Find the index of subset data frame from its original data frame in R-CodePudding

I have a data frame in R, say a1 here for a toy example, and I have a row subset from it, say a2. I want to find the original subset index (2,4).

I tried which or match, but did not succeed.

 set.seed(123)
 a1=data.frame(x1=rnorm(5),x2=runif(5),x3=runif(5))
 a2=a1[c(2,4),]
 a2index=rep(NA,dim(a2)[1])

Here is my a1 data.frame

    a1
           x1        x2         x3
1 -0.56047565 0.9568333 0.89982497
2 -0.23017749 0.4533342 0.24608773
3  1.55870831 0.6775706 0.04205953
4  0.07050839 0.5726334 0.32792072
5  0.12928774 0.1029247 0.95450365

a2 is a row subset of a1:

 a2
           x1        x2        x3
2 -0.23017749 0.4533342 0.2460877
4  0.07050839 0.5726334 0.3279207

I managed to obtain the index using double loop. But it is too slow, is there a way to speed it up?

Thanks for help.

for (i in 1:dim(a2)[1] )
   for (j in 1:dim(a1)[1])
     if (all(a2[i,]==a1[j,])){
       a2index[i]=j
       break;
       } 

# return the index vector (2,4) 
a2index``

CodePudding user response：

You can use match after transposing your data frames, so that it can have column-wise comparison.

match(as.data.frame(t(a2)), as.data.frame(t(a1)))
[1] 2 4