Combination of values in pandas data frame

Question

This is my pandas dataframe:

       Item          Support_Count
0      BREAD              4
1      MILK               4
2      DIAPER             4
3      BEER               3

How will i generate all possible unique combinations of 2 and 3 set of items from the 1st column 'Item'.

Example(2 item sets): (BREAD,MILK) ,(BREAD,DIAPER),(BREAD,BEER),(MILK,DIAPER) etc.

Example (3 item sets): (BREAD,MILK,DIAPER),(BREAD,MILK,BEER),(MILK,DIAPER,BEER) etc.

Answer 1

You can use the itertools library:

import itertools
list(itertools.combinations(df['Item'], 2))

[('BREAD', 'MILK'),
 ('BREAD', 'DIAPER'),
 ('BREAD', 'BEER'),
 ('MILK', 'DIAPER'),
 ('MILK', 'BEER'),
 ('DIAPER', 'BEER')]

list(itertools.combinations(df['Item'], 3))

[('BREAD', 'MILK', 'DIAPER'),
 ('BREAD', 'MILK', 'BEER'),
 ('BREAD', 'DIAPER', 'BEER'),
 ('MILK', 'DIAPER', 'BEER')]

Note: The number of combinations grows very quickly so generating all possible combinations may not be efficient. I recommend looking at apriori algorithm implementations if you haven't already done so.