遗传算法python实例

以下是一个简单的遗传算法 Python 实现示例：

import random

# 遗传算法参数
POPULATION_SIZE = 100
PARENTS_SIZE = 50
GENERATIONS = 100
MUTATION_RATE = 0.05

# 目标函数：计算x的平方根
def fitness(x):
   return x ** 0.5

# 随机生成种群
def generate_population(size):
   population = []
   for i in range(size):
      individual = {'x': random.uniform(0, 100)}
      individual['fitness'] = fitness(individual['x'])
      population.append(individual)
   return population

# 选择父母
def select_parents(population, size):
   parents = []
   for i in range(size):
      candidates = random.sample(population, 3)
      candidates.sort(key=lambda x: x['fitness'], reverse=True)
      parents.append(candidates[0])
   return parents

# 交叉
def crossover(parents):
   offspring = []
   for i in range(len(parents)):
      parent1 = parents[i]
      parent2 = parents[(i + 1) % len(parents)]
      child = {'x': (parent1['x'] + parent2['x']) / 2.0}
      child['fitness'] = fitness(child['x'])
      offspring.append(child)
   return offspring

# 变异
def mutate(individual):
   if random.random() < MUTATION_RATE:
      individual['x'] = random.uniform(0, 100)
      individual['fitness'] = fitness(individual['x'])

# 遗传算法主程序
def run_ga():
   population = generate_population(POPULATION_SIZE)
   for i in range(GENERATIONS):
      parents = select_parents(population, PARENTS_SIZE)
      offspring = crossover(parents)
      for child in offspring:
         mutate(child)
      population = parents + offspring
      population.sort(key=lambda x: x['fitness'], reverse=True)
      population = population[:POPULATION_SIZE]
      print('Generation %d: Best fitness = %.2f' % (i, population[0]['fitness']))

   return population[0]['x']

# 运行遗传算法
best_x = run_ga()
print('Best solution: x = %.2f, fitness = %.2f' % (best_x, fitness(best_x)))

该遗传算法的目标是最大化函数`fitness(x) = x ** 0.5`，即计算`x`的平方根。在每一代中，它随机生成一个种群，然后选择父母进行交叉产生下一代，并使用变异操作添加随机性。该算法在每一代中保留最适应的个体，并在每一代结束时输出最佳适应度。在最后一代中，它输出最佳解决方案（`x`的值）和相应的适应度。