How to convert Pandas Groupby result to Dataframe

Asked Sep 12 '20 at 06:36

Active Sep 12 '20 at 06:53

Viewed 41 times

I have used the groupby method on a Pandas Dataframe and I want to convert the result to a DataFrame. I have used unstack but it only converts the results to features and not a proper dataframe.

I have tried approaches answered in other SO Questions like this but it was not useful.

Original DataFrame - badges

UserId      Class   
-1      |   B
-1      |   B
1       |   A   
100     |   A   
100     |   B
102     |   B   
102     |   B
107     |   A

Code
badges = badges.groupby('UserId')['Class'].value_counts().unstack().fillna(0)

Output

Class        A  |  B    
UserId          
-1      |   0.0 | 2.0   
1       |   1.0 | 0.0   
100     |   1.0 | 1.0   
102     |   0.0 | 2.0   
107     |   1.0 | 0.0

Required Output

UserId  |    A  |  B
    
-1      |   0.0 | 2.0   
1       |   1.0 | 0.0   
100     |   1.0 | 1.0   
102     |   0.0 | 2.0   
107     |   1.0 | 0.0

edited Sep 12 '20 at 06:53

asked Sep 12 '20 at 06:36

Ishan Dutta

2

can you share the dataset? – Soumendra Mishra Sep 12 '20 at 06:49
after you do the groupby , try `badges.rename_axis(None,axis=1).reset_index()` – anky Sep 12 '20 at 06:50
I have added the dataset @SoumendraMishra – Ishan Dutta Sep 12 '20 at 06:55
@anky your code did not work. And I want `UserId` to be a feature not index. – Ishan Dutta Sep 12 '20 at 06:56
@IshanDutta it did work for me , `badges = badges.groupby('UserId')['Class'].value_counts().unstack().fillna(0).rename_axis(None,axis=1).reset_index()` – anky Sep 12 '20 at 06:58

How to convert Pandas Groupby result to Dataframe

0 Answers0