Creates/trains a new SVM instance

model = cv.SVM()
model = cv.SVM(...)

The first variant creates an empty model. Use cv.SVM.train to train the model. Since SVM has several parameters, you may want to find the best parameters for your problem, it can be done with cv.SVM.trainAuto.

The second variant accepts the same parameters as the train method, in which case it forwards the call after construction.