[FIXED] Is there a way to force SimpleImputer to return a pandas dataframe?

Issue

The following code

from sklearn.impute import SimpleImputer
import pandas as pd

df = pd.DataFrame(dict(
    x=[1, 2, np.nan],
    y=[2, np.nan, 0]
))

SimpleImputer().fit_transform(df)

Returns

array([[1. , 2. ],
       [2. , 1. ],
       [1.5, 0. ]])

Is there a way to use an imputer that returns a pandas dataframe instead of a numpy array? Is there a scikit-learn implementation for that?

Solution

If you want to keep the columns (e.g. for using ColumnTransformers in a later step), you can create a wrapper around SimpleImputer:

df = pd.DataFrame({"A": [1, 2, np.NaN], "B": [3, np.NaN, 4], "C": [np.NaN, 5, 6]})

class PandasSimpleImputer(SimpleImputer):
    """A wrapper around `SimpleImputer` to return data frames with columns.
    """

    def fit(self, X, y=None):
        self.columns = X.columns
        return super().fit(X, y)

    def transform(self, X):
        return pd.DataFrame(super().transform(X), columns=self.columns)


PandasSimpleImputer().fit_transform(df)

>>>
    A   B   C
0   1.0 3.0 5.5
1   2.0 3.5 5.0
2   1.5 4.0 6.0

Answered By - nocibambi