Inverse to cbind() function R

Question

Suppose I have a list of matrices:

matrix <- matrix(1:4, nrow = 2, ncol = 2)
list <- list(matrix, matrix, matrix)

And a matrix created by function cbind():

long.matrix <- do.call(cbind, list)

      [,1] [,2] [,3] [,4] [,5] [,6]
[1,]    1    3    1    3    1    3
[2,]    2    4    2    4    2    4

I want to reverse the process to get list of matrices from the long.matrix.

I can do it manually with the for loop, but I am searching for something like: function(long.matrix, 3) which I think should exist. Is there a such thing?

Answer 1

Brute-force solution:

f <- function(long.matrix, num)
           lapply(split(long.matrix, 
                        rep(seq(num), each=(ncol(long.matrix)/num)*nrow(long.matrix))), 
                  function(x) matrix(x, nrow=nrow(long.matrix))
           )

f(long.matrix, 3)
## $`1`
##      [,1] [,2]
## [1,]    1    3
## [2,]    2    4
## 
## $`2`
##      [,1] [,2]
## [1,]    1    3
## [2,]    2    4
## 
## $`3`
##      [,1] [,2]
## [1,]    1    3
## [2,]    2    4

rep builds the categories for split, to split the data. As R is column-major, here we take the first four, second four, an third four entries.

Filling in the values for the current dimensions of your example long.matrix and 3, the function reduces to this:

lapply(split(long.matrix, rep(seq(3), each=4)), function(x) matrix(x, nrow=2))

Note:

(r <- rep(seq(3), each=4) )
## [1] 1 1 1 1 2 2 2 2 3 3 3 3
split(long.matrix, r)
## $`1`
## [1] 1 2 3 4
## 
## $`2`
## [1] 1 2 3 4
## 
## $`3`
## [1] 1 2 3 4

Each of those is then passed to matrix to get the desired format.