特征转换器

class CategoricalSelector(BaseEstimator, TransformerMixin):
    """Meta-transformer for selecting categorical features and transform them using OneHotEncoder.

    Parameters
    ----------

    threshold : int, default=10
        Maximum number of unique values per feature to consider the feature
        to be categorical.

    minimum_fraction: float, default=None
        Minimum fraction of unique values in a feature to consider the feature
        to be categorical.
    """

    def __init__(self, threshold=10, minimum_fraction=None):
        """Create a CategoricalSelector object."""
        self.threshold = threshold
        self.minimum_fraction = minimum_fraction


    def fit(self, X, y=None):
        """Do nothing and return the estimator unchanged
        This method is just there to implement the usual API and hence
        work in pipelines.
        Parameters
        ----------
        X : array-like
        """
        X = check_array(X, accept_sparse='csr')
        return self


    def transform(self, X):
        """Select categorical features and transform them using OneHotEncoder.

        Parameters
        ----------
        X: numpy ndarray, {n_samples, n_components}
            New data, where n_samples is the number of samples and n_components is the number of components.

        Returns
        -------
        array-like, {n_samples, n_components}
        """
        selected = auto_select_categorical_features(X, threshold=self.threshold)
        X_sel, _, n_selected, _ = _X_selected(X, selected)

        if n_selected == 0:
            # No features selected.
            raise ValueError('No categorical feature was found!')
        else:
            ohe = OneHotEncoder(categorical_features='all', sparse=False, minimum_fraction=self.minimum_fraction)
            return ohe.fit_transform(X_sel)

class ContinuousSelector(BaseEstimator, TransformerMixin):
    """Meta-transformer for selecting continuous features and transform them using PCA.

    Parameters
    ----------

    threshold : int, default=10
        Maximum number of unique values per feature to consider the feature
        to be categorical.

    svd_solver : string {'auto', 'full', 'arpack', 'randomized'}
        auto :
            the solver is selected by a default policy based on `X.shape` and
            `n_components`: if the input data is larger than 500x500 and the
            number of components to extract is lower than 80% of the smallest
            dimension of the data, then the more efficient 'randomized'
            method is enabled. Otherwise the exact full SVD is computed and
            optionally truncated afterwards.
        full :
            run exact full SVD calling the standard LAPACK solver via
            `scipy.linalg.svd` and select the components by postprocessing
        arpack :
            run SVD truncated to n_components calling ARPACK solver via
            `scipy.sparse.linalg.svds`. It requires strictly
            0 < n_components < X.shape[1]
        randomized :
            run randomized SVD by the method of Halko et al.

    iterated_power : int >= 0, or 'auto', (default 'auto')
        Number of iterations for the power method computed by
        svd_solver == 'randomized'.

    """

    def __init__(self, threshold=10, svd_solver='randomized' ,iterated_power='auto', random_state=42):
        """Create a ContinuousSelector object."""
        self.threshold = threshold
        self.svd_solver = svd_solver
        self.iterated_power = iterated_power
        self.random_state = random_state


    def fit(self, X, y=None):
        """Do nothing and return the estimator unchanged
        This method is just there to implement the usual API and hence
        work in pipelines.
        Parameters
        ----------
        X : array-like
        """
        X = check_array(X)
        return self


    def transform(self, X):
        """Select continuous features and transform them using PCA.

        Parameters
        ----------
        X: numpy ndarray, {n_samples, n_components}
            New data, where n_samples is the number of samples and n_components is the number of components.

        Returns
        -------
        array-like, {n_samples, n_components}
        """
        selected = auto_select_categorical_features(X, threshold=self.threshold)
        _, X_sel, n_selected, _ = _X_selected(X, selected)

        if n_selected == 0:
            # No features selected.
            raise ValueError('No continuous feature was found!')
        else:
            pca = PCA(svd_solver=self.svd_solver, iterated_power=self.iterated_power, random_state=self.random_state)
            return pca.fit_transform(X_sel)