Pandas dataframe get expanding/rolling value counts
Question:
Is it possible to create value counts for a series in a cumulative/expanding fashion? E.g. for the following series
pd.Series(['a', 'a', 'b', 'c', 'c'])
I’d want
[ 'a', 'b', 'c'
1, 0, 0
2, 0, 0
2, 1, 0
2, 1, 1
2, 1, 2]
Answers:
Encode the series into indicator variables and use cumsum
s.str.get_dummies().cumsum()
a b c
0 1 0 0
1 2 0 0
2 2 1 0
3 2 1 1
4 2 1 2
Is it possible to create value counts for a series in a cumulative/expanding fashion? E.g. for the following series
pd.Series(['a', 'a', 'b', 'c', 'c'])
I’d want
[ 'a', 'b', 'c'
1, 0, 0
2, 0, 0
2, 1, 0
2, 1, 1
2, 1, 2]
Encode the series into indicator variables and use cumsum
s.str.get_dummies().cumsum()
a b c
0 1 0 0
1 2 0 0
2 2 1 0
3 2 1 1
4 2 1 2