Re: [R] Doing %o% that operates on columns instead of atomics

Is this what you want: > my.foo <- function( a, b ) a - b > z <- combn(ncol(y), 2) # get all the combinations of 2 columns > result <- do.call(cbind, lapply(seq(ncol(z)), function(.cols){ + my.foo(y[,z[1,.cols]], y[, z[2, .cols]]) + })) > > > result [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8]

Re: [R] Doing %o% that operates on columns instead of atomics

Maybe something like the code below? There is surely a more elegant way of handling the indices, maybe with plyr? -s > xx <- array(1:6,c(3,2)); xx [,1] [,2] [1,]14 [2,]25 [3,]36 > yy <- array(1:12,c(3,4)); yy [,1] [,2] [,3] [,4] [1,]147 1

Re: [R] Doing %o% that operates on columns instead of atomics

CORRECTION: > my.foo <- function( a, b ) { c <- a - b; } #really more complex, but just > to illustrate -- View this message in context: http://www.nabble.com/Doing--o--that-operates-on-columns-instead-of-atomics-tp22701363p22701400.html Sent from the R help mailing list archive at Nabble.c