how to multiply a N*N*M matrix with 1*M array

I have a matrix A with size N*N*M and array b with the size of 1*M.

I want to multiply the elements of i^th dimension of matrix A (i.e. A(:,:,i) ) with the i^th dimension of array b (i.e. b(1,i)). Meaning that: e.g.:

A(:,:,1)=[1 1; 1 1]
A(:,:,2)=[1 1; 1 1]
A(:,:,3)=[1 1; 1 1]
b=[1 2 3]
=> A times b
Results:
C(:,:,1)=[1 1; 1 1]
C(:,:,2)=[2 2; 2 2]
C(:,:,3)=[3 3; 3 3]

I am not looking for writing a loop and go through every single layer separately.

I was wondering if you guys know a fast way to do it. Thanks

Answer 1

[Stephen23]

This works correctly, and is fast and efficient:

A(:,:,1)=[1 1; 1 1];
A(:,:,2)=[1 1; 1 1];
A(:,:,3)=[1 1; 1 1];
b=[1 2 3];

bsxfun(@times,A,reshape(b,1,1,[]))

Answer 2

[Walter Roberson]

gpuarray() and pagefun() if you want fast ways of doing it.

Otherwise, writing a loop is going to have the best performance. You can hide the loop by using arrayfun() but that will be slower.