[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: nonlinear optimization code (was Suggestion for improving Octave)
From: |
Jskud |
Subject: |
Re: nonlinear optimization code (was Suggestion for improving Octave) |
Date: |
Thu, 12 May 2005 09:16:46 -0700 |
Hi.
I have secured permission from Yinyu Ye to release his self-contained
"solnp" dense constrained nonlinear solver, under GPL. It is just two
functions, written entirely in Octave. It is simpler to use, works
well, and ran twice as fast as a reimplementation of the
method-of-multipliers I wrote.
I've heavily commented and restructured the code, added diagnostics,
added references, corrected confusing typo's, and made minor changes to
avoid globals (but as far as I can tell, the code was bug free, so
what's out there is safe to use), and created a developer-friently test
suite for it.
I still need to port the (substantial) documentation to texi, and
create some automatic validation to run as part of a build.
Rather than have many different versions floating around, I really want
to get an official and up-to-date version submitted as part of Octave.
If someone else would like to take up the challenge of releasing the
updated solnp into Octave, I would be happy to pass the baton: 3 .m
files, plain .tex file, and some notes -- I can't justify silently
sitting on it any longer, and it is still not at the top of my very long
to do list.
/Jskud
> From bug-request at octave dot org Wed May 11 17:08:33 2005
> Subject: Suggestion for improving Octave
> From: "John W. Eaton" <jwe at bevo dot che dot wisc dot edu>
> To: "Tech Nossomy" <technossomy at mail dot com>
> Cc: bug-octave at bevo dot che dot wisc dot edu
> Date: Wed, 11 May 2005 18:07:47 -0400
>
> On 11-May-2005, Tech Nossomy wrote:
>
> | Hi
> |
> | I have no experience to speak of in Octave or programming systems
> | like it, but going through your to-do list for Octave, it seems that
> | Octave might benefit from porting to Coin Symphony. Coin-Symphony is
> | a MIP solver that builds under C++/g++ and is published under GPL.
>
> I think that COIN is distributed under the terms of the CPL (Common
> Public License), not the GPL (GNU General Public License). The COIN
> FAQ at http://www.coin-or.org/faqs.html says this:
>
> Q: What license is used by COIN?
> A: All current COIN-OR components use the Common Public License. The
> CPL is approved and linked to by the Open Source Initiative. The
> author of the CPL is IBM, and they have a CPL FAQ, based on their
> understanding of the CPL.
>
> Although the CPL is a free software license, it is unfortunately not
> compatible with the GPL, so COIN cannot be combined with Octave, which
> is distributed under the terms of the GPL. For details about why the
> CPL and GPL are not compatible, see the entry for the CPL at
> http://www.gnu.org/licenses/license-list.html.
>
> Do you think it would be possible to have COIN released under terms
> that are compatible with the GPL?
>
> jwe
- Re: nonlinear optimization code (was Suggestion for improving Octave),
Jskud <=