Cross validation is very slow! How can I improve the performance?

Hello guys,

I'm performing a per pixel supervised multiclass SVM machine learning classification. I would like to do a cross-validation of my model to see its accuracy. This the code I'm using right now. I have four classes and the size of my training data is a matrix of 170000x111.

    SVMModel = fitcecoc(trainingData,classes);
    cp = cvpartition(classes, 'k', 10);
    CVM = crossval(SVMModel, 'CVPartition, cp);

The problem of this is the crossval function. It's taking a loooot of time and I don't know why. Maybe it's because of the size of the training data, or maybe I'm not using it well? I don't know.

Could someone tell if this is normal? If not, how could I improve the performance?

In addition, I would like to generate a confusion matrix of the result, but I didn't found any easy way in the web. Someone knows how to generate a confusion matrix from my crossval results?

Thank in advance for your help and your time!

[John D'Errico]

CV with a huge dataset is of course going to take a long time. Get a bigger, much faster computer.