org.deeplearning4j.nn.conf.layers

Class AbstractLSTM.Builder<T extends AbstractLSTM.Builder<T>>

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<T>

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

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

Direct Known Subclasses:

GravesLSTM.Builder, LSTM.Builder

Enclosing class:

AbstractLSTM

public abstract static class AbstractLSTM.Builder<T extends AbstractLSTM.Builder<T>>
extends BaseRecurrentLayer.Builder<T>

Field Summary

Fields

Modifier and Type	Field and Description
protected double	forgetGateBiasInit Set forget gate bias initalizations.
protected IActivation	gateActivationFn Activation function for the LSTM gates.
protected boolean	helperAllowFallback When using CuDNN and an error is encountered, should fallback to the non-CuDNN implementatation be allowed? If set to false, an exception in CuDNN will be propagated back to the user.

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

inputWeightConstraints, recurrentConstraints, rnnDataFormat, weightInitFnRecurrent

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 and Description
Builder()

Method Summary

Modifier and Type	Method and Description
T	forgetGateBiasInit(double biasInit) Set forget gate bias initalizations.
T	gateActivationFunction(Activation gateActivationFn) Activation function for the LSTM gates.
T	gateActivationFunction(IActivation gateActivationFn) Activation function for the LSTM gates.
T	gateActivationFunction(String gateActivationFn) Activation function for the LSTM gates.
T	helperAllowFallback(boolean allowFallback) When using a helper (CuDNN or MKLDNN in some cases) and an error is encountered, should fallback to the non-helper implementation be allowed? If set to false, an exception in the helper will be propagated back to the user.

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

constrainInputWeights, constrainRecurrent, dataFormat, weightInitRecurrent, weightInitRecurrent, weightInitRecurrent

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, weightInit, weightNoise

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

build, constrainAllParameters, constrainBias, constrainWeights, dropOut, dropOut, name

Methods inherited from class java.lang.Object

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

Field Detail

forgetGateBiasInit

protected double forgetGateBiasInit

Set forget gate bias initalizations. Values in range 1-5 can potentially help with learning or longer-term dependencies.

gateActivationFn

protected IActivation gateActivationFn

Activation function for the LSTM gates. Note: This should be bounded to range 0-1: sigmoid or hard sigmoid, for example

helperAllowFallback

protected boolean helperAllowFallback

When using CuDNN and an error is encountered, should fallback to the non-CuDNN implementatation be allowed? If set to false, an exception in CuDNN will be propagated back to the user. If false, the built-in (non-CuDNN) implementation for LSTM/GravesLSTM will be used

Constructor Detail

Builder

public Builder()

Method Detail

forgetGateBiasInit

public T forgetGateBiasInit(double biasInit)

Set forget gate bias initalizations. Values in range 1-5 can potentially help with learning or longer-term dependencies.

gateActivationFunction

public T gateActivationFunction(String gateActivationFn)

Activation function for the LSTM gates. Note: This should be bounded to range 0-1: sigmoid or hard sigmoid, for example

Parameters:

gateActivationFn - Activation function for the LSTM gates

gateActivationFunction

public T gateActivationFunction(Activation gateActivationFn)

Activation function for the LSTM gates. Note: This should be bounded to range 0-1: sigmoid or hard sigmoid, for example

Parameters:

gateActivationFn - Activation function for the LSTM gates

gateActivationFunction

public T gateActivationFunction(IActivation gateActivationFn)

Activation function for the LSTM gates. Note: This should be bounded to range 0-1: sigmoid or hard sigmoid, for example

Parameters:

gateActivationFn - Activation function for the LSTM gates

helperAllowFallback

public T helperAllowFallback(boolean allowFallback)

When using a helper (CuDNN or MKLDNN in some cases) and an error is encountered, should fallback to the non-helper implementation be allowed? If set to false, an exception in the helper will be propagated back to the user. If false, the built-in (non-helper) implementation for LSTM/GravesLSTM will be used

Parameters:

allowFallback - Whether fallback to non-helper implementation should be used