Differential evolution vs particle swarm optimization

Clone this repo:

Branches

  1. 73159e7 update readme by BradleyWood · 6 years ago master
  2. becd775 update readme by BradleyWood · 6 years ago
  3. 459409b added result tables by BradleyWood · 6 years ago
  4. fdf49af added gifs by BradleyWood · 6 years ago
  5. 8b2a901 refactoring - moved class by BradleyWood · 6 years ago

DE, PSO Comparative study

The purpose of the project was to implement both the PSO and DE algorithms and perform a comparative study.

Build instructions

mvn clean compile assembly:single

Running

You must specify the algorithm to test by using -de or -pso as a command line argument.

java -jar <path to jar>.jar <-de or -pso> -display <Benchmark name>

To run the benchmarks and output the test results

java -jar <path to jar>.jar <-de or -pso> -test

or

java -jar <path to jar>.jar <-de or -pso> -test <Benchmark name>

Benchmark Examples

Differential Evolution

Ackley
alt text

Bent Cigar
alt text

Discus
alt text

Griewank
alt text

Rastrigin
alt text

Rosenbrock
alt text

Particle swarm optimization

Ackley
alt text

Bent Cigar
alt text

Discus
alt text

Griewank
alt text

Rastrigin
alt text

Rosenbrock
alt text