ICSI'2022 Benchmark Functions

English | 中文

This is the official repository of the ICSI'2022 benchmark functions. We encourage all researchers to test their algorithms on the ICSI'2022 test suite. At the same time, we are holding an algorithm competition based on this test set. All researchers who study swarm intelligence algorithms and evolutionary algorithms are welcome to participate in this competition. We will provide generous bonuses as rewards for the excellent contestants in this competition.

Next we will tell you how to install and use the test set.

This repository contains:

1. Test functions for python.
2. Test functions for MATLAB.
3. The definition of each function in this test suite. ICSI_2022_Benchmark_functions.pdf

Table of Contents

• ICSI'2022 Benchmark Functions
  • Table of Contents
  • Background
  • Install
    • Installation for MATLAB
    • Installation for python
  • Usage
  • Definition of test functions
  • Related repository
  • Maintainers
  • License

Background

ICSI is the abbreviation of International Conference on Swarm Intelligence, which has successfully held 12 sessions so far. This conference is one of the top conferences in the field of swarm intelligence. The 13th International Swarm Intelligence Conference (ICSI’2022) will be held in Xi’an, China in July 2022. website of ICSI'2022

In order to make researchers who study swarm intelligence optimization algorithms and evolutionary algorithms better test their algorithms, we have designed a set of test functions to provide a better test platform. At the same time, we held an algorithm competition based on this test set. All researchers who swarm intelligence algorithms and evolutionary algorithms are welcome to participate in this competition. We will provide generous bonuses as rewards for the excellent contestants in this competition.

Install

This test suite provides python and MATLAB support. Please make sure you install them.

Installation for MATLAB

For MATLAB users, please use MATLAB and run following command to compile file icsi2022.cpp.

mex icsi2022.cpp -DWINDOWS

If you can run test.mwithout any mistake, the compilation is successful.

Note: Please make sure that the compiler supported by MATLAB has been installed in the system. For how to install and use the compiler in MATLAB, please refer to this website.

Installation for python

First make sure that the following dependencies are installed in the system

Python >= 3.6
Cython==0.29.23
matplotlib==3.3.4
numpy==1.20.1

For Linux users, please run the following command to compile and install

$ bash c.sh

For Windows users, please make sure you have a suitable compiler in your system. we recommend to use MinGW. You can also use Microsoft Visual C++ 14.0 or higher.

First run the following command to compile after the compiler has been installed in your system.

gcc -c icsi2022.c icsi2022.h
ar rcs libicsi2022.a icsi2022.o

Then run the following command to finish the compilation.

python setup.py build_ext --inplace

If you can run check.pywithout any mistake, the compilation is successful.

Usage

After successful installation, you can run a simple command to use this test set function.

Here is a simple example for MATLAB users

y = icsi2022(x, i)
% x is a m*D matrix
% i is from 1 to 10 which represent the serial number of 10 functions in icsi'2022 test suite.
% 

Here is a simple example for python users

import numpy as np
import icsi2022
# x is a m*D matrix

y=icsi2022.eval(x,i)
# i is from 0 to 9 which represent the serial number of 10 functions in icsi'2022 test suite.

Definition of test functions

Please refer to the documentation icsi2022.pdf to obtain the information of test functions.

Related repository

• Fireworks Algorithm — The official repository for Fireworks Algorithm.

Maintainers

@moshangsang24

If you have any problems during use, you can contact me via email [email protected] or directly submit issue on github.

License

MIT © CIL-Lab