search

numbbo / coco-experiment

Experimental code of the COCO framework
Other
2 stars 1 forks source link

Inflation of logger files #14

Open ttusar opened 1 week ago

ttusar commented 1 week ago

Currently, the observer creates several files for each problem: dat, tdat, rdat, mdat (and adat in the bi-objective case). However, rdat and mdat are often empty.

Is this an issue? Should/can we do something about it?

brockho commented 1 week ago

I agree that it would be nicer to have (c)leaner output folders. The only place where to do something about it, is most likely the logger itself, maybe deleting the file if it turns out that no actual data was written? As @nikohansen said yesterday, the .rdat files are anyway of no use with the newest exampleexperiment...