鲸鱼算法python代码

这是一个简单的鲸鱼算法的Python代码示例：

import random
import math

# 鲸鱼算法实现类
class WhaleOptimizationAlgorithm:
    def __init__(self, num_of_search_agents, max_iter, dim, obj_func):
        self.num_of_search_agents = num_of_search_agents
        self.max_iter = max_iter
        self.dim = dim
        self.obj_func = obj_func

    # 初始化种群
    def init_population(self):
        return [[random.uniform(-10, 10) for j in range(self.dim)] for i in range(self.num_of_search_agents)]

    # 计算适应度
    def fitness(self, x):
        return self.obj_func(x)

    # 鲸鱼算法实现函数
    def optimize(self):
        # 初始化种群
        X = self.init_population()

        # 最优解
        best_solution_fitness = float("inf")
        best_solution_idx = -1
        best_solution = []

        # 迭代
        for t in range(self.max_iter):
            a = 2 - t * ((2) / self.max_iter)

            # 对每个鲸鱼进行迭代
            for i in range(self.num_of_search_agents):
                r1 = random.random()
                r2 = random.random()

                A = 2 * a * r1 - a
                C = 2 * r2

                # 寻找当前最优解
                if random.random() >= 0.5:
                    D = abs(C * best_solution - X[i])
                    X[i] = best_solution - A * D
                else:
                    D1 = abs(best_solution - X[i])
                    D2 = abs(self.best_solution - X[i])
                    X[i] = D1 * math.exp(b * C) * math.cos(2 * math.pi * C) + best_solution

                # 边界处理
                for j in range(self.dim):
                    X[i][j] = max(min(X[i][j], 10), -10)

                # 计算适应度
                fitness = self.fitness(X[i])

                # 更新最优解
                if fitness < best_solution_fitness:
                    best_solution_fitness = fitness
                    best_solution_idx = i
                    best_solution = X[i]

        return (best_solution, best_solution_fitness)

其中，`num_of_search_agents` 表示搜索代理的数量，`max_iter` 表示最大迭代次数，`dim` 表示搜索空间的维度，`obj_func` 是目标函数。在 `optimize()` 函数中，我们首先初始化种群，然后进行迭代，对每个鲸鱼进行操作，更新最优解。最后返回最优解及其适应度。