Updated Sat, 16 Apr 2011 02:45:41 +0000
These routines assist in the manipulation of matrices with same shape but different content. For example, performing the product A*b is trivial for matrix A and vector b, but what would you do if you had several such products to form? Examples abound: rotations, Jacobians, covariances, etc. Using the frontal routines, you'd collect them all in a three-dimensional matrix or third-order tensor, with each k-th frontal panel of A(:,:,k) and b(:,:,k) storing one such a related matrix and vector. Then calling
C = frontal_mtimes(A, b);
would do the equivalent of
for k=1:size(A,3), C(:,:,k) = A(:,:,k) * b(:,:,k); end
but using internally different algorithms depending on the dimensions of A (including a C-mex option). If you like operator overloading, you can do instead:
A = frontal(A);
b = frontal(b);
C = A*b;
You might want to compile the file frontal_mtimes_helper.c, but it's not required.
After you've unzipped it, test your installation running:
(don't do addpath(genpath('c:\work\fx\frontal\frontal\')))
Felipe G. Nievinski (2023). frontal (https://www.mathworks.com/matlabcentral/fileexchange/30764-frontal), MATLAB Central File Exchange. Retrieved .
MATLAB Release Compatibility
Platform CompatibilityWindows macOS Linux
Inspired by: Multiple matrix multiplications, with array expansion enabled, MTIMESX - Fast Matrix Multiply with Multi-Dimensional Support, testit
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!Start Hunting!
Discover Live Editor
Create scripts with code, output, and formatted text in a single executable document.