水墨风格的网站,金融投资网站,网上开店的基本流程,好看的网站首页图片回归预测 | Matlab基于SO-GRU蛇群算法优化门控循环单元的数据多输入单输出回归预测 目录 回归预测 | Matlab基于SO-GRU蛇群算法优化门控循环单元的数据多输入单输出回归预测效果一览基本介绍程序设计参考资料 效果一览 基本介绍 1.Matlab基于SO-GRU蛇群算法优化门控循环单元的数…回归预测 | Matlab基于SO-GRU蛇群算法优化门控循环单元的数据多输入单输出回归预测 目录 回归预测 | Matlab基于SO-GRU蛇群算法优化门控循环单元的数据多输入单输出回归预测效果一览基本介绍程序设计参考资料 效果一览 基本介绍 1.Matlab基于SO-GRU蛇群算法优化门控循环单元的数据多输入单输出回归预测完整源码和数据 2.优化参数为学习率隐含层节点正则化参数。 3.多特征输入单输出的回归预测。程序内注释详细直接替换数据就可以用。 4.程序语言为matlab程序可出预测效果图迭代优化图相关分析图运行环境matlab2020b及以上。评价指标包括:R2、MAE、MSE、RMSE和MAPE等。 5.代码特点参数化编程、参数可方便更改、代码编程思路清晰、注释明细。 程序设计 完整源码和数据获取方式(资源处下载)Matlab基于SO-GRU蛇群算法优化门控循环单元的数据多输入单输出回归预测。
function [fval,Xfood,gbest_t] SO(N,T,lb,ub,dim,fobj)
%initial
vec_flag[1,-1];
Threshold0.25;
Thresold2 0.6;
C10.5;
C2.05;
C32;
Xinitialization(N,dim,ub,lb);
for i1:Nfitness(i)feval(fobj,X(i,:));
end
[GYbest, gbest] min(fitness);
Xfood X(gbest,:);
%Diving the swarm into two equal groups males and females
Nmround(N/2);%eq.(23)
NfN-Nm;
XmX(1:Nm,:);
XfX(Nm1:N,:);
fitness_mfitness(1:Nm);
fitness_ffitness(Nm1:N);
[fitnessBest_m, gbest1] min(fitness_m);
Xbest_m Xm(gbest1,:);
[fitnessBest_f, gbest2] min(fitness_f);
Xbest_f Xf(gbest2,:);
for t 1:Tdisp([ ,num2str(t), ε ])Tempexp(-((t)/T)); %eq.(4)QC1*exp(((t-T)/(T)));%eq.(5)if Q1 Q1; end% Exploration Phase (no Food)
if QThresholdfor i1:Nmfor j1:1:dimrand_leader_index floor(Nm*rand()1);X_randm Xm(rand_leader_index, :);flag_index floor(2*rand()1);Flagvec_flag(flag_index);Amexp(-fitness_m(rand_leader_index)/(fitness_m(i)eps));%eq.(7)Xnewm(i,j)X_randm(j)Flag*C2*Am*((ub(j)-lb(j))*randlb(j));%eq.(6)endendfor i1:Nffor j1:1:dimrand_leader_index floor(Nf*rand()1);X_randf Xf(rand_leader_index, :);flag_index floor(2*rand()1);Flagvec_flag(flag_index);Afexp(-fitness_f(rand_leader_index)/(fitness_f(i)eps));%eq.(9)Xnewf(i,j)X_randf(j)Flag*C2*Af*((ub(j)-lb(j))*randlb(j));%eq.(8)endend
else %Exploitation Phase (Food Exists)if TempThresold2 %hotfor i1:Nmflag_index floor(2*rand()1);Flagvec_flag(flag_index);for j1:1:dimXnewm(i,j)Xfood(j)C3*Flag*Temp*rand*(Xfood(j)-Xm(i,j));%eq.(10)endendfor i1:Nfflag_index floor(2*rand()1);Flagvec_flag(flag_index);for j1:1:dimXnewf(i,j)Xfood(j)Flag*C3*Temp*rand*(Xfood(j)-Xf(i,j));%eq.(10)endendelse %coldif rand0.6 %fightfor i1:Nmfor j1:1:dimFMexp(-(fitnessBest_f)/(fitness_m(i)eps));%eq.(13)Xnewm(i,j)Xm(i,j) C3*FM*rand*(Q*Xbest_f(j)-Xm(i,j));%eq.(11)endendfor i1:Nffor j1:1:dimFFexp(-(fitnessBest_m)/(fitness_f(i)eps));%eq.(14)Xnewf(i,j)Xf(i,j)C3*FF*rand*(Q*Xbest_m(j)-Xf(i,j));%eq.(12)endendelse%matingfor i1:Nmfor j1:1:dimMmexp(-fitness_f(i)/(fitness_m(i)eps));%eq.(17)Xnewm(i,j)Xm(i,j) C3*rand*Mm*(Q*Xf(i,j)-Xm(i,j));%eq.(15endendfor i1:Nffor j1:1:dimMfexp(-fitness_m(i)/(fitness_f(i)eps));%eq.(18)Xnewf(i,j)Xf(i,j) C3*rand*Mf*(Q*Xm(i,j)-Xf(i,j));%eq.(16)endend
参考资料 [1] https://blog.csdn.net/kjm13182345320/article/details/129215161 [2] https://blog.csdn.net/kjm13182345320/article/details/128105718
