I have a dataframe that has five colums: G1, G2, G3, G4 and Tax. I’m grouping by G and Tax for each G column to count how many rows has the value 1. But the dimension of result is different for each G because some G’s only have 0 values.

This is my code:

self.instance.df.groupby(["G1", "Tax"]).size().drop(0)
self.instance.df.groupby(["G2", "Tax"]).size().drop(0)

I’m dropping the index 0 because i only need 1 values.

The result i’m expecting is:

G1  Tax
1   1.1   5
    1.2   7
    1.5   3
    1.7   0

and the result i’m getting:

 G1 Tax
 1  1.1   5
    1.2   7
    1.5   3