Find combinations of rows grouped by a column by picking 1 row from each group in dataframe

Question

I have a dataframe which can be grouped by column. Each row in a particular group has a unique id. By picking 1 row from each group, I want to form all possible combinations possible.

I have tried to solve it by combn() and expand.grid(). But was not able to get the desired solution.

I have following type of data

Col1  id  Unique id
A     1     A_1
A     2     A_2
B     1     B_1
C     1     C_1
C     2     C_2
C     3     C_3

I want something like this:

Groups or dataframes for following type:

(A_1,B_1,C_1)
(A_1,B_1,C_2)
(A_1,B_1,C_3)
(A_2,B_1,C_1)
(A_2,B_1,C_2)
(A_2,B_1,C_3)

Here I have shown only 3 groups that are A,B,C. I real dataset I can have any number of groups and each row can have any number of ids. Please help me for this with code or logic whatever possible.

Answer 1

You can split unique_id by Col1 then use expand.grid().

expand.grid(split(df$Unique_id, f = df$Col1))

    A   B   C
1 A_1 B_1 C_1
2 A_2 B_1 C_1
3 A_1 B_1 C_2
4 A_2 B_1 C_2
5 A_1 B_1 C_3
6 A_2 B_1 C_3