Select equivalent rows [A-B & B-A]

Question

My problem seems to be very simple but I'm not able to solve it since hours…

I have a matrix such this one:

      [,1] [,2]
[1,]    1    2
[2,]    2    1
[3,]    2    1
[4,]    3    4

I want to select the rows which have the same information, without regard to the order of the column. For instance row1 (1;2) and row2 (2;1). Then, i want to delete them, except one.

I have written this function, but it doesn't work…

f<-function(x){
     i<-1
     repeat
     {
     a<-c()
     a<-c(which(x[i,1]==x[,2] & x[i,2]==x[,1]))
          if(!is.null(a)) {x<-x[-c(a),]}
          if(i>=nrow(x)) {break} else {i<-i+1}
     }
     x
} 
f(data)

Somebody could give me a hint for this ?

Does the order of the elements within a row matter? For example, does it have to be `1,2` or can it also be `2,1`? — Blue Magister, Oct 29 '13 at 01:36

score 16 · Accepted Answer · answered Oct 29 '13 at 02:11

Like this:

unique(t(apply(mat, 1, sort)))

Note that output rows are sorted, so for example an "unmatched" row like c(5, 1) in the original data will appear as c(1, 5) in the output. If instead you want the output rows to be as they are in the input, then you can do:

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

Select equivalent rows [A-B & B-A]

1 Answers1

Linked

Related