#9617: Broken export of LP files....
-------------------------+--------------------------------------------------
Reporter: ncohen | Owner: jason, jkantor
Type: defect | Status: needs_review
Priority: major | Milestone: sage-4.5.2
Component: numerical | Keywords: CPLEX, GLPK, glpsol, MPS, file
Author: | Upstream: N/A
Reviewer: | Merged:
Work_issues: |
-------------------------+--------------------------------------------------
Comment(by leif):
The good news: Your patch here seems to work fine.
The bad news: I've again exported (now valid) MPS and LP files for the
{{{edge_disjoint_spanning_trees()}}} example with 30 vertices and run
{{{glpsol}}} on those files on both a 64-bit machine and on the 32-bit
machine (P4 Prescott) where that example did not terminate. The logs are
slightly different, but ''both terminate'':
{{{
GLPSOL: GLPK LP/MIP Solver, v4.44
Parameter(s) specified in the command line:
--freemps edst-30_vertices.mps
Reading problem data from `edst-30_vertices.mps'...
edst-30_vertices.mps:8: warning: missing model name in field 3
Objective: R0000000
1720 rows, 3650 columns, 10040 non-zeros
1370 integer variables, all of which are binary
8412 records were read
GLPK Integer Optimizer, v4.44
1720 rows, 3650 columns, 10040 non-zeros
1370 integer variables, all of which are binary
Preprocessing...
1670 rows, 3605 columns, 9790 non-zeros
1325 integer variables, all of which are binary
Scaling...
A: min|aij| = 1.000e+00 max|aij| = 1.000e+00 ratio = 1.000e+00
Problem data seem to be well scaled
Constructing initial basis...
Size of triangular part = 1670
Solving LP relaxation...
GLPK Simplex Optimizer, v4.44
1670 rows, 3605 columns, 9790 non-zeros
0: obj = 0.000000000e+00 infeas = 2.610e+02 (0)
500: obj = 0.000000000e+00 infeas = 2.610e+02 (0)
1000: obj = 0.000000000e+00 infeas = 2.330e+02 (0)
1500: obj = 0.000000000e+00 infeas = 1.172e+00 (0)
* 1589: obj = 0.000000000e+00 infeas = 2.132e-14 (0)
OPTIMAL SOLUTION FOUND
Integer optimization begins...
+ 1589: mip = not found yet >= -inf (1; 0)
+ 4531: mip = not found yet >= 0.000000000e+00 (26; 0)
+ 8462: mip = not found yet >= 0.000000000e+00 (62; 0)
+ 14689: mip = not found yet >= 0.000000000e+00 (106; 0)
+ 28430: >>>>> 0.000000000e+00 >= 0.000000000e+00 0.0% (223; 3)
+ 28430: mip = 0.000000000e+00 >= tree is empty 0.0% (0; 451)
INTEGER OPTIMAL SOLUTION FOUND
Time used: 19.1 secs
Memory used: 5.6 Mb (5848196 bytes)
}}}
(64-bit machine, time due to high sysload)
{{{
GLPSOL: GLPK LP/MIP Solver, v4.44
Parameter(s) specified in the command line:
--freemps /home/leif/Sage/new-mps-lp/edst-30_vertices.mps
Reading problem data from `/home/leif/Sage/new-mps-lp/edst-
30_vertices.mps'...
/home/leif/Sage/new-mps-lp/edst-30_vertices.mps:8: warning: missing model
name in field 3
Objective: R0000000
1720 rows, 3650 columns, 10040 non-zeros
1370 integer variables, all of which are binary
8412 records were read
GLPK Integer Optimizer, v4.44
1720 rows, 3650 columns, 10040 non-zeros
1370 integer variables, all of which are binary
Preprocessing...
1670 rows, 3605 columns, 9790 non-zeros
1325 integer variables, all of which are binary
Scaling...
A: min|aij| = 1.000e+00 max|aij| = 1.000e+00 ratio = 1.000e+00
Problem data seem to be well scaled
Constructing initial basis...
Size of triangular part = 1670
Solving LP relaxation...
GLPK Simplex Optimizer, v4.44
1670 rows, 3605 columns, 9790 non-zeros
0: obj = 0.000000000e+00 infeas = 2.610e+02 (0)
500: obj = 0.000000000e+00 infeas = 2.610e+02 (0)
1000: obj = 0.000000000e+00 infeas = 2.330e+02 (0)
1500: obj = 0.000000000e+00 infeas = 1.172e+00 (0)
* 1589: obj = 0.000000000e+00 infeas = 2.132e-14 (0)
OPTIMAL SOLUTION FOUND
Integer optimization begins...
+ 1589: mip = not found yet >= -inf (1; 0)
+ 3426: mip = not found yet >= 0.000000000e+00 (10; 0)
+ 4687: mip = not found yet >= 0.000000000e+00 (21; 0)
+ 5604: mip = not found yet >= 0.000000000e+00 (32; 0)
+ 6640: mip = not found yet >= 0.000000000e+00 (44; 0)
+ 7802: mip = not found yet >= 0.000000000e+00 (58; 0)
+ 9123: mip = not found yet >= 0.000000000e+00 (72; 0)
+ 11067: mip = not found yet >= 0.000000000e+00 (86; 0)
+ 13203: mip = not found yet >= 0.000000000e+00 (101; 0)
+ 15411: mip = not found yet >= 0.000000000e+00 (117; 1)
+ 18254: mip = not found yet >= 0.000000000e+00 (135; 1)
+ 21870: mip = not found yet >= 0.000000000e+00 (156; 1)
Time used: 60.1 secs. Memory used: 3.8 Mb.
+ 26285: mip = not found yet >= 0.000000000e+00 (186; 4)
+ 29923: >>>>> 0.000000000e+00 >= 0.000000000e+00 0.0% (253; 15)
+ 29923: mip = 0.000000000e+00 >= tree is empty 0.0% (0; 535)
INTEGER OPTIMAL SOLUTION FOUND
Time used: 67.4 secs
Memory used: 4.9 Mb (5146500 bytes)
}}}
(32-bit machine/Pentium 4 Prescott)
That means the bug causing the non-termination sits somewhere in the Sage
library interface to GLPK.
All files were generated on the 64-bit machine; I've not yet tried to
generate them on the 32-bit machine (they might differ, so I'll check that
later).
(The LP/MPS files of that example are too big to upload them here, approx.
260KB.)
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/9617#comment:3>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--
You received this message because you are subscribed to the Google Groups
"sage-trac" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/sage-trac?hl=en.
