CODEWITHSUNDEEP
pythonpython-2.7python-3.xpandas
Shubham R
Shubham R

Reputation: 7644

Find count of combinations of column in pandas df

i have a table in pandas df

product_id_x   product_id_y
1              2
1              3
1              4
3              7
3              11
3              14
3              2
and so on around (1000 rows)

i want to find the count of combinations for each product_id_x with product_id_y.

ie. 1 has combinations with 1-2,1-3,1-4(total 3 combinations) similarly 3 has total 4 combinations.

and create a dataframe df2 which has

product_id_x   combinations
1               3
3               4

and so on ..(distinct product_id_x's)

what approach should i follow? my skills on python are at a beginner level. Thanks in advance.

Upvotes: 1

Views: 1037

Answers (2)

piRSquared
piRSquared

Reputation: 294218

size counts the number of rows each of the column value pairs happen together. count counts the same thing but where they are not null. Since you did not mention anything about nulls, I'll use size after a groupby, then unstack

df.groupby(['product_id_x', 'product_id_y']).size().unstack(fill_value=0)

enter image description here

Upvotes: 2

AKS
AKS

Reputation: 19811

You can use groupby with agg on product_id_x column:

df2 = df.groupby(['product_id_x']).agg(['count'])

Or, you can directly use size function on the group to get size of each group:

df2 = df.groupby(['product_id_x']).size()

Upvotes: 2

Related Questions