I have an implementation for mostly arbitrary feed forward networks in OCaml.
Training can be done with gradient descent (fixed learning rate or
adaptive) and momentum; I have yet to integrate in a second order
optimizer (a wrapper around Nocedal's L-BFGS code), but it should be
straight forward. I haven't really had the need as I'm not working with
neural nets right now. The code is written in an imperative style. I haven't
released the code yet, but can provide it if you're interested.

liam

On Wed, May 25, 2005 at 02:56:31PM +0200, Charles VEJNAR wrote:
> Good morning,
> 
> I am looking for Neural Network library written in OCaml.
> 
> Did you know an implementation ?
> 
> Thank you
> 
> Charles
> 
> _______________________________________________
> Caml-list mailing list. Subscription management:
> http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list
> Archives: http://caml.inria.fr
> Beginner's list: http://groups.yahoo.com/group/ocaml_beginners
> Bug reports: http://caml.inria.fr/bin/caml-bugs

-- 
Liam Stewart :: liam@squozen.org