Skip to content

Numpy implementation of the Noiseless Functions of the Black-Box Optimization Benchmarking Suite

Notifications You must be signed in to change notification settings

maxhuettenrauch/bbob_numpy

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

15 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

bbob_numpy

This repository contains numpy implementations of the functions found in

https://coco.gforge.inria.fr/downloads/download16.00/bbobdocfunctions.pdf

which allow accessing current x_opt and f_opt.

It's a bit work in progress, so I put some TODOs into the code. Feel free to extend or correct them. When plotting some functions in 2D they looked different than the plots in the pdf, so there may be some errors in implementation.

The file bbobbenchmarks.py is taken from the pycma package (https://github.com/CMA-ES/pycma) with some modifications.

How to use

In order to create a suite with all objective functions in dimension 2 with 2 instances each do

test_suite_options = {'name': 'full',
                      'dim': [2],
                      'n_instances': 2}
test_suite = Suite(test_suite_options)

# query the optimal function value
for p in test_suite:
    print(p.f_obj.f_opt)

About

Numpy implementation of the Noiseless Functions of the Black-Box Optimization Benchmarking Suite

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages