The new dload() function in 7.10
in [Matlab]
|
From: Steven Lord on 24 Mar 2010 23:09 "Tim " <evitaerc(a)gmail.com> wrote in message news:ho90fp$6f4$1(a)fred.mathworks.com... > Yes Edric, that is definitely something I hope gets implemented in the > future (along with sparse distributed matrices eventually supporting > inversion via MUMPS or whatever). Unless you EXPLICITLY require the inverse of the sparse matrix (keeping in mind that even if the inverse exists it may not be sparse, of course) and are aware of the consequences of inverting a matrix, don't invert the matrix. If you're trying to invert in order to solve a system of equations, use backslash (\) to solve the system rather than multiplying by the inverse. -- Steve Lord slord(a)mathworks.com comp.soft-sys.matlab (CSSM) FAQ: http://matlabwiki.mathworks.com/MATLAB_FAQ
From: Tim on 25 Mar 2010 00:27
"Steven Lord" <slord(a)mathworks.com> wrote in message <hoek5j$72b$1(a)fred.mathworks.com>... > > If you're trying to invert in order to solve a system of equations, > use backslash (\) to solve the system rather than multiplying by the > inverse. > The backslash is exactly what I mean by matrix inversion. I don't know anyone who actually runs inv(A) unless they're trying to numerically verify some matrix theorem. The backslash operator (\) is currently not implemented for sparse distributed matrices, as of 2009b/PCT v4.2 (and not mentioned in the release note of 2010a). I fully realize the challenges in efficiently solving sparse systems explicitly. I'm just wondering whether this is on the roadmap for the near future. |