萤火虫算法(FA)优化极限学习机(ELM)的分类预测，多特征输入模型 FA-ELM分类预测模型 多特征输入单输出的二分类及多

function [nbest, fit_gen] = FA(costfhandle, u0, Lb, Ub, para, t_train, p_train) %% 检查输入参数 if nargin < 5 para = [10, 100, 0.25, 0.20, 1]; end if nargin < 4 Ub = []; end if nargin < 3 Lb = []; end if nargin < 2 disp('Usuage: FA_mincon(@cost,u0,Lb,Ub,para)'); end %% 得到初始参数 num = para(1); MaxGeneration = para(2); alpha = para(3); betamin = para(4); gamma = para(5); %% 检查上界向量与下界向量长度是否相同 if length(Lb) ~= length(Ub) disp('Simple bounds/limits are improper!') return end %% 计算待优化参数维度 dim = length(u0); %% 初始化目标函数值 zn = ones(num, 1) * 10^100; %% 初始化萤火虫位置 [ns, Lightn] = initial(num, dim, Lb, Ub, u0); %% 迭代开始 for k = 1 : MaxGeneration % 更新alpha（可选） alpha = alpha_new(alpha, MaxGeneration); % 对每个萤火虫计算目标函数值 for i = 1 : num zn(i) = costfhandle(ns(i, :), t_train, p_train); Lightn(i) = zn(i); end % 根据亮度排序 [Lightn, Index] = sort(zn); ns_tmp = ns; for i = 1 : num ns(i, :) = ns_tmp(Index(i), :); end % 找出当前最优种群 nso = ns; Lighto = Lightn; nbest = ns(1, :); cv = nbest; err = objfun(cv, t_train, p_train); % 保存最优值 fit_gen(k) = err; % 向较优方向移动 ns = movepop(num, dim, ns, Lightn, nso, Lighto, alpha, betamin, gamma, Lb, Ub); end