Creating a column with row number when another column value changes python

Question

I have a dataframe like this

package_num   delivery_date
122           12/21
122           12/21
134           12/30
156           12/30
156           12/30

I want to be able to do use something like MySQL row_number() but on python. I want a column of row number that starts with 1 and goes up increments of 1 until package number changes. So my desired outcome is:

package_num   delivery_date    row_number
122           12/21            1
122           12/21            2
134           12/30            1
156           12/30            1
156           12/30            2

meW · Answer 1 · 2019-01-03T05:52:22.297

2

A bit extended version using list comprehension:

val = df.groupby('package_num').count().values.ravel()
df['row_numer'] = [j  for i in val for j in range(1, i+1)]

edited Jan 03 '19 at 05:52

answered Jan 03 '19 at 05:43

meW

3,832
7
27

score 1 · Answer 2 · edited Jan 03 '19 at 05:51

1

You can use cumcount() , like this

df['row_number']= df.groupby('package_num').cumcount()+1


package_num   delivery_date    row_number
122           12/21            1
122           12/21            2
134           12/30            1
156           12/30            1
156           12/30            2

edited Jan 03 '19 at 05:51

meW

3,832
7
27

answered Jan 03 '19 at 05:40

Courage

772
8
20

1

groupby should be applied on "package_num" column – Ernest S Kirubakaran Jan 03 '19 at 05:46
1

@ErnestSKirubakaran Oh right, thanks – Courage Jan 03 '19 at 05:51

Creating a column with row number when another column value changes python

2 Answers2