[SciPy-user] a small OpenOpt example howto benchmark solvers

Nils Wagner nwagner@iam.uni-stuttgart...
Wed Nov 21 13:04:07 CST 2007


On Wed, 21 Nov 2007 10:41:12 +0200
  dmitrey <dmitrey.kroshko@scipy.org> wrote:
> Hi all,
> I have got a letter from Nils Wagner about interest of 
>NLP solvers 
> comparison, and maybe some of scipy-user mailist 
>subscribers could be 
> interested as well.
> 
> So if anyone is interested - here's an example of 
>ALGENCAN, cobyla and 
> lincher comparison:
> http://openopt.blogspot.com/2007/11/example-of-oo-nlp-solvers-benchmark.html
> 
> Please let me note that usually ALGENCAN is much more 
>better, and 
> choosing correct stop criteria for your problem 
>(gradtol, ftol, xtol 
> etc) is very essential.
> 
> Regards, D
> _______________________________________________
> SciPy-user mailing list
> SciPy-user@scipy.org
> http://projects.scipy.org/mailman/listinfo/scipy-user

  
Hi Dmitrey,

Do you have a reference for your benchmark example ?
I'm curious about it.

Nils


More information about the SciPy-user mailing list