CODEWITHSUNDEEP
pythonpython-3.xpandaspandas-groupby
Ajay Chinni
Ajay Chinni

Reputation: 850

Runing Count with rest condition on another row

I have dataframe where I want to keep on increasing the value until val changes and when id changes reset the count value

data = [['p1','1'],
        ['p1','1'],
        ['p1','2'],
        ['p2','3'],
        ['p2','5'],
        ['p3','2']]

df = pd.DataFrame(data = data,columns = ['id','val'])

Desired output

   id val count
0  p1   1     1  
1  p1   1     1
2  p1   2     2
3  p2   3     1
4  p2   5     2
5  p3   2     1

I could only come up with this which does the cumsum but I want a counter

df_1.groupby('id')['val'].cumsum()

My unwanted output

   id  val  count
0  p1    1      1
1  p1    1      2
2  p1    2      4
3  p2    3      3
4  p2    5      8
5  p3    2      2

Upvotes: 1

Views: 51

Answers (1)

BENY
BENY

Reputation: 323316

You can try with transform +factorize

df['count']=df.groupby('id').val.transform(lambda x : x.factorize()[0]+1)
0    1
1    1
2    2
3    1
4    2
5    1
Name: val, dtype: int32

Upvotes: 1

Related Questions