Prev: how to import 3D ultrasound data in the Workspace
Next: Counting number of gold pixels in a picture
From: Mohammad A. Mezher on 29 Mar 2010 09:08
Hi everyone,

I had a problem regarding the output size from my evaluation function

the short code is
for i = 1:3
test = (indices == i); train = ~test;
svmStruct = svmtrain(A(train,,groups(train),'Kernel_Function', Kernel_FunctionValue,'Method', MethodValue);
ytest=d(test);
[classes,ypred] = svmclassify(svmStruct,A(test,);
classperf(cp,classes,test);
[AUC(:,i),tpr(:,i),fpr(:,i)]=svmroccurve(ypred,ytest);

end

the problem is in the returned values [AUC(:,i),tpr(:,i),fpr(:,i)] from the evaluation function named here "svmroccurve"

my question
why sometimes the returned value have different size
and if that right how can i manipulate to take the average of the returend value to compute AUC and to draw roc curve?????

could anyone kindly answer this ASAP pleaaaaas

regards
 | 
Pages: 1
Prev: how to import 3D ultrasound data in the Workspace
Next: Counting number of gold pixels in a picture