010 - momentum: 0.000000
2023-10-16 21:54:00,556 epoch 9 - iter 462/1546 - loss 0.00714745 - time (sec): 19.90 - samples/sec: 1837.52 - lr: 0.000009 - momentum: 0.000000
2023-10-16 21:54:07,065 epoch 9 - iter 616/1546 - loss 0.00761023 - time (sec): 26.41 - samples/sec: 1842.09 - lr: 0.000009 - momentum: 0.000000
2023-10-16 21:54:13,656 epoch 9 - iter 770/1546 - loss 0.00769417 - time (sec): 33.00 - samples/sec: 1841.10 - lr: 0.000008 - momentum: 0.000000
2023-10-16 21:54:20,344 epoch 9 - iter 924/1546 - loss 0.00778298 - time (sec): 39.69 - samples/sec: 1855.33 - lr: 0.000008 - momentum: 0.000000
2023-10-16 21:54:26,945 epoch 9 - iter 1078/1546 - loss 0.00794647 - time (sec): 46.29 - samples/sec: 1864.63 - lr: 0.000007 - momentum: 0.000000
2023-10-16 21:54:33,540 epoch 9 - iter 1232/1546 - loss 0.00813811 - time (sec): 52.89 - samples/sec: 1864.94 - lr: 0.000007 - momentum: 0.000000
2023-10-16 21:54:40,134 epoch 9 - iter 1386/1546 - loss 0.00823860 - time (sec): 59.48 - samples/sec: 1871.30 - lr: 0.000006 - momentum: 0.000000
2023-10-16 21:54:46,726 epoch 9 - iter 1540/1546 - loss 0.00873508 - time (sec): 66.07 - samples/sec: 1876.32 - lr: 0.000006 - momentum: 0.000000
2023-10-16 21:54:46,975 ----------------------------------------------------------------------------------------------------
2023-10-16 21:54:46,975 EPOCH 9 done: loss 0.0087 - lr: 0.000006
2023-10-16 21:54:48,998 DEV : loss 0.11864420771598816 - f1-score (micro avg) 0.7941
2023-10-16 21:54:49,011 saving best model
2023-10-16 21:54:49,452 ----------------------------------------------------------------------------------------------------
2023-10-16 21:54:56,289 epoch 10 - iter 154/1546 - loss 0.00586620 - time (sec): 6.83 - samples/sec: 1818.10 - lr: 0.000005 - momentum: 0.000000
2023-10-16 21:55:03,238 epoch 10 - iter 308/1546 - loss 0.00524542 - time (sec): 13.78 - samples/sec: 1808.72 - lr: 0.000004 - momentum: 0.000000
2023-10-16 21:55:10,143 epoch 10 - iter 462/1546 - loss 0.00563260 - time (sec): 20.69 - samples/sec: 1780.72 - lr: 0.000004 - momentum: 0.000000
2023-10-16 21:55:17,083 epoch 10 - iter 616/1546 - loss 0.00560671 - time (sec): 27.63 - samples/sec: 1810.14 - lr: 0.000003 - momentum: 0.000000
2023-10-16 21:55:23,981 epoch 10 - iter 770/1546 - loss 0.00551268 - time (sec): 34.53 - samples/sec: 1798.02 - lr: 0.000003 - momentum: 0.000000
2023-10-16 21:55:30,938 epoch 10 - iter 924/1546 - loss 0.00541465 - time (sec): 41.48 - samples/sec: 1806.94 - lr: 0.000002 - momentum: 0.000000
2023-10-16 21:55:37,821 epoch 10 - iter 1078/1546 - loss 0.00581212 - time (sec): 48.37 - samples/sec: 1798.64 - lr: 0.000002 - momentum: 0.000000
2023-10-16 21:55:44,637 epoch 10 - iter 1232/1546 - loss 0.00550965 - time (sec): 55.18 - samples/sec: 1802.49 - lr: 0.000001 - momentum: 0.000000
2023-10-16 21:55:51,547 epoch 10 - iter 1386/1546 - loss 0.00535177 - time (sec): 62.09 - samples/sec: 1793.62 - lr: 0.000001 - momentum: 0.000000
2023-10-16 21:55:58,452 epoch 10 - iter 1540/1546 - loss 0.00556052 - time (sec): 69.00 - samples/sec: 1791.86 - lr: 0.000000 - momentum: 0.000000
2023-10-16 21:55:58,718 ----------------------------------------------------------------------------------------------------
2023-10-16 21:55:58,719 EPOCH 10 done: loss 0.0055 - lr: 0.000000
2023-10-16 21:56:00,754 DEV : loss 0.1224876418709755 - f1-score (micro avg) 0.7966
2023-10-16 21:56:00,767 saving best model
2023-10-16 21:56:01,611 ----------------------------------------------------------------------------------------------------
2023-10-16 21:56:01,612 Loading model from best epoch ...
2023-10-16 21:56:03,412 SequenceTagger predicts: Dictionary with 13 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-BUILDING, B-BUILDING, E-BUILDING, I-BUILDING, S-STREET, B-STREET, E-STREET, I-STREET
2023-10-16 21:56:09,090
Results:
- F-score (micro) 0.8125
- F-score (macro) 0.7224
- Accuracy 0.7061
By class:
precision recall f1-score support
LOC 0.8716 0.8467 0.8590 946
BUILDING 0.6347 0.5730 0.6023 185
STREET 0.6667 0.7500 0.7059 56
micro avg 0.8259 0.7995 0.8125 1187
macro avg 0.7243 0.7232 0.7224 1187
weighted avg 0.8250 0.7995 0.8117 1187
2023-10-16 21:56:09,090 ----------------------------------------------------------------------------------------------------
|