|
CSAIndividual.py
1 import numpy as np
2 import ObjFunction
3
4
5 class CSAIndividual:
6
7 '''
8 individual of clone selection algorithm
9 '''
10
11 def __init__(self, vardim, bound):
12 '''
13 vardim: dimension of variables
14 bound: boundaries of variables
15 '''
16 self.vardim = vardim
17 self.bound = bound
18 self.fitness = 0.
19 self.trials = 0
20
21 def generate(self):
22 '''
23 generate a random chromsome for clone selection algorithm
24 '''
25 len = self.vardim
26 rnd = np.random.random(size=len)
27 self.chrom = np.zeros(len)
28 for i in xrange(0, len):
29 self.chrom = self.bound[0, i] + \
30 (self.bound[1, i] - self.bound[0, i]) * rnd
31
32 def calculateFitness(self):
33 '''
34 calculate the fitness of the chromsome
35 '''
36 self.fitness = ObjFunction.GrieFunc(
37 self.vardim, self.chrom, self.bound)
CSA.py
1 import numpy as np
2 from CSAIndividual import CSAIndividual
3 import random
4 import copy
5 import matplotlib.pyplot as plt
6
7
8 class CloneSelectionAlgorithm:
9
10 '''
11 the class for clone selection algorithm
12 '''
13
14 def __init__(self, sizepop, vardim, bound, MAXGEN, params):
15 '''
16 sizepop: population sizepop
17 vardim: dimension of variables
18 bound: boundaries of variables
19 MAXGEN: termination condition
20 params: algorithm required parameters, it is a list which is consisting of[beta, pm, alpha_max, alpha_min]
21 '''
22 self.sizepop = sizepop
23 self.vardim = vardim
24 self.bound = bound
25 self.MAXGEN = MAXGEN
26 self.params = params
27 self.population = []
28 self.fitness = np.zeros(self.sizepop)
29 self.trace = np.zeros((self.MAXGEN, 2))
30
31 def initialize(self):
32 '''
33 initialize the population of ba
34 '''
35 for i in xrange(0, self.sizepop):
36 ind = CSAIndividual(self.vardim, self.bound)
37 ind.generate()
38 self.population.append(ind)
39
40 def evaluation(self):
41 '''
42 evaluation the fitness of the population
43 '''
44 for i in xrange(0, self.sizepop):
45 self.population.calculateFitness()
46 self.fitness = self.population.fitness
47
48 def solve(self):
49 '''
50 the evolution process of the clone selection algorithm
51 '''
52 self.t = 0
53 self.initialize()
54 self.evaluation()
55 bestIndex = np.argmax(self.fitness)
56 self.best = copy.deepcopy(self.population[bestIndex])
57 while self.t < self.MAXGEN:
58 self.t += 1
59 tmpPop = self.reproduction()
60 tmpPop = self.mutation(tmpPop)
61 self.selection(tmpPop)
62 best = np.max(self.fitness)
63 bestIndex = np.argmax(self.fitness)
64 if best > self.best.fitness:
65 self.best = copy.deepcopy(self.population[bestIndex])
66
67 self.avefitness = np.mean(self.fitness)
68 self.trace[self.t - 1, 0] = \
69 (1 - self.best.fitness) / self.best.fitness
70 self.trace[self.t - 1, 1] = (1 - self.avefitness) / self.avefitness
71 print("Generation %d: optimal function value is: %f; average function value is %f" % (
72 self.t, self.trace[self.t - 1, 0], self.trace[self.t - 1, 1]))
73 print("Optimal function value is: %f; " % self.trace[self.t - 1, 0])
74 print "Optimal solution is:"
75 print self.best.chrom
76 self.printResult()
77
78 def reproduction(self):
79 '''
80 reproduction
81 '''
82 tmpPop = []
83 for i in xrange(0, self.sizepop):
84 nc = int(self.params[1] * self.sizepop)
85 for j in xrange(0, nc):
86 ind = copy.deepcopy(self.population)
87 tmpPop.append(ind)
88 return tmpPop
89
90 def mutation(self, tmpPop):
91 '''
92 hypermutation
93 '''
94 for i in xrange(0, self.sizepop):
95 nc = int(self.params[1] * self.sizepop)
96 for j in xrange(1, nc):
97 rnd = np.random.random(1)
98 if rnd < self.params[0]:
99 # alpha = self.params[
100 # 2] + self.t * (self.params[3] - self.params[2]) / self.MAXGEN
101 delta = self.params[2] + self.t * \
102 (self.params[3] - self.params[3]) / self.MAXGEN
103 tmpPop[i * nc + j].chrom += np.random.normal(0.0, delta, self.vardim)
104 # tmpPop[i * nc + j].chrom += alpha * np.random.random(
105 # self.vardim) * (self.best.chrom - tmpPop[i * nc +
106 # j].chrom)
107 for k in xrange(0, self.vardim):
108 if tmpPop[i * nc + j].chrom[k] < self.bound[0, k]:
109 tmpPop[i * nc + j].chrom[k] = self.bound[0, k]
110 if tmpPop[i * nc + j].chrom[k] > self.bound[1, k]:
111 tmpPop[i * nc + j].chrom[k] = self.bound[1, k]
112 tmpPop[i * nc + j].calculateFitness()
113 return tmpPop
114
115 def selection(self, tmpPop):
116 '''
117 re-selection
118 '''
119 for i in xrange(0, self.sizepop):
120 nc = int(self.params[1] * self.sizepop)
121 best = 0.0
122 bestIndex = -1
123 for j in xrange(0, nc):
124 if tmpPop[i * nc + j].fitness > best:
125 best = tmpPop[i * nc + j].fitness
126 bestIndex = i * nc + j
127 if self.fitness < best:
128 self.population = copy.deepcopy(tmpPop[bestIndex])
129 self.fitness = best
130
131 def printResult(self):
132 '''
133 plot the result of clone selection algorithm
134 '''
135 x = np.arange(0, self.MAXGEN)
136 y1 = self.trace[:, 0]
137 y2 = self.trace[:, 1]
138 plt.plot(x, y1, 'r', label='optimal value')
139 plt.plot(x, y2, 'g', label='average value')
140 plt.xlabel("Iteration")
141 plt.ylabel("function value")
142 plt.title("Clone selection algorithm for function optimization")
143 plt.legend()
144 plt.show()
运行程序:
1 if __name__ == "__main__":
2
3 bound = np.tile([[-600], [600]], 25)
4 csa = CSA(50, 25, bound, 500, [0.3, 0.4, 5, 0.1])
5 csa.solve()
ObjFunction见简单遗传算法-python实现。 |
|