Hello,
(Sorry it was my mistake - I initially thought that the .par files did not contain parameters from optimize(X, 0, 0, 0), but they actually do.)
I still have the following problem:
I am using a custom evaluation/orchestration setup that runs many optimization jobs with different configurations and custom names. As a result, I end up with many .par files, and from the .par file alone I cannot reliably determine which parameter corresponds to which optimize() call.
The order of parameters in .par file is not sufficient for me, because the orchestration can inject different optimization configurations, and
the injection order does not necessarily match the order of the optimize() calls in the script.
So the injector does not know the order of the optimize calls in the compiled startegy file.
I would be fine with writing this information from the script itself at the end of optimization, something like
if(is(ENDOFALLOPTIMIZE) && Train){
printf(... finalValue(variable) ...);
}
So what would be the recommended way to retrieve/save the mapping between the parameters in the .par file and their corresponding optimize() calls?
(I could also parse the generated HTML report, since it contains the variable names and values. However, with 1000 jobs, this would open 1000 browser windows. I would like to run the optimizations with Zorro S in headless mode, using only command-line commands, without opening any browser windows.)
Thank you!