ESIndividual.py
import numpy as np
import ObjFunction class ESIndividual: '''
individual of evolutionary strategy
''' def __init__(self, vardim, bound):
'''
vardim: dimension of variables
bound: boundaries of variables
'''
self.vardim = vardim
self.bound = bound
self.fitness = 0.
self.trials = 0 def generate(self):
'''
generate a random chromsome for evolutionary strategy
'''
len = self.vardim
rnd = np.random.random(size=len)
self.chrom = np.zeros(len)
for i in xrange(0, len):
self.chrom[i] = self.bound[0, i] + \
(self.bound[1, i] - self.bound[0, i]) * rnd[i] def calculateFitness(self):
'''
calculate the fitness of the chromsome
'''
self.fitness = ObjFunction.GrieFunc(
self.vardim, self.chrom, self.bound)
ES.py
import numpy as np
from ESIndividual import ESIndividual
import random
import copy
import matplotlib.pyplot as plt class EvolutionaryStrategy: '''
the class for evolutionary strategy
''' def __init__(self, sizepop, vardim, bound, MAXGEN, params):
'''
sizepop: population sizepop
vardim: dimension of variables
bound: boundaries of variables
MAXGEN: termination condition
params: algorithm required parameters, it is a list which is consisting of[delta_max, delta_min]
'''
self.sizepop = sizepop
self.vardim = vardim
self.bound = bound
self.MAXGEN = MAXGEN
self.params = params
self.population = []
self.fitness = np.zeros(self.sizepop)
self.trace = np.zeros((self.MAXGEN, 2)) def initialize(self):
'''
initialize the population of es
'''
for i in xrange(0, self.sizepop):
ind = ESIndividual(self.vardim, self.bound)
ind.generate()
self.population.append(ind) def evaluation(self):
'''
evaluation the fitness of the population
'''
for i in xrange(0, self.sizepop):
self.population[i].calculateFitness()
self.fitness[i] = self.population[i].fitness def solve(self):
'''
the evolution process of the evolutionary strategy
'''
self.t = 0
self.initialize()
self.evaluation()
bestIndex = np.argmax(self.fitness)
self.best = copy.deepcopy(self.population[bestIndex])
while self.t < self.MAXGEN:
self.t += 1
tmpPop = self.mutation()
self.selection(tmpPop)
best = np.max(self.fitness)
bestIndex = np.argmax(self.fitness)
if best > self.best.fitness:
self.best = copy.deepcopy(self.population[bestIndex]) self.avefitness = np.mean(self.fitness)
self.trace[self.t - 1, 0] = \
(1 - self.best.fitness) / self.best.fitness
self.trace[self.t - 1, 1] = (1 - self.avefitness) / self.avefitness
print("Generation %d: optimal function value is: %f; average function value is %f" % (
self.t, self.trace[self.t - 1, 0], self.trace[self.t - 1, 1]))
print("Optimal function value is: %f; " % self.trace[self.t - 1, 0])
print "Optimal solution is:"
print self.best.chrom
self.printResult() def mutation(self):
'''
mutate the population by a random normal distribution
'''
tmpPop = []
for i in xrange(0, self.sizepop):
ind = copy.deepcopy(self.population[i])
delta = self.params[0] + self.t * \
(self.params[1] - self.params[0]) / self.MAXGEN
ind.chrom += np.random.normal(0.0, delta, self.vardim)
for k in xrange(0, self.vardim):
if ind.chrom[k] < self.bound[0, k]:
ind.chrom[k] = self.bound[0, k]
if ind.chrom[k] > self.bound[1, k]:
ind.chrom[k] = self.bound[1, k]
ind.calculateFitness()
tmpPop.append(ind)
return tmpPop def selection(self, tmpPop):
'''
update the population
'''
for i in xrange(0, self.sizepop):
if self.fitness[i] < tmpPop[i].fitness:
self.population[i] = tmpPop[i]
self.fitness[i] = tmpPop[i].fitness def printResult(self):
'''
plot the result of evolutionary strategy
'''
x = np.arange(0, self.MAXGEN)
y1 = self.trace[:, 0]
y2 = self.trace[:, 1]
plt.plot(x, y1, 'r', label='optimal value')
plt.plot(x, y2, 'g', label='average value')
plt.xlabel("Iteration")
plt.ylabel("function value")
plt.title("Evolutionary strategy for function optimization")
plt.legend()
plt.show()
运行程序：
if __name__ == "__main__":
bound = np.tile([[-600], [600]], 25)
es = ES(60, 25, bound, 1000, [10, 1])
es.solve()ObjFunction见简单遗传算法-python实现。