Class SparseLinearSVM

java.lang.Object

smile.classification.AbstractClassifier<SparseArray>

smile.classification.SparseLinearSVM

All Implemented Interfaces:

Serializable, ToDoubleFunction<SparseArray>, ToIntFunction<SparseArray>, Classifier<SparseArray>

public class SparseLinearSVM
extends AbstractClassifier<SparseArray>

Sparse linear support vector machines for classification.

See Also:

• Serialized Form

Nested Class Summary

Nested classes/interfaces inherited from interface Classifier

Classifier.Trainer<T,M>

Field Summary

Fields inherited from class AbstractClassifier

classes

Constructor Summary

Constructors

Constructor	Description
SparseLinearSVM(int p, KernelMachine<SparseArray> svm)	Constructor.

Method Summary

Modifier and Type	Method	Description
double	intercept()	Returns the intercept.
int	predict(SparseArray x)	Predicts the class label of an instance.
double	score(SparseArray x)	The raw prediction score.
double[]	weights()	Returns the linear weights.

Methods inherited from class AbstractClassifier

classes, numClasses

Methods inherited from class Object

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

Methods inherited from interface Classifier

applyAsDouble, applyAsInt, online, predict, predict, predict, predict, predict, predict, predict, soft, update, update, update

Constructor Details

SparseLinearSVM

public SparseLinearSVM(int p,
 KernelMachine<SparseArray> svm)

Constructor.

Parameters:

p - the dimension of input vector.

svm - The kernel machine.

Method Details

weights

public double[] weights()

Returns the linear weights.

Returns:

the linear weights.

intercept

public double intercept()

Returns the intercept.

Returns:

the intercept.

predict

public int predict(SparseArray x)

Description copied from interface: Classifier

Predicts the class label of an instance.

Parameters:

x - the instance to be classified.

Returns:

the predicted class label.

score

public double score(SparseArray x)

Description copied from interface: Classifier

The raw prediction score.

Parameters:

x - the instance to be classified.

Returns:

the raw prediction score.