Missing values imputation in python

Question

After applying Imputer.fit_transform() on my dataset I am losing the column names on the transformed data frame. Is there any way to impute it without losing column names??

Answer 1

As I said in the comment to the question, just replace (re-assign) the values in the dataframe with the data returned from the Imputer.

Lets say this is your dataframe:

import numpy as np
import pandas as pd

df = pd.DataFrame(data=[[1,2,3], 
                        [3,4,4],
                        [3,5,np.nan], 
                        [6,7,8],
                        [3,np.nan,1]],
                  columns=['A', 'B', 'C'])

Current df:

   A    B    C
0  1  2.0  3.0
1  3  4.0  4.0
2  3  5.0  NaN
3  6  7.0  8.0
4  3  NaN  1.0

If you are sending whole the df to Imputer, just use this:

df[df.columns] = Imputer().fit_transform(df)

If you are sending only some columns, then use those columns only to assign the results:

columns_to_impute = ['B', 'C']
df[columns_to_impute] = Imputer().fit_transform(df[columns_to_impute])

Output:

     A    B    C
0  1.0  2.0  3.0
1  3.0  4.0  4.0
2  3.0  5.0  4.0
3  6.0  7.0  8.0
4  3.0  4.5  1.0