[Coin-osi-devel] More params, glpk 4.9, mip gap

Kendall Bailey krbailey at gmail.com
Thu Apr 6 15:26:45 EDT 2006


I'm preparing to submit an AMPL driver program to the COIN project,
based on OSI solvers.  I've added a boolean attribute to the OsiGlpk
class to select the branch and bound routine to invoke.  My goal is to
simply expose as much solver functionality to AMPL as I can.  I've
made the AMPL driver (I'm calling it 'coinamp' similar to ILOGs
'cplexamp') makefile able to disable the use of the new boolean, so
when I submit the code others can build it against a stock OsiGlpk. 
Do you have plans to make the lpx_intopt() function callable from
OsiGlpk's branchAndBound()?  Do you want a patch?  Is lpx_intopt()
always better than lpx_integer() and OsiGlpk should just switch over?

Thanks,
Kendall Bailey


On 3/16/06, Brady Hunsaker <hunsaker at engr.pitt.edu> wrote:

> In the specific case of the two integer optimization routines, I haven't
> yet looked at updating OsiGlpk for GLPK 4.9.  It may make sense to
> change the branchAndBound call so that the new routine is the default.
> I know this doesn't address the question of allowing users their choice
> of routines, however.
>
> Brady
>




More information about the Osi mailing list