Matrix/Linear Algebra Library?

[Oliver Dathe]

I used LinAlg of the Shark-Project for SVD but is has no sparse matrix 
support and is C++.
http://shark-project.sourceforge.net/

GNU Scientific Library
http://www.gnu.org/software/gsl/
http://www.gnu.org/software/gsl/manual/html_node/Singular-Value-Decomposition.html
It seem to have no sparse matrix support but there are bindings in
http://www.dsource.org/projects/bindings/browser/trunk/gsl

SVDLIBC
http://tedlab.mit.edu/~dr/SVDLIBC/
It has sparse matrix support but you have to make bindings yourself or 
try bcd for that
http://www.dsource.org/projects/bcd

NumPy
http://www.scipy.org/doc/numpy_api_docs/numpy.linalg.linalg.html#svd
I've never used Python but there is Kirks well known pyd for interfacing 
between Python and D.

Sure you need SVD? Often it is used for least squares fitting or 
pseudoinverse. So depending on what you are doing there may be weaker 
alternatives like Moore-Penrose pseudoinverse.