使用条件的熊猫缺失值（按其他列分组）

Question

I have a data frame like this

df

Col A    Col B    Col C 

25       1         2          
NaN      3         1
27       2         3 
29       3         1

I want to fill Nan values in col A based on Col C and Col B.

My output df should be like this

25       1         2          
29       3         1
27       2         3 
29       3         1

I have tried this code df.groupby(['Col B','Col C']).ffill()

but didnt worked.Any suggestion will be helpful

Answer 1

Here you go:

df['Col A'] = df["Col A"].fillna(df.groupby(['Col B','Col C'])["Col A"].transform(lambda x: x.mean()))
print(df)

Prints:

   Col A  Col B  Col C
0   25.0      1      2
1   29.0      3      1
2   27.0      2      3
3   29.0      3      1

Answer 2

You can try

df.fillna(df.groupby(['ColB','ColC']).transform('first'),inplace=True)
df
Out[386]: 
   ColA  ColB  ColC
0  25.0     1     2
1  29.0     3     1
2  27.0     2     3
3  29.0     3     1