Don't let that INV go past your eyes; to solve that system, FACTORIZE!
14 May 2009
06 Sep 2011)
A simple-to-use object-oriented method for solving linear systems and least-squares problems.
% Test code for the FACTORIZE object.
% To run all the tests, use test_all. The SPQR mexFunction from SuiteSparse
% is required. The FACTORIZE method works without SPQR, but it will not use
% COD for sparse matrices in that case (which this test relies upon). The
% output of this test in MATLAB R2011a is given in test_all.txt.
% reset_rand - resets the state of rand
% test_accuracy - test the accuracy of the factorize object
% test_all - test the Factorize package (factorize, inverse, and related)
% test_all_cod - test the COD factorization
% test_all_svd - tests the svd factorization method for a range of problems.
% test_cod - test the COD, COD_SPARSE and RQ functions
% test_disp - test the display method of the factorize object
% test_errors - tests error handling for the factorize object methods
% test_factorize - test the accuracy of the factorization object
% test_function - test various functions applied to a factorize object
% test_functions - test various functions applied to a factorize object
% test_performance - compare performance of factorization/solve methods.
% test_svd - test factorize(A,'svd') and factorize(A,'cod') for a given matrix
% Copyright 2011, Timothy A. Davis, University of Florida