org.apache.reef.examples.group.bgd.data

Class SparseExample

java.lang.Object

org.apache.reef.examples.group.bgd.data.SparseExample

All Implemented Interfaces:

Serializable, Example

public final class SparseExample
extends Object
implements Example

Example implementation on a index and value array.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
SparseExample(double label, float[] values, int[] indices)

Method Summary

Methods

Modifier and Type	Method and Description
void	addGradient(Vector gradientVector, double gradient) Adds the current example's gradient to the gradientVector, assuming that the gradient with respect to the prediction is gradient.
int	getFeatureLength()
double	getLabel() Access to the label.
double	predict(Vector w) Computes the prediction for this Example, given the model w.

Methods inherited from class java.lang.Object

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

Constructor Detail

SparseExample

public SparseExample(double label,
             float[] values,
             int[] indices)

Method Detail

getFeatureLength

public int getFeatureLength()

getLabel

public double getLabel()

Description copied from interface: Example

Access to the label.

Specified by:

getLabel in interface Example

Returns:

the label

predict

public double predict(Vector w)

Description copied from interface: Example

Computes the prediction for this Example, given the model w.

w.dot(this.getFeatures())

Specified by:

predict in interface Example

Parameters:

w - the model

Returns:

the prediction for this Example, given the model w.

addGradient

public void addGradient(Vector gradientVector,
               double gradient)

Description copied from interface: Example

Adds the current example's gradient to the gradientVector, assuming that the gradient with respect to the prediction is gradient.

Specified by:

addGradient in interface Example