[f1,f2]=simplefeat('*h4*.png');
t(:,1)=f1;
t(:,2)=f2;
train=t(1:10,:);
test=t(11:20,:);
[f1,f2]=simplefeat('*tfr*.png');
t(:,1)=f1;
t(:,2)=f2;
train(11:20,:)=t(1:10,:);
test(11:20,:)=t(11:20,:);
% plot the training data for the first half as blue circles and the
% second half as green diamonds
plot(train(1:10,1),train(1:10,2),'bo',train(11:20,1),train(11:20,2),'gd');
hold on
% plot the test data for the first half as blue x's and the second half
% as green plusses
plot(test(1:10,1),test(1:10,2),'bx',test(11:20,1),test(11:20,2),'g+');
hold off
% plot decision boundary
ninc=200;
[pltx,plty]=finddecisionboundary(train(1:10,:),train(11:20,:),1,2,ninc)
% plot all of the points that were on the decision boundary
hold on
plot(pltx,plty,'k.');
hold off
% generate "labels" for each row showing that the first half is in class 1
group=[repmat(1,10,1); repmat(2,10,1)];
% train a linear classifier with the training data and evaluate it with
% the test data
cl=classify(test,train,group);
% the resulting vector shows which class the classifier thinks each test
% point is in
cl
% count true +, false +, true -, false - and build confusion matrix
tp=length(find(cl(1:10)==1));
fp=length(find(cl(11:20)==1));
fn=length(find(cl(1:10)==2));
tn=length(find(cl(11:20)==2));
confmat=[tp fn; fp tn]