DISH_python

Implementation of the DISH algorithm in Python.

Full paper

Usage

Sample use can also be seen at the end of the file main.py.

dim = 10 #dimension size of the optimized problem
NP = round(25 * math.log(dim) * math.sqrt(dim)) #population size (recomended setting)
maxFEs = 5000 #maximum number of objective function evaluations
H = 5 #archive size
minPopSize = 4

sphere = Sphere(dim) #defined test function
de = DISH(dim, maxFEs, sphere, H, NP, minPopSize) #initialize DISH
resp = de.run() #run the optimization
print(resp) #print the results

Output resp then includes optimized values features and value of objective function ofv. Also, the id of particle is included.

File descriptions

• main.py
  • The main file contains the main class DISH and one sample test function class Sphere.