请帮我用python写出一个遗传算法求函数最大值，函数值如下：x**2 * np.exp(-x/2) * ((math.sin(5 * (x**2))) ** 2)

好的，下面是一个使用Python实现的遗传算法求解函数最大值的示例代码：

import numpy as np
import math
import random

# 定义目标函数
def fitness_func(x):
    return x**2 * np.exp(-x/2) * ((math.sin(5 * (x**2))) ** 2)

# 初始化种群
def init_population(pop_size, chrom_length):
    population = np.zeros((pop_size, chrom_length))
    for i in range(pop_size):
        population[i, :] = np.random.permutation(np.arange(chrom_length))
    return population

# 计算适应度
def cal_fitness(population):
    fitness_value = np.zeros((population.shape[0], 1))
    for i in range(population.shape[0]):
        chrom = population[i, :]
        x = chrom2x(chrom)
        fitness_value[i, 0] = fitness_func(x)
    return fitness_value

# 将染色体转换为实数
def chrom2x(chrom):
    x = np.zeros((1,))
    for i in range(chrom.shape[0]):
        x += chrom[i] * (2 ** i)
    x = x / (2 ** chrom.shape[0] - 1) * 10
    return x

# 选择操作
def selection(population, fitness_value):
    idx = np.random.choice(population.shape[0], size=2, replace=False)
    if fitness_value[idx[0]] > fitness_value[idx[1]]:
        return population[idx[0], :]
    else:
        return population[idx[1], :]

# 交叉操作
def crossover(parent1, parent2):
    child = np.zeros(parent1.shape)
    cross_point = np.random.randint(1, parent1.shape[0])
    child[:cross_point] = parent1[:cross_point]
    child[cross_point:] = parent2[cross_point:]
    return child

# 变异操作
def mutation(child, mutation_rate):
    for i in range(child.shape[0]):
        if np.random.rand() < mutation_rate:
            child[i] = 1 - child[i]
    return child

# 遗传算法主函数
def GA(pop_size, chrom_length, max_iter, mutation_rate):
    population = init_population(pop_size, chrom_length)
    best_fitness = 0
    best_chrom = np.zeros((chrom_length,))
    for i in range(max_iter):
        fitness_value = cal_fitness(population)
        idx = np.argmax(fitness_value)
        if fitness_value[idx] > best_fitness:
            best_fitness = fitness_value[idx]
            best_chrom = population[idx, :]
        new_population = np.zeros((pop_size, chrom_length))
        for j in range(pop_size):
            parent1 = selection(population, fitness_value)
            parent2 = selection(population, fitness_value)
            child = crossover(parent1, parent2)
            child = mutation(child, mutation_rate)
            new_population[j, :] = child
        population = new_population
    best_x = chrom2x(best_chrom)
    return best_x, best_fitness

# 调用遗传算法函数求解函数最大值
best_x, best_fitness = GA(pop_size=100, chrom_length=10, max_iter=500, mutation_rate=0.01)
print("最优解：x = %.4f, f(x) = %.4f" % (best_x, best_fitness))

这段代码使用遗传算法求解函数 `x**2 * np.exp(-x/2) * ((math.sin(5 * (x**2))) ** 2)` 的最大值，其中 `pop_size` 表示种群大小，`chrom_length` 表示染色体长度，`max_iter` 表示最大迭代次数，`mutation_rate` 表示变异概率。在 `GA` 函数中，我们首先初始化种群，然后循环进行选择、交叉和变异操作，直到达到最大迭代次数或找到最优解为止。

最后输出最优解的 `x` 值和函数最大值。你可以根据需要修改代码中的参数和函数，以获得更好的结果。