[GiNaC-list] advice on a GiNaC based project.

[Alexei Sheplyakov]

Hello!

On Fri, May 23, 2008 at 12:47:55AM -0700, Bhalchandra Thatte wrote:

> I wish to write a few programs that do something like this: 
> take a graph or a tree
> each vertex or edge of the graph has a variable associated with it.
> walk on the graph or find certain subgraphs, and compute certain
> multivariable polynomials in terms of the variables associated
> with vertices or edges.

I think boost.graph (for graph manipulations) and GiNaC (for symbolic
computation) is a reasonable choice.

> later possibly do numerical optimization of the polynomials using
> an optimization library.

Be forewarned: GiNaC does not provide any kind of numerical optimization.

> I thought GiNaC would be ideal for this type of applications because
> C++ will give me complete control over the data structures and
> the efficiency.

I suggest to think twice if you really need a low level programming
language (such as C++) at all.

Best regards,
	Alexei

-- 
All science is either physics or stamp collecting.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: Digital signature
Url : http://www.cebix.net/pipermail/ginac-list/attachments/20080523/d7f167ed/attachment.sig