help-glpk
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Help-glpk] [Fwd: Re: [Fwd: Re: [Fwd: graceful tree labeling example]]]


From: Andrew Makhorin
Subject: [Help-glpk] [Fwd: Re: [Fwd: Re: [Fwd: graceful tree labeling example]]]
Date: Fri, 06 Oct 2017 22:21:05 +0300

-------- Forwarded Message --------
From: Mike Appleby <address@hidden>
To: Michael Hennebry <address@hidden>
Cc: Heinrich Schuchardt <address@hidden>, address@hidden
Subject: Re: [Help-glpk] [Fwd: Re: [Fwd: graceful tree labeling
example]]
Date: Fri, 6 Oct 2017 11:13:17 -0500

On Fri, Oct 6, 2017 at 9:36 AM, Michael Hennebry
<address@hidden> wrote:

> Constraint-enforced integer might be a useful variable type.
> The solver would not need to branch on them,
> but could use their integrality for things like Gomory cuts.

Indeed, if I select the right combination of options for
branching/backtracking/cuts, then I usually get comparable solution
times for the all-integer version and the version that uses floats for
vertex and edge labels. But all my input trees are small, and I only
search a subset of the mip-related options, so I might be skipping the
options necessary to speed up the float version.

As you can probably tell, I have no understanding of how the solver
works or what the options do. I am a monkey poking it with a stick :).






reply via email to

[Prev in Thread] Current Thread [Next in Thread]