Open HeleNoir opened 2 years ago
I came across an issue: the COCO bindings seem to not include the final_target_fvalue1
function, which returns the known optimum. While the optimizer should not know this value and this is not a problem for benchmarking, we still need the value for our experimental analysis as it includes additional comparisons.
@luleyleo Could you please look into it?
In addition, there is #138 and I would like to somehow log problem-related values, e.g. the known optimum or if the target has been hit. I will try to solve this myself, but I might need advice at some point.
To properly develop the evaluation, suitable examples are of help. I'm thinking of using
examples\coco.rs
as a common template and adding an example that will be used in the Tech Report.