Trainable automatic radial basis Support Vector Classifier by LIBSVM
[W,KERNEL] = RBLIBSVC(A)
This routine computes a classifier by LIBSVC using a radial basis kernel with an optimised standard deviation by REGOPTC. The resulting classifier W is identical to LIBSVC(A,KERNEL). As the kernel optimisation is based on internal cross-validation the dataset A should be sufficiently large. Moreover it is very time-consuming as the kernel optimisation needs about 100 calls to LIBSVC.
Note that LIBSVC is basically a multi-class classifier. The kernel is thereby the same for all classes. Use MCLASSC to find a multi-class classifier composed by two-class classifiers with possibly different kernels.