I have a question about matching two matrices in R to perform a very simple calculation. I have working code (below), but I feel like there must be a more efficient and 'R-like' way to do it. Any clues very welcome.
The Problem
I have two matrices with related information. m1 contains a bunch of references. m2 contains data on those references (zeros or ones). I want to know which rows of m1 have a '0' in the first column and a '1' in the second column when you look up the data from m2. Here's a toy example:
> m1 <- matrix(data = c(51,52,53,51,54,55,56,57), nrow = 4, ncol = 2)
> m1
[,1] [,2]
[1,] 51 54
[2,] 52 55
[3,] 53 56
[4,] 51 57
> m2 <- matrix(data = c(0,0,1,0,0,1,1), nrow = 7, ncol = 1)
> rownames(m2) <- c(51,52,53,54,55,56,57)
> m2
[,1]
51 0
52 0
53 1
54 0
55 0
56 1
57 1
General properties are that I can already guarantee that every entry in m1 has a corresponding row name in m2, and I need to do this many millions of times on much larger matrices, so speed is useful.
What I want to do is use m2 to figure out which rows of m1 have a zero in the first column, and a 1 in the second column. In this case, only the final row of m1 has that property.
My Solution
I have a relatively O.K. solution to this, which uses apply() and is not too bad:
> is.zero.one <- function(line, m2){
+ start = m2[as.character(line[1]),]
+ end = m2[as.character(line[2]),]
+ if(start==0 && end==1){return(TRUE)}
+ else{return(FALSE)}
+}
> apply(m1, 1, is.zero.one, m2)
[1] FALSE FALSE FALSE TRUE
This works fine. But it feels clunky.
My question
Does anyone know of a smarter/faster/more natural way to do this? I had a look at match() and related functions, but couldn't come up with a solution. Ditto for searching around related questions here. Part of the reason for my question is that I'm not a very good R programmer, so even if this turns out to be a decent solution, I'd be very interested to see how others would solve it.
Thanks for any help.