Package org.deeplearning4j.nn.conf.layers.samediff

Class SameDiffLambdaVertex

java.lang.Object

org.deeplearning4j.nn.conf.graph.GraphVertex

org.deeplearning4j.nn.conf.layers.samediff.SameDiffVertex

org.deeplearning4j.nn.conf.layers.samediff.SameDiffLambdaVertex

All Implemented Interfaces:

Serializable, Cloneable, TrainingConfig

public abstract class SameDiffLambdaVertex
extends SameDiffVertex

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
class	SameDiffLambdaVertex.VertexInputs

Field Summary

Fields

Modifier and Type	Field	Description
protected SameDiffLambdaVertex.VertexInputs	inputs

Fields inherited from class org.deeplearning4j.nn.conf.layers.samediff.SameDiffVertex

biasUpdater, dataType, gradientNormalization, gradientNormalizationThreshold, regularization, regularizationBias, updater

Constructor Summary

Constructors

Constructor	Description
SameDiffLambdaVertex()

Method Summary

Modifier and Type	Method	Description
GraphVertex	clone()
void	defineParametersAndInputs(SDVertexParams params)	Define the parameters - and inputs - for the network.
SDVariable	defineVertex(SameDiff sameDiff, Map<String,SDVariable> layerInput, Map<String,SDVariable> paramTable, Map<String,SDVariable> maskVars)	Define the vertex
abstract SDVariable	defineVertex(SameDiff sameDiff, SameDiffLambdaVertex.VertexInputs inputs)	The defineVertex method is used to define the foward pass for the vertex
protected SameDiffLambdaVertex.VertexInputs	getInputs(SameDiff sd)
void	initializeParameters(Map<String,INDArray> params)	Set the initial parameter values for this layer, if required

Methods inherited from class org.deeplearning4j.nn.conf.layers.samediff.SameDiffVertex

applyGlobalConfig, applyGlobalConfigToLayer, feedForwardMaskArrays, getGradientNormalization, getGradientNormalizationThreshold, getLayerName, getMemoryReport, getOutputType, getRegularizationByParam, getUpdaterByParam, getVertexParams, instantiate, isPretrainParam, maxVertexInputs, minVertexInputs, numParams, paramReshapeOrder, setDataType, validateInput

Methods inherited from class org.deeplearning4j.nn.conf.graph.GraphVertex

equals, hashCode

Methods inherited from class java.lang.Object

finalize, getClass, notify, notifyAll, toString, wait, wait, wait

Field Detail

inputs

protected transient SameDiffLambdaVertex.VertexInputs inputs

Constructor Detail

SameDiffLambdaVertex

public SameDiffLambdaVertex()

Method Detail

defineVertex

public abstract SDVariable defineVertex(SameDiff sameDiff,
                                        SameDiffLambdaVertex.VertexInputs inputs)

The defineVertex method is used to define the foward pass for the vertex

Parameters:

sameDiff - SameDiff instance to use to define the vertex

inputs - Layer input variable

Returns:

The output variable (orresponding to the output activations for the vertex)

defineVertex

public SDVariable defineVertex(SameDiff sameDiff,
                               Map<String,SDVariable> layerInput,
                               Map<String,SDVariable> paramTable,
                               Map<String,SDVariable> maskVars)

Description copied from class: SameDiffVertex

Define the vertex

Specified by:

defineVertex in class SameDiffVertex

Parameters:

sameDiff - SameDiff instance

layerInput - Input to the layer - keys as defined by SameDiffVertex.defineParametersAndInputs(SDVertexParams)

paramTable - Parameter table - keys as defined by SameDiffVertex.defineParametersAndInputs(SDVertexParams)

maskVars - Masks of input, if available - keys as defined by SameDiffVertex.defineParametersAndInputs(SDVertexParams)

Returns:

The final layer variable corresponding to the activations/output from the forward pass

defineParametersAndInputs

public void defineParametersAndInputs(SDVertexParams params)

Description copied from class: SameDiffVertex

Define the parameters - and inputs - for the network. Use SDLayerParams.addWeightParam(String, long...) and SDLayerParams.addBiasParam(String, long...). Note also you must define (and optionally name) the inputs to the vertex. This is required so that DL4J knows how many inputs exists for the vertex.

Specified by:

defineParametersAndInputs in class SameDiffVertex

Parameters:

params - Object used to set parameters for this layer

initializeParameters

public void initializeParameters(Map<String,INDArray> params)

Description copied from class: SameDiffVertex

Set the initial parameter values for this layer, if required

Specified by:

initializeParameters in class SameDiffVertex

Parameters:

params - Parameter arrays that may be initialized

clone

public GraphVertex clone()

Specified by:

clone in class GraphVertex

getInputs

protected SameDiffLambdaVertex.VertexInputs getInputs(SameDiff sd)