clear,clc, close all
%%
data=xlsread('seeds.xlsx');
Total_Run= ;n= ;Sampling_Num= ;Sampling_iter= ;k= ;
Max_iter=1;
Accuracy_Proposed=zeros(Max_iter,4);Index_Proposed=zeros(Max_iter,4);Accuracy_Original=zeros(Max_iter,4);Index_Original=zeros(Max_iter,4);
for iter=1:Max_iter
     [Data_Base_Model,Data_Bootstrap] = Real_dataset(data,Total_Run,n,Sampling_Num,Sampling_iter,k);
    %% Parameters......................................................
    K= ;                              % Number of Classes
    Pi= ;u= ;epsilon= ;      % General Parameters
    C= ; beta= ; mu= ; eta= ;       % there is no output for eta more than 2!
    degree= ;                         % Kernel degree
    Ic= ;                         % Stop Criterion
    for j=1:5
        % type='Linear';                % Kernel Types: Linear, polynomial, RBF, Tanh
        %% Model Selection
        % model=['Proposed';'Original'];                 % Models: Proposed, Original
        for i=1:2
            if j==1
                type='Linear';
            elseif j==2
                type='polynomial';
            elseif j==3
                type='RBF';
            elseif j==4
                type='Tanh';
            end
            switch i
                case 1                                  % Proposed
                    model='Proposed';
                    data = Data_Bootstrap{1,1};
                     [f1,f2,class,macc,index, Accuracy,f1max,f2max] = SVM(data,K,type,Ic,Pi,u,epsilon,C,beta,mu,eta,degree,model,j,iter); % Proceeding both
                    Accuracy_Proposed(iter,j)=macc; Index_Proposed(iter,j)=index;
                   % figure(j)
                   % plot(Accuracy,'r')
                  % plot(f1max,f2max,'ro')
                case 2                                  % Original
                    model='Original';
                    data = Data_Base_Model{1,1};
                     [f1,f2,class,macc,index, Accuracy,f1max,f2max] = SVM(data,K,type,Ic,Pi,u,epsilon,C,beta,mu,eta,degree,model,j,iter); % Proceeding both
                    Accuracy_Original(iter,j)=macc; Index_Original(iter,j)=index;
                    %figure(j)
                   % hold on
                    %plot(Accuracy,'b')   %figure(3),legend('Proposed','RSSVM')
                   % plot(f1max,f2max,'bo')
            end
        end
    end
end
%% Main
Setting=[Pi, u, epsilon, C, beta, mu, eta, degree];
Mixed_Proposed=[];
Mixed_Original=[];
for kernel_number=1:4
    Mixed_Proposed=[Mixed_Proposed, Accuracy_Proposed(:,kernel_number),Index_Proposed(:,kernel_number)];
    Mixed_Original=[Mixed_Original, Accuracy_Original(:,kernel_number),Index_Original(:,kernel_number)];
end
Mixed_Proposed=[Setting;Mixed_Proposed];
Mixed_Original=[Setting;Mixed_Original];
xlswrite('Mixed_Proposed', Mixed_Proposed);
xlswrite('Mixed_Original', Mixed_Original);
disp('finished')

    %% Real dataset function code......................................................

function [Data_Base_Model,Data_Bootstrap] = Real_dataset(data,Total_Run,n,Sampling_Num,Sampling_iter,k)
%UNTITLED2 Summary of this function goes here
%   Detailed explanation goes here

% Total_Run=1;         % Total iteration(r)
% n=150;               % Dimension of each creation
Data_Base_Model{Total_Run}={};
Data_Bootstrap{Total_Run}={};
for ij=1:Total_Run
%% Ba Jaygozari- ya 11, 12 % darand ya 13 , 14
%data1 = DatasetCreatorIris(n); % sakhtegi
%data = sortrows(data1, size(data1, 2)); % Ta inja baraye dataye sakhtegist khatte 11 , 12     
%D=load('IrisTrain.mat');                  %real  Mahalle dataset
%data=D.data_6July;                        % Real data
SamplingData=data(:,1:size(data,2)-1); 
Data_Base_Model{ij}=data(:,1:size(data,2));
%% R1,R2,R3
    r1=0;r2=0;r3=0;
for i=1:size(data,1)
    if data(i,end)==1
        r1=r1+1;
    elseif data(i,end)==2
        r2=r2+1;
    elseif data(i,end)==3
        r3=r3+1;
    end
end
%%
% k=3;                                                    % Number of classes
% Sampling_iter=25;                                       % Number of Samples  
% Sampling_Num=7;                                         % Volume of Samples  
dimension=Total_Run*Sampling_iter*Sampling_Num;         % Dimension of random numbers
R1=randi(r1,1,dimension);
R2=randi(r2,1,dimension); R2=R2+r1;
R3=randi(r3,1,dimension); R3=R3+r1+r2;
R=[R1,R2,R3];              % Creating all random numbers
Sample_Matrix=zeros(1,size(SamplingData,2));            % Accelerating
Sample_Mean=zeros(1,size(SamplingData,2));              % Accelerating
m=0;
for i=1:Sampling_iter*k
    m=(i-1)*Sampling_Num;
    for j=1:Sampling_Num
        m=m+1;
        Sample_Matrix(1,:)=Sample_Matrix+SamplingData(R(m),:);
    end
    Sample_Mean(i,:)=Sample_Matrix/Sampling_Num;
    Sample_Matrix=zeros(1,size(SamplingData,2));
end
Sample_Mean(1:Sampling_iter,size(data,2))=1;
Sample_Mean(Sampling_iter+1:2*Sampling_iter,size(data,2))=2;
Sample_Mean(2*Sampling_iter+1:3*Sampling_iter,size(data,2))=3;
Data_Bootstrap{ij}=Sample_Mean;
end
end
    %% Find vertices function code......................................................

function [V] = Find_V(K)
%UNTITLED3 Summary of this function goes here
%   Detailed explanation goes here
V=zeros(K,K-1);         % Accelerating 
V(2,1)=1;               % Assumption
for j=3:K
    for k=1:j-2
        m=0;            % m is not a parameter, just helps us to calculate the summation easier
        for i=1:j-1
            m=m+V(i,k);
        end
        V(j,k)=m/(j-1);
        m=0;
        for i=1:j-2
            m=m+V(j,i)^2; 
        end
        V(j,j-1)=sqrt(1-m);
    end
end

inputArg1 = V;
end

    %% Finding class function code......................................................

function class = Find_class(f1,f2,dim)
%UNTITLED2 Summary of this function goes here
%   Detailed explanation goes here

class=zeros(1,dim);
center=[0.5,sqrt(3)/6];

m1=sqrt(3)/3; m2=-sqrt(3)/3;

for i=1:dim
    if f1(i)<=0.5
        if f2(i)<=center(2)
            class(i)=1;
        elseif (f2(i)-center(2))/(f1(i)-center(1))>=m2
            class(i)=1;
        else
            class(i)=3;
        end
    elseif f1(i)>0.5
        if f2(i)<=center(2)
            class(i)=2;
        elseif (f2(i)-center(2))/(f1(i)-center(1))<=m1
            class(i)=2;
        else
            class(i)=3;
        end 
    end   
end
end

    %% Finding E & F matrices code......................................................

function [outputArg1,outputArg2] = Find_E_F(K,V,dim,data)
%UNTITLED Summary of this function goes here
%   Detailed explanation goes here
      
E=zeros(K,K-1,K-1);
F=zeros(K,K-1,K-1);
for j=1:K-1
    for i=1:K
        mm=0;
        for m=1:K
            if i~=m
                mm=mm+1;
                E(i,mm,j)=2*(V(i,j)-V(m,j))';
                F(i,mm,j)=2*(V(i,j)^2-V(m,j)^2)';
            end
        end
    end
end
% Assigning E_hat and F_hat based on data to their E and F arrays.
E_hat=zeros(2*dim,dim,K-1);
F_hat=zeros(2*dim,1,K-1);
for j=1:K-1
    for i=1:dim
        Ci=data(i,end);
        E_hat(2*i-1:2*i,i,j)=E(Ci,:,j);
        F_hat(2*i-1:2*i,1,j)=F(Ci,:,j);
    end
end


outputArg1 = E_hat;
outputArg2 = F_hat;
end

    %% Finding H & B matrices code......................................................

function [H,B] = Find_H_B(E_hat,F_hat,Kernel,e1,e2,epsilon,Pi,u,model)
%UNTITLED2 Summary of this function goes here
%   Detailed explanation goes here


switch model

    case 'Proposed'
        H= E_hat(:,:,1)*(Kernel+eye(size(Kernel,1)))*E_hat(:,:,1)' + ...
            E_hat(:,:,2)*(Kernel+eye(size(Kernel,1)))*E_hat(:,:,2)';

        B= (F_hat(:,:,1)+((u-epsilon)*Pi + u)*e2) + (F_hat(:,:,2)+((u-epsilon)*Pi + u)*e2);
        %B= (F_hat(:,:,1)+epsilon*e2) + (F_hat(:,:,2)+epsilon*e2);


    case 'Original'

        H= E_hat(:,:,1)*(Kernel+e1*e1')*E_hat(:,:,1)' + ...
            E_hat(:,:,2)*(Kernel+e1*e1')*E_hat(:,:,2)';

        B= (F_hat(:,:,1)+epsilon*e2) + (F_hat(:,:,2)+epsilon*e2);



end
end

    %% Dual function code......................................................

function [Data_Base_Model,Data_Bootstrap] = Dual_dataset(Total_Run,n,Sampling_Num,Sampling_iter,k)
%UNTITLED2 Summary of this function goes here
%   Detailed explanation goes here

% Total_Run=1;         % Total iteration(r)
% n=150;               % Dimension of each creation
Data_Base_Model{Total_Run}={};
Data_Bootstrap{Total_Run}={};
for ij=1:Total_Run
%% Ba Jaygozari- ya 11, 12 % darand ya 13 , 14
data1 = DatasetCreatorIris(n); % sakhtegi
data = sortrows(data1, size(data1, 2)); % Ta inja baraye dataye sakhtegist khatte 11 , 12     
%D=load('IrisTrain.mat');                  %real  Mahalle dataset
%data=D.data_6July;                        % Real data
SamplingData=data(:,1:4); 
Data_Base_Model{ij}=data(:,1:5);
%% R1,R2,R3
    r1=0;r2=0;r3=0;
for i=1:size(data,1)
    if data(i,5)==1
        r1=r1+1;
    elseif data(i,5)==2
        r2=r2+1;
    elseif data(i,5)==3
        r3=r3+1;
    end
end
%%
% k=3;                                                    % Number of classes
% Sampling_iter=25;                                       % Number of Samples  
% Sampling_Num=7;                                         % Volume of Samples  
dimension=Total_Run*Sampling_iter*Sampling_Num;         % Dimension of random numbers
R1=randi(r1,1,dimension);
R2=randi(r2,1,dimension); R2=R2+r1;
R3=randi(r3,1,dimension); R3=R3+r1+r2;
R=[R1,R2,R3];              % Creating all random numbers
Sample_Matrix=zeros(1,size(SamplingData,2));            % Accelerating
Sample_Mean=zeros(1,size(SamplingData,2));              % Accelerating
m=0;
for i=1:Sampling_iter*k
    m=(i-1)*Sampling_Num;
    for j=1:Sampling_Num
        m=m+1;
        Sample_Matrix(1,:)=Sample_Matrix+SamplingData(R(m),:);
    end
    Sample_Mean(i,:)=Sample_Matrix/Sampling_Num;
    Sample_Matrix=zeros(1,size(SamplingData,2));
end
Sample_Mean(1:Sampling_iter,5)=1;
Sample_Mean(Sampling_iter+1:2*Sampling_iter,5)=2;
Sample_Mean(2*Sampling_iter+1:3*Sampling_iter,5)=3;
Data_Bootstrap{ij}=Sample_Mean;
end
end

    %% SVM function code......................................................
function [f1,f2,class,macc,index, Accuracy,f1max,f2max] = SVM(data,K,type,Ic,Pi,u,epsilon,C,beta,mu,eta,degree,model,j,iter)
%UNTITLED2 Summary of this function goes here
%   Detailed explanation goes here
%Normalising data......................................
data(:,1:end-1)=data(:,1:end-1)/(max(max(data(:,1:end-1))));
realclass=data(:,size(data,2));
dim=size(data,1);
V=Find_V(K);             % V matrix
A=data; A(:,end)=[];                           % Removing Label Column
e1=ones(size(A,1),1);e2=ones(2*size(A,1),1);   % Vector e

Kernel = My_Kernel(A,dim,degree,mu,beta,type);
[E_hat,F_hat]=Find_E_F(K,V,dim,data);               % producing E and F matrices for E_hat and F_hat.
[H,B]=Find_H_B(E_hat,F_hat,Kernel,e1,e2,epsilon,Pi,u,model);
% Algorithm 2 .................................
[f1,f2,class,Accuracy,f1max,f2max] = Algorithm2(K,dim,Ic,eta,H,B,C,Kernel,E_hat,realclass);
% Output
[macc,index]=max(Accuracy);
SIMplot(f1max,f2max,realclass,model,dim,j,iter);
end
    %% Kernel functions code......................................................
function Kernel = My_Kernel(A,dim,degree,mu,beta,type)
%UNTITLED3 Summary of this function goes here
%   Detailed explanation goes here
Kernel=zeros(dim,dim);


switch type

    case 'Linear'

        % Linear
        Kernel = A*A';

    case 'polynomial'
        % polynomial
        for i=1:dim
            for j=1:dim
                Kernel(i,j)=(A(i,:)*A(j,:)'+1)^degree;
            end
        end
    case 'RBF'
        % % RBF
        for i=1:dim
            for j=1:dim
                Kernel(i,j)=exp((-mu*norm(A(i,:)-A(j,:),2)^2));

            end
        end
        %
    case 'Tanh'
        % Tangent hyperbolic
        for i=1:dim
            for j=1:dim
                Kernel(i,j)=tanh(mu*A(i,:)*A(j,:)'+ beta);
            end
        end
    case 'Laplace'
     % Laplace
  for i = 1:dim
    for j = 1:dim
    Kernel(i, j) = exp(-mu * norm(A(i, :) - A(j, :), 1));
    end
  end
end
    %% Plots code......................................................

function []= ShowPlot(f1,f2)
%UNTITLED Summary of this function goes here
%   Detailed explanation goes here

figure()
plot(f1,f2,'o')
for i=1:41
    plot(f1(i),f2(i),'ro')
    hold on
end
for i=42:81
    plot(f1(i),f2(i),'go')
    hold on
end
for i=82:120
    plot(f1(i),f2(i),'bo')
end
%hold off
xlabel('f_1'),ylabel('f_2')
pause(0.01)
x=[0.5, 0.5, 0.75, 0.5, 0.25];
y=[0,sqrt(3)/6, 0.433, sqrt(3)/6, 0.433];
plot(x,y,'--');
% [macc,index]=max(Accuracy);
end
    %% Plots code......................................................










<pre>An Error occurred while handling another error:
yii\web\HeadersAlreadySentException: Headers already sent in  on line 0. in /var/www/html/prof-homepages/vendor/yiisoft/yii2/web/Response.php:366
Stack trace:
#0 /var/www/html/prof-homepages/vendor/yiisoft/yii2/web/Response.php(339): yii\web\Response-&gt;sendHeaders()
#1 /var/www/html/prof-homepages/vendor/yiisoft/yii2/web/ErrorHandler.php(136): yii\web\Response-&gt;send()
#2 /var/www/html/prof-homepages/vendor/yiisoft/yii2/base/ErrorHandler.php(135): yii\web\ErrorHandler-&gt;renderException()
#3 [internal function]: yii\base\ErrorHandler-&gt;handleException()
#4 {main}
Previous exception:
yii\web\HeadersAlreadySentException: Headers already sent in  on line 0. in /var/www/html/prof-homepages/vendor/yiisoft/yii2/web/Response.php:366
Stack trace:
#0 /var/www/html/prof-homepages/vendor/yiisoft/yii2/web/Response.php(339): yii\web\Response-&gt;sendHeaders()
#1 /var/www/html/prof-homepages/vendor/yiisoft/yii2/base/Application.php(656): yii\web\Response-&gt;send()
#2 /var/www/html/prof-homepages/vendor/faravaghi/yii2-filemanager/models/Files.php(696): yii\base\Application-&gt;end()
#3 /var/www/html/prof-homepages/vendor/faravaghi/yii2-filemanager/controllers/FilesController.php(484): faravaghi\filemanager\models\Files-&gt;getFile()
#4 [internal function]: faravaghi\filemanager\controllers\FilesController-&gt;actionGetFile()
#5 /var/www/html/prof-homepages/vendor/yiisoft/yii2/base/InlineAction.php(57): call_user_func_array()
#6 /var/www/html/prof-homepages/vendor/yiisoft/yii2/base/Controller.php(180): yii\base\InlineAction-&gt;runWithParams()
#7 /var/www/html/prof-homepages/vendor/yiisoft/yii2/base/Module.php(528): yii\base\Controller-&gt;runAction()
#8 /var/www/html/prof-homepages/vendor/yiisoft/yii2/web/Application.php(103): yii\base\Module-&gt;runAction()
#9 /var/www/html/prof-homepages/vendor/yiisoft/yii2/base/Application.php(386): yii\web\Application-&gt;handleRequest()
#10 /var/www/html/prof-homepages/backend/web/index.php(16): yii\base\Application-&gt;run()
#11 {main}</pre>