smile.base.svm

Class LASVM<T>

java.lang.Object

smile.base.svm.LASVM<T>

All Implemented Interfaces:

java.io.Serializable

public class LASVM<T>
extends java.lang.Object
implements java.io.Serializable

LASVM is an approximate SVM solver that uses online approximation. It reaches accuracies similar to that of a real SVM after performing a single sequential pass through the training examples. Further benefits can be achieved using selective sampling techniques to choose which example should be considered next. LASVM requires considerably less memory than a regular SVM solver. This becomes a considerable speed advantage for large training sets.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
LASVM(smile.math.kernel.MercerKernel<T> kernel, double C, double tol) Constructor.
LASVM(smile.math.kernel.MercerKernel<T> kernel, double Cp, double Cn, double tol) Constructor.

Method Summary

Modifier and Type	Method and Description
KernelMachine<T>	fit(T[] x, int[] y) Trains the model.
KernelMachine<T>	fit(T[] x, int[] y, int epoch) Trains the model.

Methods inherited from class java.lang.Object

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

Constructor Detail

LASVM

public LASVM(smile.math.kernel.MercerKernel<T> kernel,
             double C,
             double tol)

Constructor.

Parameters:

C - the soft margin penalty parameter.

tol - the tolerance of convergence test.

LASVM

public LASVM(smile.math.kernel.MercerKernel<T> kernel,
             double Cp,
             double Cn,
             double tol)

Constructor.

Parameters:

Cp - the soft margin penalty parameter for positive instances.

Cn - the soft margin penalty parameter for negative instances.

tol - the tolerance of convergence test.

Method Detail

fit

public KernelMachine<T> fit(T[] x,
                            int[] y)

Trains the model.

Parameters:

x - training samples.

y - training labels.

fit

public KernelMachine<T> fit(T[] x,
                            int[] y,
                            int epoch)

Trains the model.

Parameters:

x - training samples.

y - training labels.

epoch - the number of epochs, usually 1 or 2 is sufficient.