parallel computing in R - r

Question

I am familiar with Matlab, but recently started to use R. I encounterd a problem when using parallel computing in R.
I want to use a matrix or a 3-d array as an output after parellel computing. In Matlab, an example what I want to do is as follows.
X=zeros(10,5,100);
Y=zeros(100,2);
parfor i=1:100;
X(:,:,i) = randn(10,5);
Y(i,:) = randn(1,2);
end
However, as long as I investigated, foreach in R seem return only vectors(list?) and a matrix or an array seem not allowed. I'm wondering how I need to wright a code to implement what Matlab does.

Answer 1

Here is my suggested solution based on the Matlab code in the question.
#install.packages("foreach")
#install.packages("doParallel")
library(foreach)
library(doParallel)
X <- array(c(rep(0,10), rep(0,5), rep(0,100)),dim = c(10,5,100))
Y <- array(c(rep(0,100), rep(0,2)),dim = c(100,2))
X=foreach(i=1:100) %dopar% {
X[ , , i]= array(c( rnorm(10), rnorm(5)),dim = c(10,5))
X
}
Y=foreach(i=1:100) %dopar% {
Y[ i , ]= array(c( rnorm(1), rnorm(2)),dim = c(1,2))
Y
}

Answer 2

You can use the following R code to replicate your Matlab code. Note that R is vectorized, so for loops are often unnecessary.
X <- rep(0, 10*5*100)
Y <- rep(0, 100*2)
dim(X) <- c(10,5,100)
dim(Y) <- c(100,2)
set.seed(1234) # This is just for replication. Can omit if you want fully random numbers
X[] <- rnorm(length(X))
Y[] <- rnorm(length(Y))
apply(X, c(1,2), mean)
apply(Y, 2, mean)
Output:
> apply(X, c(1,2), mean)
[,1] [,2] [,3] [,4] [,5]
[1,] 0.055024278 -0.0205575146 -0.071816174 -0.087659326 0.12707484
[2,] -0.018254630 0.0821710032 -0.005589141 0.049391162 -0.01413225
[3,] -0.009338280 0.0284398951 -0.004083183 0.013750904 -0.02076238
[4,] 0.039544807 -0.0425689205 0.016054568 0.013936608 -0.06183537
[5,] 0.053641184 0.1362104005 0.069674908 0.008190821 -0.21042331
[6,] 0.065895652 -0.0098767327 -0.082148255 0.038705556 -0.05255018
[7,] 0.007917696 -0.0002747114 -0.045812106 -0.062452164 0.23984287
[8,] -0.173435275 0.0328011859 -0.053835173 0.057308693 -0.03760174
[9,] 0.047481900 -0.0225967973 -0.161777736 0.005625679 -0.05406814
[10,] -0.031460338 0.0628553230 -0.176667084 -0.098304874 0.06060704
> apply(Y, 2, mean)
[1] 0.06083109 -0.03338719
You can also simplify this a little with the following commands:
set.seed(1234)
X <- rnorm(10*5*100)
Y <- rnorm(100*2)
dim(X) <- c(10,5,100)
dim(Y) <- c(100,2)
apply(X, c(1,2), mean)
apply(Y, 2, mean)