ws.palladian.classification

Class LibSvmLearner

java.lang.Object

ws.palladian.core.AbstractLearner<LibSvmModel>

ws.palladian.classification.LibSvmLearner

All Implemented Interfaces:

Learner<LibSvmModel>

public final class LibSvmLearner
extends AbstractLearner<LibSvmModel>

A wrapper classifier for the LIBSVM machine learning library.

Since:

2.0

Version:

2.0

Author:

Klemens Muthmann, Philipp Katz

Constructor Summary

Constructors

Constructor and Description
LibSvmLearner(LibSvmKernel kernel) Creates a new completely initialized LibSvmLearner using a linear kernel.

Method Summary

Modifier and Type	Method and Description
String	toString()
LibSvmModel	train(Dataset dataset)

Methods inherited from class ws.palladian.core.AbstractLearner

train, train

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Constructor Detail

LibSvmLearner

public LibSvmLearner(LibSvmKernel kernel)

Creates a new completely initialized LibSvmLearner using a linear kernel. It can be used to either train a new model or classify unlabeled FeatureVectors.

Parameters:

kernel - The kernel to use with the SVM predictor. This implementation currently supports the LinearKernel and the RBFKernel. The kernel is required to transfer the data to a higher dimensional space, where it is separable. Please read on the theoretical details of SVM to learn more. If you do not care you are probably fine using either kernel. Just try them.

Method Detail

train

public LibSvmModel train(Dataset dataset)

toString

public String toString()

Overrides:

toString in class AbstractLearner<LibSvmModel>