写一个遗传算法，用python

编写一个简单的遗传算法的Python代码示例如下：

import random

# 假设我们要解决的是0-1背包问题，定义一个个体（解）
def create_population(size):
    return [random.choices([0, 1], k=num_items) for _ in range(size)]

# 计算每个个体的适应度（目标函数）
def fitness(individual):
    total_value = sum(item * value for item, value in zip(individual, values))
    total_weight = sum(weight * item for item, weight in zip(individual, weights))
    return total_value / total_weight if total_weight > 0 else float('inf')

# 选择操作
def selection(population, fitness_values):
    return random.choices(population, weights=[fitness_val for fitness_val in fitness_values])

# 交叉操作（这里用单点交叉）
def crossover(parent1, parent2):
    cut_point = random.randint(1, len(parent1)-1)
    child1 = parent1[:cut_point] + parent2[cut_point:]
    child2 = parent2[:cut_point] + parent1[cut_point:]
    return child1, child2

# 变异操作
def mutate(individual):
    for i in range(len(individual)):
        if random.random() < mutation_rate:
            individual[i] = 1 - individual[i]
    return individual

# 主循环
def genetic_algorithm(population_size, num_generations, crossover_rate, mutation_rate):
    population = create_population(population_size)
    
    for _ in range(num_generations):
        fitness_values = [fitness(indiv) for indiv in population]
        next_generation = [selection(population, fitness_values)]
        
        while len(next_generation) < population_size:
            parents = random.sample(population, 2)
            children = crossover(parents[0], parents[1])
            children = [mutate(child) for child in children]
            next_generation.extend(children)
        
        population = next_generation
    
    # 返回最适应的解
    return max(population, key=fitness)

# 示例参数
num_items, values, weights = 5, [10, 20, 30, 40, 50], [1, 2, 3, 4, 5]
population_size = 100
num_generations = 100
crossover_rate = 0.8
mutation_rate = 0.01

solution = genetic_algorithm(population_size, num_generations, crossover_rate, mutation_rate)
print("最优解：", solution)

请注意，这只是一个基础版本，实际应用中可能需要根据问题特性进行适当修改。同时，遗传算法的性能很大程度上取决于参数的选择，上述例子中的参数仅作为演示用途。