libuplift.base

Base classes for all estimators and trasformers.

Classes

UpliftRegressorMixin	Mixin class for all uplift regression estimators in
UpliftClassifierMixin	Mixin class for all uplift classification estimators in
UpliftTransformerMixin

Functions

is_uplift(estimator)

Returns True if the given estimator is an uplift model.

Module Contents

class libuplift.base.UpliftRegressorMixin

Bases: _BaseUpliftMixin, sklearn.base.RegressorMixin

Mixin class for all uplift regression estimators in libuplift.

score(X, y, trt, n_trt=None, sample_weight=None)

Score test data.

By default difference between model predicted and sample ATE is returned (e_sate).

class libuplift.base.UpliftClassifierMixin

Bases: _BaseUpliftMixin, sklearn.base.ClassifierMixin

Mixin class for all uplift classification estimators in libuplift.

predict_action(X, pos_label=None)

Predict most beneficial action.

Only supported for binary classification or when pos_label is set. pos_label must be an intereger between 0 and self.n_classes_-1.

score(X, y, trt, n_trt=None, sample_weight=None)

Score test data.

By default difference between model predicted and sample ATE is returned (e_sate).

class libuplift.base.UpliftTransformerMixin

Bases: object

fit_transform(X, y=None, trt=None, n_trt=None, **fit_params)

libuplift.base.is_uplift(estimator)

Returns True if the given estimator is an uplift model.

Parameters:

estimatorobject

Estimator object to test.

Returns:

outbool

True if estimator is an uplift model and False otherwise.