Package org.deeplearning4j.nn.conf.layers

Class EmbeddingSequenceLayer.Builder

java.lang.Object

org.deeplearning4j.nn.conf.layers.Layer.Builder<T>

org.deeplearning4j.nn.conf.layers.BaseLayer.Builder<T>

org.deeplearning4j.nn.conf.layers.FeedForwardLayer.Builder<EmbeddingSequenceLayer.Builder>

org.deeplearning4j.nn.conf.layers.EmbeddingSequenceLayer.Builder

Enclosing class:

EmbeddingSequenceLayer

public static class EmbeddingSequenceLayer.Builder
extends FeedForwardLayer.Builder<EmbeddingSequenceLayer.Builder>

Field Summary

Fields inherited from class org.deeplearning4j.nn.conf.layers.FeedForwardLayer.Builder

nIn, nOut

Fields inherited from class org.deeplearning4j.nn.conf.layers.BaseLayer.Builder

activationFn, biasInit, biasUpdater, gainInit, gradientNormalization, gradientNormalizationThreshold, iupdater, regularization, regularizationBias, weightInitFn, weightNoise

Fields inherited from class org.deeplearning4j.nn.conf.layers.Layer.Builder

allParamConstraints, biasConstraints, iDropout, layerName, weightConstraints

Constructor Summary

Constructors

Constructor	Description
Builder()

Method Summary

Modifier and Type	Method	Description
EmbeddingSequenceLayer	build()
EmbeddingSequenceLayer.Builder	hasBias(boolean hasBias)	If true: include bias parameters in the layer.
EmbeddingSequenceLayer.Builder	inferInputLength(boolean inferInputLength)	Set input sequence inference mode for embedding layer.
EmbeddingSequenceLayer.Builder	inputLength(int inputLength)	Set input sequence length for this embedding layer.
EmbeddingSequenceLayer.Builder	outputDataFormat(RNNFormat format)
void	setWeightInitFn(IWeightInit weightInit)
EmbeddingSequenceLayer.Builder	weightInit(EmbeddingInitializer embeddingInitializer)	Initialize the embedding layer using the specified EmbeddingInitializer - such as a Word2Vec instance
EmbeddingSequenceLayer.Builder	weightInit(IWeightInit weightInit)	Weight initialization scheme to use, for initial weight values
EmbeddingSequenceLayer.Builder	weightInit(INDArray vectors)	Initialize the embedding layer using values from the specified array.

Methods inherited from class org.deeplearning4j.nn.conf.layers.FeedForwardLayer.Builder

nIn, nIn, nOut, nOut, units

Methods inherited from class org.deeplearning4j.nn.conf.layers.BaseLayer.Builder

activation, activation, biasInit, biasUpdater, dist, gainInit, gradientNormalization, gradientNormalizationThreshold, l1, l1Bias, l2, l2Bias, regularization, regularizationBias, updater, updater, weightDecay, weightDecay, weightDecayBias, weightDecayBias, weightInit, weightInit, weightNoise

Methods inherited from class org.deeplearning4j.nn.conf.layers.Layer.Builder

constrainAllParameters, constrainBias, constrainWeights, dropOut, dropOut, name

Methods inherited from class java.lang.Object

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

Constructor Detail

Builder

public Builder()

Method Detail

outputDataFormat

public EmbeddingSequenceLayer.Builder outputDataFormat(RNNFormat format)

hasBias

public EmbeddingSequenceLayer.Builder hasBias(boolean hasBias)

If true: include bias parameters in the layer. False (default): no bias.

Parameters:

hasBias - If true: include bias parameters in this layer

inputLength

public EmbeddingSequenceLayer.Builder inputLength(int inputLength)

Set input sequence length for this embedding layer.

Parameters:

inputLength - input sequence length

Returns:

Builder

inferInputLength

public EmbeddingSequenceLayer.Builder inferInputLength(boolean inferInputLength)

Set input sequence inference mode for embedding layer.

Parameters:

inferInputLength - whether to infer input length

Returns:

Builder

weightInit

public EmbeddingSequenceLayer.Builder weightInit(IWeightInit weightInit)

Description copied from class: BaseLayer.Builder

Weight initialization scheme to use, for initial weight values

Overrides:

weightInit in class BaseLayer.Builder<EmbeddingSequenceLayer.Builder>

See Also:

IWeightInit

setWeightInitFn

public void setWeightInitFn(IWeightInit weightInit)

weightInit

public EmbeddingSequenceLayer.Builder weightInit(EmbeddingInitializer embeddingInitializer)

Initialize the embedding layer using the specified EmbeddingInitializer - such as a Word2Vec instance

Parameters:

embeddingInitializer - Source of the embedding layer weights

weightInit

public EmbeddingSequenceLayer.Builder weightInit(INDArray vectors)

Initialize the embedding layer using values from the specified array. Note that the array should have shape [vocabSize, vectorSize]. After copying values from the array to initialize the network parameters, the input array will be discarded (so that, if necessary, it can be garbage collected)

Parameters:

vectors - Vectors to initialize the embedding layer with

build

public EmbeddingSequenceLayer build()

Specified by:

build in class Layer.Builder<EmbeddingSequenceLayer.Builder>