国际象棋皇后可以越过其他棋子吗（如何让8个皇后在国际象棋棋盘上和谐相处）(2)__娱乐最前沿_一个专注明星八卦的网站娱乐最前沿

当前位置：主页 > 历史文化 > 正文

国际象棋皇后可以越过其他棋子吗（如何让8个皇后在国际象棋棋盘上和谐相处）(2)

时间：2023-01-25 14:59 来源：网络整理作者：娱乐最前沿阅读：次

def fitness_function(individual): value = 0 for i in range(7): for j in range(i 1,8,1): if individual[i] != individual[j]: x_distance = np.abs(individual[j] - individual[i]) y_distance = j - i if x_distance != y_distance: value = 1 return value

Softmax function 用于把顺应度函数转化为概率漫衍

def softmax(input): input = np.array(input, dtype=np.float) input = np.exp(input) output = input / input.sum() return output

Mutation 随机变异

def mutation(individual, prob=0.1): p = np.random.rand(8) individual[p>prob] = np.random.choice(range(8), 8)[p>prob] return individual

GA函数Genetic Algorithm

def GA(size = 4): #默认种群巨细为4，可以多试几个，结果纷歧样 size = size num_generation = 0 population = [] for i in range(size): population.append(np.random.choice(range(8), 8)) while (True): print("Generation : ", num_generation) fitness_list = [] selection = [] for individual in population: fitness_value = fitness_function(individual) if fitness_value == 28: print("Find Target!") print(individual) return individual fitness_list.append(fitness_value) print(fitness_list) print() #Selection is Here天然选择在这里 prob = softmax(fitness_list) select_id = np.random.choice(range(size), size, replace=True, p=prob) for idx in select_id: selection.append(population[idx]) num_pair = int(size/2) position = np.random.choice(range(1,7,1), num_pair, replace=True) #Crossover is Here基因片断的交错交流在这里 for i in range(0, size, 2): start = position[int(i/2)] tempa = copy.deepcopy(selection[i][start:]) tempb = copy.deepcopy(selection[i 1][start:]) selection[i][start:] = tempb selection[i 1][start:] = tempa #Mutation is Here变异在这里 for i in range(size): selection[i] = copy.deepcopy(mutation(selection[i], prob=0.8)) population = selection num_generation = 1

来看当作果吧

配置初始种群数目为4

Queen = GA(size = 4)

前4代的种群内4个个另外顺应度

后四代种群内4个个另外顺应度

国际象棋皇后可以越过其他棋子吗（怎样让8个皇后在国际象棋棋盘上调和相处）(8)