Hi:
Reordering the dimensions, then doing a vectorized addition, then reordering
(back) again is faster, it seems.
> m <- 20; n <- 30; p <- 40; q <- 30
> a <- NA
> length(a) <- m * n * p * q
> dim(a) <- c(m, n, p, q)
> x <- 1:n
> a[1:m,,1:p,1:q] <- 0
> b <- a
>
> # Approach 1
> system.time({
+ c <- aperm(a,c(2,1,3,4))
+ c <- c + x
+ a <- aperm(c,c(2,1,3,4))
+ })
user system elapsed
0.019 0.001 0.019
>
> # Approach 2
> system.time({
+ b[1:m,,1:p,1:q] <- 0
+ for (mm in 1:m) {
+ for (pp in 1:p) {
+ for (qq in 1:q) {
+ b[mm,,pp,qq] <- x
+ }
+ }
+ }
+ })
user system elapsed
0.149 0.002 0.150
> all(b==a)
[1] TRUE
>
On May 18, 2011, at 11:47 PM, huron wrote:
> Hello,
>
> A simple question, although I can't find an answer via my google/forum
> search:
>
> I have a 4-dimensional array; call it A[1:M,1:N,1:P,1:Q]. I have a vector x
> that is N by 1.
>
> I would like to "quickly" add x to the 2nd dimension of A; in other words, I
> want a quicker way of doing the following:
>
> for (m in 1:M) {
> for (p in 1:P) {
> for (q in 1:Q) {
> A[m,,p,q] = A[m,,p,q] + x
> }
> }
> }
>
> Thanks in advance!
>
> --
> View this message in context:
> http://r.789695.n4.nabble.com/Add-a-vector-to-the-values-in-a-specific-dimension-of-an-array-tp3534730p3534730.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
