How to prepare input for classification learner app?

Question

My work is on speech recognition. And this is what I have so far:

FOR TRAINING

I have input vector matrix (training data) of size 11811x65 double, corresponding response of size 1*65 double.

FOR TESTING

I have a matrix of size 5942x11 double.

I want to use classification learner app having "Multiclass SVM".

How should I give input to classification learner app?

1. Do we only give training data and its corresponding output?
2. Can give both training as well as testing data simultaneously to the app?
3. ? (Maybe I'm doing it all wrong?)

Answer 1

I suppose that you have 65 observations in your training data X and 11 in the testing data Z ? right?

1. create table:

   train = array2table(X);
   train.Group = Y;% labels
   

2. train SVM using the classification learner

3. export the trained model let suppost that is named trainedsvm

4. now you can classify your testing data

   test = array2table(Z(:,1:11811));
   test.Group = U;% labels
   [predictedlabel,scores] = predict(trainedsvm,test);