Fillna if all the values of a column are null in pandas

November 17, 2021

I have to fill a column only if all the values of that column are null. For example c

df = pd.DataFrame(data = {"col1":[3, np.nan, np.nan, 21, np.nan],
                          "col2":[4, np.nan, 12, np.nan, np.nan],
                          "col3":[33, np.nan, 55, np.nan, np.nan],
                          "col4":[np.nan, np.nan, np.nan, np.nan, np.nan]})


>>> df
   col1  col2  col3  col4
0   3.0   4.0  33.0   NaN
1   NaN   NaN   NaN   NaN
2   NaN  12.0  55.0   NaN
3  21.0   NaN   NaN   NaN
4   NaN   NaN   NaN   NaN

In the above example, I have to replace the values of col4 with 100 since all the values are null/NaN.

So for the above example. I have to get the output as below.

   col1  col2  col3  col4
0   3.0   4.0  33.0   100
1   NaN   NaN   NaN   100
2   NaN  12.0  55.0   100
3  21.0   NaN   NaN   100
4   NaN   NaN   NaN   100

Tried using the below command. But its replacing values of a column only if it contains atleast 1 non-nan value

df.where(df.isnull().all(axis=1), df.fillna(100), inplace=True)

Could you please let me know how to do this.

Thanks

>Solution :

Use indexing:

df.loc[:, df.isna().all()] = 100
print(df)

# Output:
   col1  col2  col3   col4
0   3.0   4.0  33.0  100.0
1   NaN   NaN   NaN  100.0
2   NaN  12.0  55.0  100.0
3  21.0   NaN   NaN  100.0
4   NaN   NaN   NaN  100.0