使用 DEAP 库实现遗传算法解决旅行商问题

遗传算法（Genetic Algorithm, GA）是一种模拟自然选择过程的优化算法，常用于求解复杂问题，如旅行商问题（Traveling Salesman Problem, TSP）。TSP是一个经典的组合优化问题，目标是找到访问一组城市的最短路径并返回起点。 以下是使用Python的`deap`库简单演示遗传算法解决TSP的一个基本代码框架： ```python import random from deap import base, creator, tools # 创建遗传算法所需工具 creator.create("FitnessMax", base.Fitness, weights=(1.0,)) creator.create("Individual", list, fitness=creator.FitnessMax) toolbox = base.Toolbox() toolbox.register("population", tools.initRepeat, list, creator.Individual) toolbox.register("mate", tools.cxTwoPoint) toolbox.register("mutate", tools.mutShuffleIndexes, indpb=0.05) toolbox.register("select", tools.selTournament, tournsize=3) # 初始化问题的数据，例如城市列表和距离矩阵 cities = ... # 假设这是一个二维数组，表示每个城市的坐标 distances = ... # 计算城市之间的实际距离 def eval_tour(individual): total_distance = 0 for i in range(len(individual)): total_distance += distances[individual[i-1]][individual[i]] return (total_distance,), toolbox.register("evaluate", eval_tour) toolbox.register("update_fitness", tools.map, toolbox.evaluate) # 主循环 pop = toolbox.population(n=100) # 初始种群大小 for gen in range(100): # 迭代次数 offspring = toolbox.select(pop, len(pop)) # 选择 offspring = [toolbox.clone(ind) for ind in offspring] # 克隆避免直接修改原始种群 offspring = [toolbox.mate(*ind) for ind in zip(offspring[::2], offspring[1::2])] # 配对交叉 offspring = [toolbox.mutate(ind) for ind in offspring] # 变异 invalid_ind = [ind for ind in offspring if not ind.fitness.valid] # 检查是否满足约束 fitnesses = toolbox.map(toolbox.update_fitness, invalid_ind) # 更新适应度 for ind, fit in zip(invalid_ind, fitnesses): ind.fitness.values = fit pop[:] = offspring # 替换种群 best_solution = min(pop, key=lambda x: x.fitness.values[0]) # 获取最佳解决方案 ```