遗传算法(GA) - 求解 - 旅行商问题(TSP)

#!/usr/bin/env python
# coding: utf-8

from matplotlib import pyplot as plt
import numpy as np
import pandas as pd
import time


# 生成城市坐标
def gen_city(filename, city_num):
    name = ["city's name"] * city_num  # 这个并没什么用，但是不要省，省了的话还要修改代码
    x = [np.random.randint(0, 100) for _ in range(city_num)]
    y = [np.random.randint(0, 100) for _ in range(city_num)]

    x[0] = 0
    y[0] = 0
    with open(filename, "w") as f:
        for i in range(city_num):
            name[i] = f'city{i}'
            f.write(name[i] + "," + str(x[i]) + "," + str(y[i]) + "\n")
        f.write(name[0] + "," + str(x[0]) + "," + str(y[0]) + "\n")  # 最后一个节点即为起点

    # 打印城市的坐标
    position = pd.read_csv("cities.csv", names=['ind', 'lat', 'lon'])
    plt.scatter(x=position['lon'], y=position['lat'])
    plt.show()
    position.head()


def create_init_list(filename):
    data = pd.read_csv(filename, names=['index', 'lat', 'lon'])  # lat->纬度 lon->经度
    data_list = []
    for i in range(len(data)):
        data_list.append([float(data.iloc[i]['lon']), float(data.iloc[i]['lat'])])
    return data_list


def distance_matrix(coordinate_list, size):  # 生成距离矩阵，邻接矩阵
    d = np.zeros((size + 2, size + 2))
    for i in range(size + 1):
        x1 = coordinate_list[i][0]
        y1 = coordinate_list[i][1]
        for j in range(size + 1):
            if (i == j) or (d[i][j] != 0):
                continue
            x2 = coordinate_list[j][0]
            y2 = coordinate_list[j][1]
            distance = np.sqrt((x1 - x2) ** 2 + (y1 - y2) ** 2)
            if (i == 0):  # 起点与终点是同一城市
                d[i][j] = d[j][i] = d[size + 1][j] = d[j][size + 1] = distance
            else:
                d[i][j] = d[j][i] = distance
    return d


def path_length(d_matrix, path_list, size):  # 计算路径长度
    length = 0
    for i in range(size + 1):
        length += d_matrix[path_list[i]][path_list[i + 1]]
    return length


def shuffle(my_list):  # 起点和终点不能打乱
    temp_list = my_list[1:-1]
    np.random.shuffle(temp_list)
    shuffle_list = my_list[:1] + temp_list + my_list[-1:]
    return shuffle_list


def product_len_probability(my_list, d_matrix, size, p_num):  # population,   d,       size,p_num
    len_list = []  # 种群中每个个体（路径）的路径长度
    pro_list = []
    path_len_pro = []
    for path in my_list:
        len_list.append(path_length(d_matrix, path, size))
    max_len = max(len_list) + 1e-10
    gen_best_length = min(len_list)  # 种群中最优路径的长度
    gen_best_length_index = len_list.index(gen_best_length)  # 最优个体在种群中的索引
    # 使用最长路径减去每个路径的长度，得到每条路径与最长路径的差值，该值越大说明路径越小
    mask_list = np.ones(p_num) * max_len - np.array(len_list)
    sum_len = np.sum(mask_list)  # mask_list列表元素的和
    for i in range(p_num):
        if (i == 0):
            pro_list.append(mask_list[i] / sum_len)
        elif (i == p_num - 1):
            pro_list.append(1)
        else:
            pro_list.append(pro_list[i - 1] + mask_list[i] / sum_len)
    for i in range(p_num):
        # 路径列表 路径长度 概率
        path_len_pro.append([my_list[i], len_list[i], pro_list[i]])
    # 返回 最优路径 最优路径的长度 每条路径的概率
    return my_list[gen_best_length_index], gen_best_length, path_len_pro


def choose_cross(population, p_num):  # 随机产生交配者的索引，越优的染色体被选择几率越大
    jump = np.random.random()  # 随机生成0-1之间的小数
    if jump < population[0][2]:
        return 0
    low = 1
    high = p_num
    mid = int((low + high) / 2)
    # 二分搜索
    # 如果jump在population[mid][2]和population[mid-1][2]之间，那么返回mid
    while (low < high):
        if jump > population[mid][2]:
            low = mid
            mid = (low + high) // 2
        elif jump < population[mid - 1][2]:  # 注意这里一定是mid-1
            high = mid
            mid = (low + high) // 2
        else:
            return mid


def product_offspring(size, parent_1, parent_2, pm):  # 产生后代
    son = parent_1.copy()
    product_set = np.random.randint(1, size + 1)
    parent_cross_set = set(parent_2[1:product_set])  # 交叉序列集合
    cross_complete = 1
    for j in range(1, size + 1):
        if son[j] in parent_cross_set:
            son[j] = parent_2[cross_complete]
            cross_complete += 1
            if cross_complete > product_set:
                break
    if np.random.random() < pm:  # 变异
        son = veriation(son, size, pm)
    return son


def veriation(my_list, size, pm):  # 变异，随机调换两城市位置
    ver_1 = np.random.randint(1, size + 1)
    ver_2 = np.random.randint(1, size + 1)
    while ver_2 == ver_1:  # 直到ver_2与ver_1不同
        ver_2 = np.random.randint(1, size + 1)
    my_list[ver_1], my_list[ver_2] = my_list[ver_2], my_list[ver_1]
    return my_list


def main(filepath, p_num, gen, pm):
    start = time.time()
    coordinate_list = create_init_list(filepath)
    size = len(coordinate_list) - 2  # 除去了起点和终点
    d = distance_matrix(coordinate_list, size)  # 各城市之间的邻接矩阵
    path_list = list(range(size + 2))  # 初始路径
    # 随机打乱初始路径以建立初始种群路径
    population = [shuffle(path_list) for i in range(p_num)]
    # 初始种群population以及它的最优路径和最短长度
    gen_best, gen_best_length, population = product_len_probability(population, d, size, p_num)
    # 现在的population中每一元素有三项，第一项是路径，第二项是长度，第三项是使用时转盘的概率
    son_list = [0] * p_num  # 后代列表
    best_path = gen_best  # 最好路径初始化
    best_path_length = gen_best_length  # 最好路径长度初始化
    every_gen_best = [gen_best_length]  # 每一代的最优值
    for i in range(gen):  # 迭代gen代
        son_num = 0
        while son_num < p_num:  # 循环产生后代，一组父母亲产生两个后代
            father_index = choose_cross(population, p_num)  # 获得父母索引
            mother_index = choose_cross(population, p_num)
            father = population[father_index][0]  # 获得父母的染色体
            mother = population[mother_index][0]
            son_list[son_num] = product_offspring(size, father, mother, pm)  # 产生后代加入到后代列表中
            son_num += 1
            if son_num == p_num:
                break
            son_list[son_num] = product_offspring(size, mother, father, pm)  # 产生后代加入到后代列表中
            son_num += 1
        # 在新一代个体中找到最优路径和最优值
        gen_best, gen_best_length, population = product_len_probability(son_list, d, size, p_num)
        if (gen_best_length < best_path_length):  # 这一代的最优值比有史以来的最优值更优
            best_path = gen_best
            best_path_length = gen_best_length
        every_gen_best.append(gen_best_length)
    end = time.time()
    print(f"迭代用时：{(end - start)}s")
    print("史上最优路径:", best_path, sep=" ")  # 史上最优路径
    print("史上最短路径长度:", best_path_length, sep=" ")  # 史上最优路径长度

    # 打印各代最优值和最优路径
    x = [coordinate_list[point][0] for point in best_path]  # 最优路径各节点经度
    y = [coordinate_list[point][1] for point in best_path]  # 最优路径各节点纬度
    plt.figure(figsize=(8, 10))
    plt.subplot(211)
    plt.plot(every_gen_best)  # 画每一代中最优路径的路径长度
    plt.subplot(212)
    plt.scatter(x, y)  # 画点
    plt.plot(x, y)  # 画点之间的连线
    plt.grid()  # 给画布添加网格
    plt.show()


if __name__ == '__main__':
    city_num = 10  # 城市数量
    filepath = 'cities.csv'
    # gen_city(filepath, city_num)

    p_num = 100  # 种群个体数量
    gen = 1000  # 进化代数
    pm = 0.4  # 变异率
    main(filepath, p_num, gen, pm)