精英保存策略的遗传算法代码

以下是一个简单的精英保存策略的遗传算法代码：

import random

# 定义适应度函数
def fitness_function(individual):
    return sum(individual)

# 定义遗传算法的主函数
def genetic_algorithm(population_size, gene_size, generations, elitism_rate, mutation_rate):
    # 初始化种群
    population = [[random.randint(0, 1) for j in range(gene_size)] for i in range(population_size)]

    # 迭代进化
    for i in range(generations):
        # 计算适应度
        fitness = [fitness_function(individual) for individual in population]

        # 选择精英个体
        elite_size = int(elitism_rate * population_size)
        elite_index = sorted(range(len(fitness)), key=lambda k: fitness[k], reverse=True)[:elite_size]
        elite_individuals = [population[i] for i in elite_index]

        # 选择非精英个体
        non_elite_individuals = [population[i] for i in range(population_size) if i not in elite_index]

        # 交叉操作
        offspring = []
        while len(offspring) < population_size - elite_size:
            parent1 = random.choice(non_elite_individuals)
            parent2 = random.choice(non_elite_individuals)
            offspring1 = parent1[:gene_size // 2] + parent2[gene_size // 2:]
            offspring2 = parent2[:gene_size // 2] + parent1[gene_size // 2:]
            offspring.append(offspring1)
            offspring.append(offspring2)

        # 变异操作
        for individual in offspring:
            for j in range(gene_size):
                if random.random() < mutation_rate:
                    individual[j] = 1 - individual[j]

        # 更新种群
        population = elite_individuals + offspring

    # 返回最优个体
    fitness = [fitness_function(individual) for individual in population]
    best_index = fitness.index(max(fitness))
    return population[best_index]

其中，`population_size` 表示种群大小，`gene_size` 表示基因个数，`generations` 表示迭代次数，`elitism_rate` 表示精英保留率，`mutation_rate` 表示变异率。在主函数中，首先初始化种群，然后迭代进化。在每次迭代中，先计算适应度，然后选择精英个体和非精英个体。接着进行交叉和变异操作，生成新的后代个体。最后更新种群并返回最优个体。