R Remove rows if value in one column appears again in another column without loop

Question

I would like to remove cases where the combination of two columns also appears as a reverse combination in those two same columns

Below is the input and desired output.

SS of Tables

It doesn't matter if the first or second occurrence is kept.

Data:

df <- data.frame(
  "x1" = 1:6, 
  "x2" = c(2,1,4,3,6,5), 
  "x3" = c("a","b", "c","d","e","f"))

Can you please share your example input in a format I can copy/paste in to R? Images are a very unfriendly way to share data. `dput(your_input)` or `dput(head(you_input))` is an easy way to make a copy/pasteable R object. — Gregor Thomas, Sep 04 '18 at 19:51
`nopes = unname(as.list(DT[, 2:1])); res = DT[!nopes, on=names(DT)[1:2]]`? Not sure if it works since there's no example to test with. — Frank, Sep 04 '18 at 19:51
Sure, Try this: df <- data.frame("x1" = 1:6, "x2" = c(2,1,4,3,6,5), "x3" = c("a","b", "c","d","e","f")) — a.m., Sep 04 '18 at 20:30
with that data, you can do `subset(df,!duplicated(t(apply(df[-3],1,sort))))` — Onyambu, Sep 04 '18 at 21:07

score 0 · Answer 1 · edited Sep 04 '18 at 21:42

0

The first suggestion worked just fine for me:

subset(df,!duplicated(t(apply(df[-3],1,sort))))

edited Sep 04 '18 at 21:42

Frank

answered Sep 04 '18 at 21:20

a.m.

1 Answers1