Re: [Help-glpk] glpsol, arbitrary precision and large numbers

[Robbie Morrison]

------------------------------------------------------------
To:           Edd Barrett <address@hidden>
Subject:      Re: [Help-glpk] glpsol, arbitrary precision and large numbers
Message-ID:  <address@hidden>
From:         Edd Barrett <address@hidden>
Date:         Fri, 29 Jun 2012 10:42:35 +0100
------------------------------------------------------------

> On Wed, Jun 27, 2012 at 01:03:35PM +0100, Edd Barrett wrote:
>
>> Which is false, so under this assignment the system
>> in infeasible. The solver should have either tried a
>> different assignment of either variables, or if it
>> could not, then it should have reported the problem
>> infeasible? Right?
>
> Any thoughts on this? I am keen to figure out what went
> wrong here.
>
> Cheers
>
> --
> Best Regards
> Edd Barrett
>
> http://www.theunixzoo.co.uk

Hi Edd

No idea what the underlying problem is but here are
some debugging suggestions:

  * try formulating a minimum reproducible case and
    circulate that

  * check the KKT conditions if appropriate

    
http://en.wikibooks.org/wiki/GLPK/Solution_information#Karush-Kuhn-Tucker_optimality_conditions

  * export your MathProg model in MPS format and try
    solving that with GLPSOL or with other solvers

    http://en.wikibooks.org/wiki/GLPK/Interoperability#MPS_free_format

HTH, Robbie
---
Robbie Morrison
PhD student -- policy-oriented energy system simulation
Technical University of Berlin (TU-Berlin), Germany
University email (redirected) : address@hidden
Webmail (preferred)           : address@hidden
[from Webmail client]