How to use fitcsvm in matlab classifications Brain tumor Mr image?(Benign,Malignant)
1 visualización (últimos 30 días)
Mostrar comentarios más antiguos
load egitimseti.mat xdata = meas; group = label;
%svmStruct1 = svmtrain(xdata,group,'kernel_function', 'linear');
%Train the SVM Classifier
SVMModel= fitcsvm(meas,group,'KernelFunction','rbf','BoxConstraint',Inf);
sv = SVMModel.SupportVectors;
figure,
gscatter(xdata(:,1),xdata(:,2),group);
hold on
plot(sv(:,1),sv(:,2),'ko','MarkerSize',10)
legend('MALIGNANT','BENIGN','Support Vector')
hold off
%species = svmclassify(svmStruct1,feat,'showplot',false
%x = fitcdiscr(xdata,group); label= predict(SVMModel,xdata);
%data1 = [meas(:,1),meas(:,2)]; %newfeat = [feat(:,1),feat(:,2)];
%close all
%SVMModel2=fitcsvm(data1,group); %label2=predict(SVMModel2,newfeat);
%En yüksek doğruluk(accurasy) icin trainset
data = meas; groups = ismember(label,'BENIGN'); groups = ismember(label,'MALIGNANT'); [train,test] = crossvalind('HoldOut',groups);%test ve egitim icin veriler karıstırıldı.0 ve 1 cp = classperf(groups);
%svmStruct = svmtrain(data(train,:),groups(train),'showplot',false,'kernel_function','linear');%yapi olustu%r %classes = svmclassify(svmStruct,data(test,:),'showplot',false);%svm ile sınıflandır SVMModelson=fitcsvm(data(train,:),groups(train)); class=predict(SVMModelson,data(test,:));
classperf(cp,class,test);
indicies = crossvalind('Kfold',label,10); cp = classperf(label);%for ;
%svmStruct = svmtrain(xdata(train,:),group(train),'boxconstraint',Inf,'showplot',false,'kernel_function','rbf');
%classes = svmclassify(svmStruct,xdata(test,:),'showplot',false);
SVMModelnormalize=fitcsvm(xdata(train,:),group(train));
classes=predict(SVMModelnormalize,xdata(test,:));
classperf(cp,classes,test);
%end
%çapraz geçerlilik sonucu
Accuracy = cp.CorrectRate*100; sprintf("Dogruluk oranı %d dir",Accuracy);
Respuestas (1)
Bernhard Suhm
el 20 de Dic. de 2017
Can you illustrate what your data looks like, and where you get what type of error? And which release of MATLAB are you using? Looks like you were on an old release at first, since svmtrain has long been depracated.
2 comentarios
Image Analyst
el 20 de Dic. de 2017
You can help us to help you.
Of course it always starts by you answering direct questions people ask, which you aren't doing (yet).
Ver también
Categorías
Más información sobre Image Data Workflows en Help Center y File Exchange.
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!