CODEWITHSUNDEEP
matlabcross-validation
anonymous2015
anonymous2015

Reputation: 13

Cross-validation using linear discriminant analysis

The following code performs 10-fold cross-validation using linear discriminant analysis:

load fisheriris
indices = crossvalind('Kfold',species,10);
cp = classperf(species); % initializes the CP object

for i = 1:10
    test = (indices == i); train = ~test;
    class = classify(meas(test),meas(train),species(train));
    % updates the CP object with the current classification results
    classperf(cp,class,test)  
end

cp.CorrectRate

How can this be modified to use fitcdiscr instead of classify on line 7? When I try, I get an error (Wrong number of arguments). I do not know what arguments are needed or not.

Upvotes: 0

Views: 830

Answers (1)

mf13
mf13

Reputation: 21

fitcdiscr returns a model based on the training data with true labels. Therefore in order to obtain the predicted classes (class), we need to use the model method predict.

load fisheriris 
indices = crossvalind('Kfold',species,10);
cp = classperf(species);
for i = 1:10
    test = (indices == i); train = ~test;
    Mdl = fitcdiscr(meas(train,:), species(train,:));
    class = Mdl.predict(meas(test,:));
    classperf(cp,class,test);
end
cp.CorrectRate

I tested this with the old function (classify) and the CorrectRate was the same.

Upvotes: 1

Related Questions