|
From: | Domingo Alvarez Duarte |
Subject: | Re: MPS files for maximization lack OBJSENSE |
Date: | Sat, 14 May 2022 16:13:54 +0200 |
User-agent: | Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101 Thunderbird/91.8.1 |
Hello Julian !
Meanwhile I implemented read/write OBJSENSE for freemps format on my fork of GLPK with several extensions and here https://github.com/mingodad/GLPK/actions/runs/2324178367 you can find the binaries built with github actions to try.
With your proposed example the solution is the same on both, and here is the generated freemps:
====
* Problem: simple_max
* Class: LP
* Rows: 3
* Columns: 2
* Non-zeros: 6
* Format: Free MPS
*
NAME simple_max
OBJSENSE MAX
ROWS
L Constraint1
L Constraint2
N Objective
COLUMNS
x Constraint1 1 Constraint2 3
x Objective 1
y Constraint1 2 Constraint2 1
y Objective 1
RHS
RHS1 Constraint1 15 Constraint2 20
ENDATA
====
Cheers !
When using glpsol (GLPSOL: GLPK LP/MIP Solver, v4.65) for the attached .mod file thus
glpsol -m simple_max.mod --w
freemps simple_max.mps
I get the (correct) output
0: obj = -0.000000000e+00 inf = 0.000e+00 (2)
2: obj = 1.000000000e+01 inf = 0.000e+00 (0)
OPTIMAL LP SOLUTION FOUND
but the (attached) MPS file is set up for the default sense (minimization). Thus, when running
glpsol simple_max.mps
I get the output
0: obj = 0.000000000e+00 inf = 0.000e+00 (0)
OPTIMAL LP SOLUTION FOUND
The same happens if --wmps is used.
When writing out the MPS file glpsol must either
- Use the OBJSENSE section (preferably)
OBJSENSE
MAX
- Negate the objective coefficients - although this yields the wrong sign of the objective function and duals
How can this bug in glpsol be fixed?
Thanks,
JulianThe University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. Is e buidheann carthannais a th’ ann an Oilthigh Dhùn Èideann, clàraichte an Alba, àireamh clàraidh SC005336.
--
Dr. J. A. Julian Hall, Reader, School of Mathematics,
University of Edinburgh, James Clerk Maxwell Building,
Peter Guthrie Tait Road, EDINBURGH, EH9 3FD, UK.
Room: 5418 Phone: [+44](131) 650 5075 Email: J.A.J.Hall@ed.ac.uk
Web: https://www.maths.ed.ac.uk/school-of-mathematics/people/a-z?person=47
![]()
My working hours may not be your working hours. Do not feel pressure to reply to this email outside your working hours.
[Prev in Thread] | Current Thread | [Next in Thread] |