how to do cross validation for multi svm classification ?

Question

Hadeer tawfik 2016-10-23

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/308764-how-to-do-cross-validation-for-multi-svm-classification

编辑： Hadeer tawfik 2016-10-24

I am totally confused about how to do cross validation for this multisvm code,I have tried the crossval function but it is not working, and shows me this error

Error using crossval (line 160)
At least two inputs are needed.
Error in deteectsoilme (line 27)
CVMdl = crossval(results);
 *here is  the code*
                      srcFiles = dir('E:\sense\Soil Detection_Upload\Soil Detection_Code\all\*.jpg');
                        for i = 1 : length(srcFiles)
                            filename = strcat('E:\sense\Soil Detection_Upload\Soil Detection_Code\all\',srcFiles(i).name);
                        Img = imread(filename);
                        Img = imresize(Img,[256,256]);
                        %figure, imshow(Img); title('Query Image');
                        % Enhance Contrast
                        I = imadjust(Img,stretchlim(Img));
                        %figure, imshow(I);title('Contrast Enhanced');
                        % Extract Features from query image
                        [Feature_Vector] = Extract_FeaturesofSoilforall(I);
                        whos Feature_Vector
                        % Load Training Features
                        [X,T] = cataractdataset;
                         TrainFeat = X;
                         Train_Label = T;
                        test = Feature_Vector;
                        result = multisvm(TrainFeat,Train_Label,test);
                        %disp(result)
                        CVMdl = crossval(results);% Trying crossval function 
                        %if result == 1
                           %helpdlg(' grade1');
                           %disp(' grade1 ');
                        %elseif result == 2
                            %helpdlg(' last ');
                            %disp('last');
                        %elseif result == 3
                            %helpdlg('normal');
                            %disp(' normal ');
                        end
                        % Tabulate the results using a confusion matrix.
                        confMat = confusionmat(Train_Label, results)
                        diag(confMat)

and this is the multisvm code:

function [itrfin] = multisvm( T,C,test )
%Inputs: T=Training Matrix, C=Group, test=Testing matrix
%Outputs: itrfin=Resultant class
itrind=size(test,1);
itrfin=[];
Cb=C;
Tb=T;
for tempind=1:itrind
    tst=test(tempind,:);
    C=Cb;
    T=Tb;
    u=unique(C);
    N=length(u);
    c4=[];
    c3=[];
    j=1;
    k=1;
    if(N>2)
        itr=1;
        classes=0;
        cond=max(C)-min(C);
        while((classes~=1)&&(itr<=length(u))&& size(C,2)>1 && cond>0)
        %This while loop is the multiclass SVM Trick
            c1=(C==u(itr));
            newClass=c1;
            %svmStruct = svmtrain(T,newClass,'kernel_function','rbf'); % I am using rbf kernel function, you must change it also
            svmStruct = svmtrain(T,newClass,'kernel_function','rbf');
            classes = svmclassify(svmStruct,tst);
              % This is the loop for Reduction of Training Set
              for i=1:size(newClass,2)
                  if newClass(1,i)==0;
                      c3(k,:)=T(i,:);
                      k=k+1;
                  end
              end
          T=c3;
          c3=[];
          k=1;
              % This is the loop for reduction of group
              for i=1:size(newClass,2)
                  if newClass(1,i)==0;
                      c4(1,j)=C(1,i);
                      j=j+1;
                  end
              end
          C=c4;
          c4=[];
          j=1;
          cond=max(C)-min(C); % Condition for avoiding group 
                              %to contain similar type of values 
                              %and the reduce them to process
              % This condition can select the particular value of iteration
              % base on classes
              if classes~=1
                  itr=itr+1;
              end    
          end
      end
valt=Cb==u(itr);    % This logic is used to allow classification
val=Cb(valt==1);    % of multiple rows testing matrix
val=unique(val);
itrfin(tempind,:)=val;  
end
end