org.tribuo.regression.sgd.objectives

Class Huber

java.lang.Object

org.tribuo.regression.sgd.objectives.Huber

All Implemented Interfaces:

com.oracle.labs.mlrg.olcut.config.Configurable, com.oracle.labs.mlrg.olcut.provenance.Provenancable<com.oracle.labs.mlrg.olcut.provenance.ConfiguredObjectProvenance>, SGDObjective<DenseVector>, RegressionObjective

public class Huber
extends Object
implements RegressionObjective

Huber loss, i.e., a mixture of l2 and l1 losses.

Field Summary

Fields

Modifier and Type	Field and Description
static double	DEFAULT_COST

Constructor Summary

Constructors

Constructor and Description
Huber() Huber Loss using the default cost DEFAULT_COST.
Huber(double cost) Huber loss using the supplied cost.

Method Summary

Modifier and Type	Method and Description
com.oracle.labs.mlrg.olcut.provenance.ConfiguredObjectProvenance	getProvenance()
com.oracle.labs.mlrg.olcut.util.Pair<Double,SGDVector>	loss(DenseVector truth, SGDVector prediction) Deprecated.
com.oracle.labs.mlrg.olcut.util.Pair<Double,SGDVector>	lossAndGradient(DenseVector truth, SGDVector prediction) Scores a prediction, returning the loss and a vector of per output dimension gradients.
void	postConfig() Used by the OLCUT configuration system, and should not be called by external code.
String	toString()

Methods inherited from class java.lang.Object

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

Field Detail

DEFAULT_COST

public static final double DEFAULT_COST

See Also:

Constant Field Values

Constructor Detail

Huber

public Huber()

Huber Loss using the default cost DEFAULT_COST.

Huber

public Huber(double cost)

Huber loss using the supplied cost. Cost must be positive.

Parameters:

cost - The cost.

Method Detail

postConfig

public void postConfig()

Used by the OLCUT configuration system, and should not be called by external code.

Specified by:

postConfig in interface com.oracle.labs.mlrg.olcut.config.Configurable

loss

@Deprecated
public com.oracle.labs.mlrg.olcut.util.Pair<Double,SGDVector> loss(DenseVector truth,
                                                                               SGDVector prediction)

Deprecated.

Description copied from interface: RegressionObjective

Scores a prediction, returning the loss.

Specified by:

loss in interface RegressionObjective

Parameters:

truth - The true regression value.

prediction - The predicted regression value.

Returns:

A pair with the loss and gradient.

lossAndGradient

public com.oracle.labs.mlrg.olcut.util.Pair<Double,SGDVector> lossAndGradient(DenseVector truth,
                                                                              SGDVector prediction)

Description copied from interface: SGDObjective

Scores a prediction, returning the loss and a vector of per output dimension gradients.

Specified by:

lossAndGradient in interface SGDObjective<DenseVector>

Specified by:

lossAndGradient in interface RegressionObjective

Parameters:

truth - The true output.

prediction - The prediction for each dimension.

Returns:

The score and per dimension gradient.

toString

public String toString()

Overrides:

toString in class Object

getProvenance

public com.oracle.labs.mlrg.olcut.provenance.ConfiguredObjectProvenance getProvenance()

Specified by:

getProvenance in interface com.oracle.labs.mlrg.olcut.provenance.Provenancable<com.oracle.labs.mlrg.olcut.provenance.ConfiguredObjectProvenance>