Pandas - count distinct values per column [duplicate]

Answer 1

I think you need groupby with nunique :

print (df)
   Id  ActivityId  ActivityCode
0   1           2             3
1   1           2             4
2   1           3             2
3   2           8             7

df = df.groupby('Id')['ActivityId'].nunique()
print (df)
Id
1    2
2    1
Name: ActivityId, dtype: int64

And for dict add Series.to_dict:

d = df.groupby('Id')['ActivityId'].nunique().to_dict()
print (d)
{1: 2, 2: 1}