I am using SVM in Matlab for classification. I directly gave the training data set and class labels for training but without any parameters. My code looks like this:
traningData = myData;
label = [1 1 0 0 0 0 0 0 1 1 1 1 0 1 1 0 1 1 1 0];
SVMStructure = svmtrain (myData, label);
... %further prediction part
I found the default SVM method is 'linear', so here in my code, it should be 'linear' then. But how about the parameter C? It says from the Matlab documentation:
The resulting structure, SVMstruct, contains the optimized parameters from the SVM algorithm, enabling you to classify new data
So does that mean Matlab automatically optimizes the paramter C for linear SVM here?
