"Row-wise" reshape of 3d matrix -> 2d matrix

I need to do a reshape of a 3d matrix to a 2d matrix, but where first row of each matrix along the 3d dimension corresponds to the first "set" of rows in the finished 2d matrix.

For example, lets say I have a matrix like this:

M(:,:,1) = [
  111  121
  211  221]

M(:,:,2) = [
  112  122
  212  222]

M(:,:,3) = [
  113  123
  213  223]

I want it to be reshaped into

M_Reshaped
  111  121
  112  122
  113  123
  211  221
  212  222
  213  223

I know about

reshape(permute(M,[1 3 2]), [], 3, 1);

But that simply stacks each matrix along the 3d dimension on top of each other, which is not what I want to achieve. I could do it in a loop, but I've got a feeling there's a more elegant and efficient way that I have not managed to figure out.

Answer 1

[Guillaume]

You simply need a different permutation of the dimensions:

 reshape(permute(M, [3 1 2]), [], size(M, 2))

[Oscar Frick]

I clearly need a better feeling for how the permute + reshape works, thanks!

[Raymond MacNeil]

Good question, Oscar.