Source code for evalml.pipelines.components.transformers.imputers.knn_imputer

"""Component that imputes missing data according to a specified imputation strategy."""

import numpy as np
import pandas as pd
import woodwork
from sklearn.impute import KNNImputer as Sk_KNNImputer

from evalml.pipelines.components.transformers import Transformer
from evalml.utils import infer_feature_types
from evalml.utils.nullable_type_utils import _get_new_logical_types_for_imputed_data


class KNNImputer(Transformer):
    """Imputes missing data using KNN according to a specified number of neighbors.  Natural language columns are ignored.

    Args:
        number_neighbors (int): Number of nearest neighbors for KNN to search for. Defaults to 3.
        random_seed (int): Seed for the random number generator. Defaults to 0.

    """

    name = "KNN Imputer"

    def __init__(self, number_neighbors=3, random_seed=0, **kwargs):
        parameters = {"number_neighbors": number_neighbors}
        parameters.update(kwargs)

        imputer = Sk_KNNImputer(
            n_neighbors=number_neighbors,
            missing_values=np.nan,
            **kwargs,
        )
        self._all_null_cols = None
        super().__init__(
            parameters=parameters,
            component_obj=imputer,
            random_seed=random_seed,
        )

    def fit(self, X, y=None):
        """Fits imputer to data. 'None' values are converted to np.nan before imputation and are treated as the same.

        Args:
            X (pd.DataFrame or np.ndarray): the input training data of shape [n_samples, n_features]
            y (pd.Series, optional): the target training data of length [n_samples]

        Returns:
            self

        Raises:
            ValueError: if the KNNImputer receives a dataframe with both Boolean and Categorical data.

        """
        X = infer_feature_types(X)

        nan_ratio = X.isna().sum() / X.shape[0]

        # Keep track of the different types of data in X
        self._all_null_cols = nan_ratio[nan_ratio == 1].index.tolist()
        self._natural_language_cols = list(
            X.ww.select("NaturalLanguage", return_schema=True).columns.keys(),
        )

        # Only impute data that is not natural language columns or fully null
        self._cols_to_impute = [
            col
            for col in X.columns
            if col not in self._natural_language_cols and col not in self._all_null_cols
        ]

        # If the Dataframe only had natural language columns, do nothing.
        if not self._cols_to_impute:
            return self

        self._component_obj.fit(X[self._cols_to_impute], y)
        return self

    def transform(self, X, y=None):
        """Transforms input by imputing missing values. 'None' and np.nan values are treated as the same.

        Args:
            X (pd.DataFrame): Data to transform.
            y (pd.Series, optional): Ignored.

        Returns:
            pd.DataFrame: Transformed X
        """
        # Record original data
        X = infer_feature_types(X)
        original_index = X.index
        original_schema = X.ww.schema

        # separate out just the columns we are imputing
        X_t = X[self._cols_to_impute]
        if not self._cols_to_impute:
            not_all_null_cols = [
                col for col in X.columns if col not in self._all_null_cols
            ]
            return X.ww[not_all_null_cols]

        # Transform the data
        X_t = self._component_obj.transform(X_t)
        X_t = pd.DataFrame(X_t, columns=self._cols_to_impute)

        # Reinit woodwork, maintaining original types where possible
        imputed_schema = original_schema.get_subset_schema(self._cols_to_impute)
        new_ltypes = _get_new_logical_types_for_imputed_data(
            impute_strategy="knn",
            original_schema=imputed_schema,
        )
        X_t.ww.init(schema=imputed_schema, logical_types=new_ltypes)

        # Add back in the unchanged original natural language columns that we want to keep
        if len(self._natural_language_cols) > 0:
            X_t = woodwork.concat_columns([X_t, X.ww[self._natural_language_cols]])
            # reorder columns to match original
            X_t = X_t.ww[[col for col in original_schema.columns if col in X_t.columns]]

        if self._cols_to_impute:
            X_t.index = original_index
        return X_t

    def fit_transform(self, X, y=None):
        """Fits on X and transforms X.

        Args:
            X (pd.DataFrame): Data to fit and transform
            y (pd.Series, optional): Target data.

        Returns:
            pd.DataFrame: Transformed X
        """
        return self.fit(X, y).transform(X, y)