Package org.nd4j.linalg.lossfunctions

Class SameDiffLoss

java.lang.Object

org.nd4j.linalg.lossfunctions.SameDiffLoss

All Implemented Interfaces:

Serializable, ILossFunction

public abstract class SameDiffLoss
extends Object
implements ILossFunction

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field	Description
protected SDVariable	scorePerExampleVariable
protected SameDiff	sd

Constructor Summary

Constructors

Modifier	Constructor	Description
protected	SameDiffLoss()

Method Summary

Modifier and Type	Method	Description
INDArray	computeGradient(INDArray labels, INDArray preOutput, IActivation activationFn, INDArray mask)	Compute the gradient of the loss function with respect to the inputs: dL/dOutput
Pair<Double,INDArray>	computeGradientAndScore(INDArray labels, INDArray preOutput, IActivation activationFn, INDArray mask, boolean average)	Compute both the score (loss function value) and gradient.
double	computeScore(INDArray labels, INDArray preOutput, IActivation activationFn, INDArray mask, boolean average)	Compute the score (loss function value) for the given inputs.
INDArray	computeScoreArray(INDArray labels, INDArray preOutput, IActivation activationFn, INDArray mask)	Compute the score (loss function value) for each example individually.
protected void	createSameDiffInstance(DataType dataType)
abstract SDVariable	defineLoss(SameDiff sd, SDVariable layerInput, SDVariable labels)	Define the loss function. NOTE: The score on a *per example* basis - should return a SDVariable with shape [minibatch], where out[i] is the score for the ith minibatch
String	name()	The opName of this function

Methods inherited from class java.lang.Object

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

Field Detail

sd

protected transient SameDiff sd

scorePerExampleVariable

protected transient SDVariable scorePerExampleVariable

Constructor Detail

SameDiffLoss

protected SameDiffLoss()

Method Detail

defineLoss

public abstract SDVariable defineLoss(SameDiff sd,
                                      SDVariable layerInput,
                                      SDVariable labels)

Define the loss function.
NOTE: The score on a *per example* basis - should return a SDVariable with shape [minibatch], where out[i] is the score for the ith minibatch

Parameters:

sd - SameDiff instance to define the loss on

layerInput - Input to the SameDiff loss function

labels - Labels placeholder

Returns:

The score on a per example basis (SDVariable with shape [minibatch])

createSameDiffInstance

protected void createSameDiffInstance(DataType dataType)

computeScore

public double computeScore(INDArray labels,
                           INDArray preOutput,
                           IActivation activationFn,
                           INDArray mask,
                           boolean average)

Compute the score (loss function value) for the given inputs.

Specified by:

computeScore in interface ILossFunction

Parameters:

labels - Label/expected preOutput

preOutput - Output of the model (neural network)

activationFn - Activation function that should be applied to preOutput

mask - Mask array; may be null

average - Whether the score should be averaged (divided by number of rows in labels/preOutput) or not @return Loss function value

computeScoreArray

public INDArray computeScoreArray(INDArray labels,
                                  INDArray preOutput,
                                  IActivation activationFn,
                                  INDArray mask)

Compute the score (loss function value) for each example individually. For input [numExamples,nOut] returns scores as a column vector: [numExamples,1]

Specified by:

computeScoreArray in interface ILossFunction

Parameters:

labels - Labels/expected output

preOutput - Output of the model (neural network)

activationFn - Activation function that should be applied to preOutput

mask -

Returns:

Loss function value for each example; column vector

computeGradient

public INDArray computeGradient(INDArray labels,
                                INDArray preOutput,
                                IActivation activationFn,
                                INDArray mask)

Compute the gradient of the loss function with respect to the inputs: dL/dOutput

Specified by:

computeGradient in interface ILossFunction

Parameters:

labels - Label/expected output

preOutput - Output of the model (neural network), before the activation function is applied

activationFn - Activation function that should be applied to preOutput

mask - Mask array; may be null

Returns:

Gradient dL/dPreOut

computeGradientAndScore

public Pair<Double,INDArray> computeGradientAndScore(INDArray labels,
                                                           INDArray preOutput,
                                                           IActivation activationFn,
                                                           INDArray mask,
                                                           boolean average)

Compute both the score (loss function value) and gradient. This is equivalent to calling computeScore(INDArray, INDArray, IActivation, INDArray, boolean) and computeGradient(INDArray, INDArray, IActivation, INDArray) individually

Specified by:

computeGradientAndScore in interface ILossFunction

Parameters:

labels - Label/expected output

preOutput - Output of the model (neural network)

activationFn - Activation function that should be applied to preOutput

mask - Mask array; may be null

average - Whether the score should be averaged (divided by number of rows in labels/output) or not

Returns:

The score (loss function value) and gradient

name

public String name()

Description copied from interface: ILossFunction

The opName of this function

Specified by:

name in interface ILossFunction

Returns: