Here's the pandas dataframe that I'm using to learn how to do this:

import pandas as pd
test_list = pd.DataFrame()
test_list["Item"] = ["A", "B", "C", "D", "E", "F", "G", "H", "I", "J", "K"]
test_list["Number"] = ["1", "2", "3", "4", "5", "6", "7", "8", "9", "10", "11"]
test_list["Combined Numbers"]= ""

Based on that dataframe above, I intend to combine up to 3 numbers, separated by commas. Following that, I intend to repeat this combined value I now have, for each of the test_list["Item"] and test_list["Number"] involved.

I've been scratching my head figuring it out so far. So far I've seen examples of groupby() function for situations like combining information based on a given criteria, like a duplicate value from a column. I'm learning to explore if I don't have anything to refer to, how can I work this out instead?

Here's my intended goal:

Thank you

CodePudding user response：

In you case as you don't have this column with the groups, you can generate a range of the length of your dataframe (with np.arange) and do the floor division by 3 (//3). Use groupby.transform to keep the original shape of your data and do the join operation on the column.

test_list["Combined Numbers"] = (
    test_list.groupby(np.arange(len(test_list))//3)
    ['Number'].transform(', '.join)
)
print(test_list)
#    Item Number Combined Numbers
# 0     A      1          1, 2, 3
# 1     B      2          1, 2, 3
# 2     C      3          1, 2, 3
# 3     D      4          4, 5, 6
# 4     E      5          4, 5, 6
# 5     F      6          4, 5, 6
# 6     G      7          7, 8, 9
# 7     H      8          7, 8, 9
# 8     I      9          7, 8, 9
# 9     J     10           10, 11
# 10    K     11           10, 11