Imputing missing value using zero for equal condition

user54nj1 0
Viewed 32

Having a dataframe as below

df1 = pd.DataFrame({'Name1':['A','B','A','B','B','C','C','C','E','E','E'],
                    'Name2':['A','B','D','C','D','D','A','B','A','B','C'],'Marks2':['Nan','Nan',6,50, 88,'Nan',140,9,'Nan',65,70]})
df1

if Name1 and Name2 are equal, then need to replace Marks2 with 0.The rest of the values need to be same.

dataframe pandas python
3 Answers
df['Marks2']=df['Marks2'].mask(df['Name1'].eq(df['Name2']), 0)
df

    Name1   Name2   Marks2
0   A   A   0
1   B   B   0
2   A   D   6
3   B   C   50
4   B   D   88
5   C   D   Nan
6   C   A   140
7   C   B   9
8   E   A   Nan
9   E   B   65
10  E   C   70
userCaFe0

If you want to always replace Marks2 (regardles of it being null or not) then:

df1.loc[df1["Name1"] == df1["Name2"], "Marks2"] = 0

If instead you only want to set Marks2 to 0 if it was null then:

df1.loc[(df1["Name1"] == df1["Name2"]) & df1["Marks2"].isna(),"Marks2"] = 0
Matteo Zanoni0

You can use .loc[] to filter on the rows where Name1 and Name2 are equal like so:

df1.loc[df1['Name1'] == df1['Name2']]

You can then specify the mark column, and set it to 0 when the above is the case

df1.loc[df1['Name1'] == df1['Name2'], 'Marks2'] = 0

   Name1 Name2 Marks2
0      A     A      0
1      B     B      0
2      A     D      6
3      B     C     50
4      B     D     88
5      C     D    Nan
6      C     A    140
7      C     B      9
8      E     A    Nan
9      E     B     65
10     E     C     70
Emi OB0
Related