emplp10.gms : Poorly-scaled model - small matrix entry implies large x using EMP

Description

This model tests lp10 from the Test Library with solver EMP. In the first test
EMP generates a standard scalar model which is solved by the default lp solver.
For the second test EMP rewrites the model as MCP and solves it with the default
MCP solver.

Contributor: Jan-H. Jagla, March 2007


Small Model of Type : GAMS


Category : GAMS Test library


Main file : emplp10.gms

$title 'Poorly-scaled model - small matrix entry implies large x using EMP' (EMPLP10,SEQ=382)

$ontext

This model tests lp10 from the Test Library with solver EMP. In the first test
EMP generates a standard scalar model which is solved by the default lp solver.
For the second test EMP rewrites the model as MCP and solves it with the default
MCP solver.

Contributor: Jan-H. Jagla, March 2007

$offtext

$set modelname lp10

*use default EMP to solve lp10
$log !!!!!!! try modeltype EMP !!!!!!!
$call gams %modelname% lo=%gams.lo% --MTYPE=emp
$if errorlevel 1 $abort Default EMP failed on %modelname%

*use EMP with NLP2MCP reformulation to solve lp10
$echo ModelType MCP           > empinfo.txt
$echo EmpInfoFile empinfo.txt > jams.opt

$log !!!!!!! try modeltype EMP with optfile=1 !!!!!!!
$call gams %modelname% lo=%gams.lo% optfile=1 --solvedasmcp=1 --MTYPE=emp
$if errorlevel 1 $abort EMP with MCP reformulation failed on %modelname%