遗传算法（GA）求解TSP问题(golang实现)

package main
 
import (
    "fmt"
    "math/rand"
    "time"
)
 
// 定义城市坐标结构体
type City struct {
    x, y float64
}
 
// 定义遗传算法解决旅行商问题的结构体
type GA struct {
    population [][]City
    nextCity   []City
    best       []City
    fitness    float64
    size       int
    mutation   float64
    crossover  float64
    elitism   bool
}
 
// 初始化遗传算法
func (ga *GA) Init(size int, mutation, crossover float64, elitism bool) {
    ga.size = size
    ga.mutation = mutation
    ga.crossover = crossover
    ga.elitism = elitism
    ga.population = make([][]City, size)
    ga.best = make([]City, len(cities))
    for i := range ga.population {
        ga.population[i] = make([]City, len(cities))
        for j := range ga.population[i] {
            ga.population[i][j] = cities[j]
        }
        rand.Shuffle(len(ga.population[i]), func(i, j int) {
            ga.population[i][i], ga.population[i][j] = ga.population[i][j], ga.population[i][i]
        })
    }
}
 
// 计算适应度函数
func (ga *GA) Fitness() {
    var total float64
    for i := range ga.population {
        var distance float64
        for j := 1; j < len(ga.population[i]); j++ {
            distance += Distance(ga.population[i][j-1], ga.population[i][j])
        }
        distance += Distance(ga.population[i][len(ga.population[i])-1], ga.population[i][0])
        if distance < ga.fitness || ga.fitness == 0 {
            copy(ga.best, ga.population[i])
            ga.fitness = distance
        }
        total += distance
    }
    fmt.Println("Best fitness:", ga.fitness)
}
 
// 交叉操作
func (ga *GA) Crossover() {
    for len(ga.population) < cap(ga.population) {
        parent1 := rand.Intn(len(ga.population))
        parent2 := rand.Intn(len(ga.population))
        if rand.Float64() < ga.crossover {
            crossPoint := rand.Intn(len(ga.population[parent1])-1) + 1
            ga.population = append(ga.population, append(ga.population[parent1][crossPoint:], ga.population[parent2][:crossPoint]...))
        }
    }
}
 
// 变异操作
func (ga *GA) Mutation() {
    for i := range ga.population {
        for j := range ga.population[i] {
            if rand.Float64() < ga.mutation {
                rand.Shuffle(len(ga.population), func(i, j int) {
                    ga.population[i][j], ga.population[i][j] = ga.population[i][j], ga.population[i][i]
                })
            }
        }
    }
}
 
// 选择操作
func (ga *GA) Selection() {
    newPopulation := make(