MCPcopy Create free account
hub / github.com/catboost/catboost / epsilon

Function epsilon

catboost/python-package/catboost/datasets.py:199–220  ·  view source on GitHub ↗

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

()

Source from the content-addressed store, hash-verified

197
198
199def 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
223def monotonic1():

Callers 15

MannWhitneyWithSignFunction · 0.50
TTestFunction · 0.50
KLDivergenceFunction · 0.50
RelativeEqualFunction · 0.50
DefaultConstructTestMethod · 0.50
MakeActionFunction · 0.50
CheckInvariantsFunction · 0.50
ValidateMethod · 0.50
SolveNewtonEquationMethod · 0.50
ComputeDCGPosWeightsMethod · 0.50

Calls 2

_download_datasetFunction · 0.85

Tested by 1

TestFloatLimitsFunction · 0.40