discrimintools.datasets.load_infidelity

discrimintools.datasets.load_infidelity(element='train')

Infidelity dataset

Parameters:

element (str, default = ‘train’) – The dataset to load. Possible values are:

• ‘train’ for training dataset.

• ‘test’ for testing dataset.

Returns:

infidelity – The infidelity dataset.

Return type:

DataFrame of shape (n_samples, n_columns)

References

[1] Ricco Rakotomalala (2020), « Régression logistique sous Python », Université Lumière Lyon 2.

Examples

>>> from discrimintools.datasets import load_infidelity
>>> from discrimintools import DISCRIM
>>> D = load_infidelity("train") # load training data
>>> y, X = D["Infidelity"], D.drop(columns=["Infidelity"]) # split into X and y
>>> clf = DISCRIM()
>>> clf.fit(X,y)
DISCRIM(priors='prop')