Originally posted by: CplexUser1453

Hello

I am using OPL 6.3 and I have a stupid question on using an *.ops file :

By exectuting oplrun on the command line :

oplrun -tune params.ops model.mod data.dat

I got the following params.ops file :

<?xml version="1.0"?>
<settings>
<category name="cplex">
<setting name="heurfreq" value="3" type="int"/>
<setting name="rinsheur" value="20" type="int"/>
</category>
</settings>

However, when I try to execute my model with the .ops file :

oplrun model.mod data.dat params.ops

I get a syntax error :

ERRORDATA_001 at 1:1-2 params.ops: Data parsing error: syntax error, unexpected <.
FATALPARSE_103 at 1:2-3 params.ops: Unexpected character in "?", try using quoted string data, or check the encoding

What am I missing?

Thanks a lot!
#DecisionOptimization
#OPLusingCPLEXOptimizer

Hi,

if you do oplrun -h

you get

Usage:
oplrun [options] model-file [data-file ...]
oplrun [options] -p project-dir [run-configuration ...]

In order to use an ops file, you need to include it into a run configuration.
(With drag and drop in the IDE, this is quite easy)

You can also turn

<setting name="heurfreq" value="3" type="int"/>
<setting name="rinsheur" value="20" type="int"/>

into OPL scripting:

execute
{
 cplex.heurfreq=3;
 cplex.rinsheur=20;
}

regards
#DecisionOptimization
#OPLusingCPLEXOptimizer

Originally posted by: CplexUser1453

Hello

Thanks a lot for your response.

I am running OPL models on a Linux machine, that's why I am asking about command line (sometimes it is more pracical than the IDE).

So if I understand correctly, the only way to apply non default parameters, using command line, is to include them in the my model .mod file?

(I had a vague recollection that there was a way to use an *.ops file on command line...)
#DecisionOptimization
#OPLusingCPLEXOptimizer

Hi,

the Ide is now available on linux for some chips.
Are you aware of this?

Regards
#DecisionOptimization
#OPLusingCPLEXOptimizer

Originally posted by: CplexUser1453

Yes, I was using OPL 6.3 for now but I will upgrade to a more recent version. Thanks for your response
#DecisionOptimization
#OPLusingCPLEXOptimizer