Classes

class BoostedRulesClassifier (estimator=DecisionTreeClassifier(max_depth=1), *, n_estimators=15, learning_rate=1.0, random_state=None)

An easy-interpretable classifier optimizing simple logical rules.

Parameters

estimator : object with fit and predict methods

Defaults to DecisionTreeClassifier with AdaBoost. For SLIPPER, should pass estimator=imodels.SlipperBaseEstimator

Expand source code

class BoostedRulesClassifier(RuleInspectionMixin, AdaBoostClassifier):
    '''An easy-interpretable classifier optimizing simple logical rules.

    Parameters
    ----------
    estimator: object with fit and predict methods
        Defaults to DecisionTreeClassifier with AdaBoost.
        For SLIPPER, should pass estimator=imodels.SlipperBaseEstimator
    '''

    def __init__(
        self,
        estimator=DecisionTreeClassifier(max_depth=1),
        *,
        n_estimators=15,
        learning_rate=1.0,
        random_state=None,
    ):
        try: # sklearn version >= 1.2
            super().__init__(
                estimator=estimator,
                n_estimators=n_estimators,
                learning_rate=learning_rate,
                random_state=random_state,
            )
        except: # sklearn version < 1.2
            super().__init__(
                base_estimator=estimator,
                n_estimators=n_estimators,
                learning_rate=learning_rate,
                random_state=random_state,
            )
            self.estimator = estimator



    def fit(self, X, y, feature_names=None, **kwargs):
        X, y, feature_names = check_fit_arguments(self, X, y, feature_names)
        classes = self.classes_  # super().fit overwrites this with the encoded labels
        names_in = getattr(self, 'feature_names_in_', None)
        super().fit(X, y, **kwargs)
        self.classes_ = classes
        if names_in is not None:  # super().fit strips this when passed a plain array
            self.feature_names_in_ = names_in
        self.complexity_ = len(self.estimators_)
        return self

Ancestors

• RuleInspectionMixin
• RulesMixin
• TextMixin
• LeavesMixin
• sklearn.ensemble._weight_boosting.AdaBoostClassifier
• sklearn.utils._metadata_requests._RoutingNotSupportedMixin
• sklearn.base.ClassifierMixin
• sklearn.ensemble._weight_boosting.BaseWeightBoosting
• sklearn.ensemble._base.BaseEnsemble
• sklearn.base.MetaEstimatorMixin
• sklearn.base.BaseEstimator
• sklearn.utils._estimator_html_repr._HTMLDocumentationLinkMixin
• sklearn.utils._metadata_requests._MetadataRequester

Subclasses

• SlipperClassifier

Methods

def fit(self, X, y, feature_names=None, **kwargs)

Build a boosted classifier/regressor from the training set (X, y).

Parameters

X : {array-like, sparse matrix} of shape (n_samples, n_features)

The training input samples. Sparse matrix can be CSC, CSR, COO, DOK, or LIL. COO, DOK, and LIL are converted to CSR.

y : array-like of shape (n_samples,)

The target values.

sample_weight : array-like of shape (n_samples,), default=None

Sample weights. If None, the sample weights are initialized to 1 / n_samples.

Returns

self : object

Fitted estimator.

Expand source code

def fit(self, X, y, feature_names=None, **kwargs):
    X, y, feature_names = check_fit_arguments(self, X, y, feature_names)
    classes = self.classes_  # super().fit overwrites this with the encoded labels
    names_in = getattr(self, 'feature_names_in_', None)
    super().fit(X, y, **kwargs)
    self.classes_ = classes
    if names_in is not None:  # super().fit strips this when passed a plain array
        self.feature_names_in_ = names_in
    self.complexity_ = len(self.estimators_)
    return self

Inherited members

• RuleInspectionMixin:
  • apply
  • get_rules

class BoostedRulesRegressor (estimator=DecisionTreeRegressor(max_depth=1), *, n_estimators=15, learning_rate=1.0, random_state=13)

An easy-interpretable regressor optimizing simple logical rules.

Parameters

estimator : object with fit and predict methods

Defaults to DecisionTreeRegressor with AdaBoost.

Expand source code

class BoostedRulesRegressor(RuleInspectionMixin, AdaBoostRegressor):
    '''An easy-interpretable regressor optimizing simple logical rules.

    Parameters
    ----------
    estimator: object with fit and predict methods
        Defaults to DecisionTreeRegressor with AdaBoost.
    '''

    def __init__(
        self,
        estimator=DecisionTreeRegressor(max_depth=1),
        *,
        n_estimators=15,
        learning_rate=1.0,
        random_state=13,
    ):
        try: # sklearn version >= 1.2
            super().__init__(
                estimator=estimator,
                n_estimators=n_estimators,
                learning_rate=learning_rate,
                random_state=random_state,
            )
        except: # sklearn version < 1.2
            super().__init__(
                base_estimator=estimator,
                n_estimators=n_estimators,
                learning_rate=learning_rate,
                random_state=random_state,
            )
            self.estimator = estimator


    def fit(self, X, y, feature_names=None, **kwargs):
        X, y, feature_names = check_fit_arguments(self, X, y, feature_names)
        names_in = getattr(self, 'feature_names_in_', None)
        super().fit(X, y, **kwargs)
        if names_in is not None:  # super().fit strips this when passed a plain array
            self.feature_names_in_ = names_in
        self.complexity_ = len(self.estimators_)
        return self

Ancestors

• RuleInspectionMixin
• RulesMixin
• TextMixin
• LeavesMixin
• sklearn.ensemble._weight_boosting.AdaBoostRegressor
• sklearn.utils._metadata_requests._RoutingNotSupportedMixin
• sklearn.base.RegressorMixin
• sklearn.ensemble._weight_boosting.BaseWeightBoosting
• sklearn.ensemble._base.BaseEnsemble
• sklearn.base.MetaEstimatorMixin
• sklearn.base.BaseEstimator
• sklearn.utils._estimator_html_repr._HTMLDocumentationLinkMixin
• sklearn.utils._metadata_requests._MetadataRequester

Methods

def fit(self, X, y, feature_names=None, **kwargs)

Build a boosted classifier/regressor from the training set (X, y).

Parameters

X : {array-like, sparse matrix} of shape (n_samples, n_features)

The training input samples. Sparse matrix can be CSC, CSR, COO, DOK, or LIL. COO, DOK, and LIL are converted to CSR.

y : array-like of shape (n_samples,)

The target values.

sample_weight : array-like of shape (n_samples,), default=None

Sample weights. If None, the sample weights are initialized to 1 / n_samples.

Returns

self : object

Fitted estimator.

Expand source code

def fit(self, X, y, feature_names=None, **kwargs):
    X, y, feature_names = check_fit_arguments(self, X, y, feature_names)
    names_in = getattr(self, 'feature_names_in_', None)
    super().fit(X, y, **kwargs)
    if names_in is not None:  # super().fit strips this when passed a plain array
        self.feature_names_in_ = names_in
    self.complexity_ = len(self.estimators_)
    return self

Inherited members

• RuleInspectionMixin:
  • apply
  • get_rules