A set of benchmark function

A library to support the benchmarking of functions for optimization evaluation, similar to algorithm-test.

Description

This is a benchmark function group for optimization algorithm evaluation.

All benchmark function (test function)

• Multimodal function

  • One global optimum solution

    • Ackley function
    • Beale function
    • Goldstein-Price function
    • Bukin function N.6
    • Levi function N.13
    • Three-hump camel function
    • Easom function
    • Eggholder function
    • McCormick function
    • Schaffer function N.2
    • Schaffer function N.4
    • Styblinski-Tang function
    • De Jong's funtion F5
    • Five-well potential function
    • Griewank function
    • Michalewicz function
    • Rastrigin function
    • Schwefel function
    • Six-hump camel function
    • Xin-She Yang function

  • Multiple global optimal solutions

    • Shuberts function

• Unimodal function

  • One global optimum solution
    • Sphere function
    • Rosenbrock function
    • Booth function
    • Matyas function
    • De Jong's function F1
    • De Jong's function F2
    • De Jong's function F3
    • De Jong's function F4
    • Ellipsoid function
    • k-tablet function
    • Weighted Sphere function
    • hyper ellipsodic function
    • Sum of different power function
    • Perm function
    • Zakharov function

Reference

• Test functions for optimization(en:Wikipedia)
• 最適化アルゴリズムを評価するベンチマーク関数まとめ

Features

• Easy to use many benchmark function

• Future

  • To implement test code to use evaluate algorithm

Installation/Requirement

python 3.XX

$ pip install git+https://github.com/keit0222/optimization-evaluation.git

upgrade library

$ pip install --upgrade git+https://github.com/keit0222/optimization-evaluation.git

Usage

example

$ python example.py

Author/Contributors

qiita: @tomochiii

License

This software is released under the MIT License, see LICENSE.txt.