上演化计算课的时候老师讲了一种DE算法的改进算法CoDE,于是看了下CoDE的论文中的算法步骤：
算法中使用的三种交叉策略：
根据不同的交叉策略采取不同的变异策略：
超参数的三种不同组合（一一对应）：
python代码实现
import numpy as np
import random
import math class Population:
def __init__(self, min_range, max_range, dim, rounds, size, object_func):
self.min_range = min_range
self.max_range = max_range
self.dimension = dim
self.rounds = rounds
self.size = size
self.cur_round = 1
self.get_object_function_value = object_func
# 初始化种群
self.individuality = [np.array([random.uniform(self.min_range, self.max_range) for s in range(self.dimension)])
for tmp in range(size)]
self.object_function_values = [self.get_object_function_value(v) for v in self.individuality]
self.mutant1 = None
self.mutant2 = None
self.mutant3 = None
#变异 def mutate1(self): #使用rand/1进行变异
self.mutant1 = []
factor = 1.0
for i in range(self.size):
r0, r1, r2 = 0, 0, 0
#随机查找变异父代
while r0 == r1 or r1 == r2 or r0 == r2 or r0 == i:
r0 = random.randint(0, self.size - 1)
r1 = random.randint(0, self.size - 1)
r2 = random.randint(0, self.size - 1)
#变异向量
tmp = self.individuality[r0] + (self.individuality[r1] - self.individuality[r2]) * factor
for t in range(self.dimension):
if tmp[t] > self.max_range or tmp[t] < self.min_range:
tmp[t] = random.uniform(self.min_range, self.max_range)
self.mutant1.append(tmp) def mutate2(self): #使用rand/2进行变异
self.mutant2 = []
factor = 1.0
for i in range(self.size):
r0, r1, r2 , r3, r4= 0, 0, 0, 0, 0
#随机查找变异父代
while r0 == r1 or r1 == r2 or r2 == r3 or r3 == r4 or r4 == r0 \
or r0 == r2 or r0 == r3 or r0 == r4 or r1 == r3 or r1 == r4 \
or r2 == r4 or r0 == i:
r0 = random.randint(0, self.size - 1)
r1 = random.randint(0, self.size - 1)
r2 = random.randint(0, self.size - 1)
r3 = random.randint(0, self.size - 1)
r4 = random.randint(0, self.size - 1)
#变异向量
tmp = self.individuality[r0] + (self.individuality[r1] - self.individuality[r2]) * factor\
+ (self.individuality[r3] - self.individuality[r4]) * factor
for t in range(self.dimension):
if tmp[t] > self.max_range or tmp[t] < self.min_range:
tmp[t] = random.uniform(self.min_range, self.max_range)
self.mutant2.append(tmp) def mutate3(self): #使用current-to-rand/1进行变异
self.mutant3 = []
factor = 0.8
for i in range(self.size):
r0, r1, r2 = 0, 0, 0
# 随机查找变异父代
while r0 == r1 or r1 == r2 or r0 == r2 or r0 == i:
r0 = random.randint(0, self.size - 1)
r1 = random.randint(0, self.size - 1)
r2 = random.randint(0, self.size - 1)
# 变异向量
tmp = self.individuality[i] + (self.individuality[r0] - self.individuality[i]) * factor \
+ (self.individuality[r1] - self.individuality[r2]) * factor
for t in range(self.dimension):
if tmp[t] > self.max_range or tmp[t] < self.min_range:
tmp[t] = random.uniform(self.min_range, self.max_range)
self.mutant3.append(tmp) #交叉
def crossover(self,mutant,CR):
for i in range(self.size):
Jrand = random.randint(0, self.dimension)
for j in range(self.dimension):
if random.random() > CR and j != Jrand:
mutant[i][j] = self.individuality[i][j]
return mutant #选择
def select(self,u1,u2,u3):
for i in range(self.size):
tmp1 = self.get_object_function_value(u1[i])
tmp2 = self.get_object_function_value(u2[i])
tmp3 = self.get_object_function_value(u3[i])
if tmp1 < tmp2 and tmp1 < tmp3 and tmp1 < self.object_function_values[i]:
self.individuality[i] = u1[i]
self.object_function_values[i] = tmp1
elif tmp2 < tmp1 and tmp2 < tmp3 and tmp2 < self.object_function_values[i]:
self.individuality[i] = u2[i]
self.object_function_values[i] = tmp2
elif tmp3 < tmp1 and tmp3 < tmp2 and tmp3 < self.object_function_values[i]:
self.individuality[i] = u3[i]
self.object_function_values[i] = tmp3 def print_best(self):
m = min(self.object_function_values)
i = self.object_function_values.index(m)
print("轮数：" + str(self.cur_round))
print("最佳个体：" + str(self.individuality[i]))
print("目标函数值：" + str(m)) def evolution(self):
while self.cur_round < self.rounds:
self.mutate1()
self.mutate2()
self.mutate3()
u1 = self.crossover(self.mutant1,CR=0.1)
u2 = self.crossover(self.mutant2,CR=0.9)
u3 = self.mutant3
self.select(u1,u2,u3)
self.print_best()
self.cur_round = self.cur_round + 1 # 测试部分
if __name__ == "__main__":
def f(x):
return x**2
p = Population(min_range=-3, max_range=3, dim=1, rounds=100, size=30, object_func=f)
p.evolution()
---***如有不对欢迎指正***---
