python遗传算法代码

以下是一个简单的 Python 遗传算法代码的示例：

import random

# 遗传算法参数
POPULATION_SIZE = 100
GENERATIONS = 50
MUTATION_RATE = 0.1

# 问题参数
TARGET = 371
NUM_GENES = 6
GENE_VALUES = [i for i in range(10)]

def fitness(individual):
    """
    计算个体的适应度
    """
    return abs(sum(individual) - TARGET)

def generate_individual():
    """
    生成随机个体
    """
    return [random.choice(GENE_VALUES) for _ in range(NUM_GENES)]

def crossover(parent1, parent2):
    """
    交叉操作
    """
    point = random.randint(1, NUM_GENES-1)
    return parent1[:point] + parent2[point:], parent2[:point] + parent1[point:]

def mutate(individual):
    """
    变异操作
    """
    for i in range(NUM_GENES):
        if random.random() < MUTATION_RATE:
            individual[i] = random.choice(GENE_VALUES)
    return individual

# 初始化种群
population = [generate_individual() for _ in range(POPULATION_SIZE)]

# 开始迭代
for generation in range(GENERATIONS):
    # 计算每个个体的适应度
    fitness_scores = [fitness(individual) for individual in population]

    # 找到适应度最好的个体
    best_individual = population[fitness_scores.index(min(fitness_scores))]
    print("Generation {}: Best individual = {}, Fitness = {}".format(generation, best_individual, fitness(best_individual)))

    # 选择操作
    parents = random.choices(population, weights=[1/fitness_score for fitness_score in fitness_scores], k=2)

    # 交叉操作
    child1, child2 = crossover(parents[0], parents[1])

    # 变异操作
    child1 = mutate(child1)
    child2 = mutate(child2)

    # 将子代加入种群
    population += [child1, child2]

    # 选择新的种群
    population = sorted(population, key=lambda individual: fitness(individual))[:POPULATION_SIZE]

这段代码实现了一个简单的数字求和问题的遗传算法。在每个迭代中，我们计算每个个体的适应度，找到适应度最好的个体，然后进行选择、交叉和变异操作来生成新的子代。最后，我们将子代加入种群，并选择新的种群。在每个迭代结束时，我们打印出最佳个体的适应度和基因序列。