in R: find observations with unique combinations a

Question

This question already has an answer here:

Select equivalent rows [A-B & B-A] [duplicate] (1 answer)

Closed 4 years ago.

I have a data frame with 1000 observations on 20 variables.

I want to select only the rows that have a unique combination across columns, regardless of their order.

That is, if a combination is ABA and another is BAA, I want the code only to return one of these combinations.

To identify unique combinations I run a simple unique command across multiple variables.

How would you write such a code?

Answer 1

We can sort the data by row using apply with MARGIN=1, then use duplicated to return the logical index, negate it and get the unique rows in the data.

dat[!duplicated(t(apply(dat, 1, sort))),]