[astro] conjugate gradient stuff

[Travis Desell]

So i think i've isolated the problem as to why the conjugate gradient 
descent isn't working.  I've
implemented the line search from the matlab code, and it's taking the 
same number of iterations
to converge as the previously implemented version -- so that isn't the 
problem.

One thing i've noticed in the matlab code is that a "step" is calculated 
and the search direction is
modified by the line search before it was used, so I was wondering if 
Malik could explain what
the code is doing a bit (since i'm not too familiar with matlab).

Basically, from the code in "linesearch.h" I'm wondering how the 
variables dser, d and step are
calculated.

-- 
----------------------------------------------------------------------------------------------------------
Travis Desell
<deselt @ cs.rpi.edu>
1-518-878-7424
Worldwide Computing Laboratory ( http://wcl.cs.rpi.edu/ )
Rensselaer Polytechnic Institute, 110 8th Street, Troy NY 12180, USA
----------------------------------------------------------------------------------------------------------