non-negative least squares

[Brian Gough]

John Regehr writes:
 > I'm interested in creating a non-negative least squares routine in GSL
 > that has an interface similar to gsl_multifit_linear.  There appear to be
 > unencumbered NNLS codes in C available for downloading so I assume that
 > getting this working in GSL is mostly a matter of translating the matrix
 > access idioms into GSL calls?
 > 
 > I haven't hacked GSL before; does anyone have any advice for the best
 > place to put this, which GSL internals should be reused, etc?

Hi,

See the GSL Design document at http://sources.redhat.com/gsl/ for details.
The latest version is available through CVS.

regards

-- 
Brian Gough

Network Theory Ltd            
15 Royal Park                 
Bristol BS8 3AL               
United Kingdom                

Tel: +44 (0)117 3179309 
Fax: +44 (0)117 9048108              
Web: http://www.network-theory.co.uk/