Download "epsilon" [1] data set. Will return two pandas.DataFrame-s, first with train part (epsilon_normalized) and second with test part (epsilon_normalized.t) of the dataset. Object class will be located in the first column of dataset. NOTE: This is a preprocessed version of
()
| 197 | |
| 198 | |
| 199 | def epsilon(): |
| 200 | """ |
| 201 | Download "epsilon" [1] data set. |
| 202 | |
| 203 | Will return two pandas.DataFrame-s, first with train part (epsilon_normalized) and second with |
| 204 | test part (epsilon_normalized.t) of the dataset. Object class will be located in the first |
| 205 | column of dataset. |
| 206 | |
| 207 | NOTE: This is a preprocessed version of the dataset. It was converted from libsvm format into |
| 208 | tsv (CatBoost doesn't support libsvm format out of the box). |
| 209 | |
| 210 | [1]: https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/binary.html#epsilon |
| 211 | """ |
| 212 | urls = ( |
| 213 | 'https://proxy.sandbox.yandex-team.ru/785711439', |
| 214 | 'https://storage.mds.yandex.net/get-devtools-opensource/250854/epsilon.tar.gz', ) |
| 215 | md5 = '5bbfac403ac673da7d7ee84bd532e973' |
| 216 | dataset_name, train_file, test_file = 'epsilon', 'train.tsv', 'test.tsv' |
| 217 | train_path, test_path = _download_dataset(urls, md5, dataset_name, train_file, test_file, cache=True) |
| 218 | return ( |
| 219 | _load_numeric_only_dataset(train_path, 400000, 2001, sep='\t'), |
| 220 | _load_numeric_only_dataset(test_path, 100000, 2001, sep='\t')) |
| 221 | |
| 222 | |
| 223 | def monotonic1(): |