Re: [boost] Numerical integration library?

From: Sohail Somani (s.somani_at_[hidden])
Date: 2006-08-01 13:55:37

> -----Original Message-----
> From: boost-bounces_at_[hidden] 
> [mailto:boost-bounces_at_[hidden]] On Behalf Of John Maddock
> Sent: Tuesday, August 01, 2006 10:36 AM
> To: boost_at_[hidden]
> Subject: Re: [boost] Numerical integration library?
> 
> Sohail Somani wrote:
> > Yes this is what I meant. It seems that there is different
> > nomenclature regarding this relatively common concept.
> >
> > Is that your library you pointed to above?
> 
> Yep, still work in progress as you can see.
> 
> BTW, I'm also interested in numeric integration, but only of 
> the "integrate
> function f from x to y" not ODE solving.  Sigh, that 
> nomenclature again :-/

If I can only suggest: Decouple evaluation of the objective function
from its derivatives. Or have another concept that is not callable.

I have an example of how one can specify a family of functions and their
derivatives:

http://sohail.taggedtype.net/software/newton_raphson.cxx

You can blame the idle hands for this one.

Sohail