org.deeplearning4j.nn.conf.layers

Class BaseLayer.Builder<T extends BaseLayer.Builder<T>>

java.lang.Object

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

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

Direct Known Subclasses:

FeedForwardLayer.Builder

Enclosing class:

BaseLayer

public abstract static class BaseLayer.Builder<T extends BaseLayer.Builder<T>>
extends Layer.Builder<T>

Field Summary

Fields

Modifier and Type	Field and Description
protected org.nd4j.linalg.activations.IActivation	activationFn
protected double	biasInit
protected org.nd4j.linalg.learning.config.IUpdater	biasUpdater
protected Distribution	dist
protected GradientNormalization	gradientNormalization
protected double	gradientNormalizationThreshold
protected org.nd4j.linalg.learning.config.IUpdater	iupdater
protected double	l1
protected double	l1Bias
protected double	l2
protected double	l2Bias
protected WeightInit	weightInit
protected IWeightNoise	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	activation(org.nd4j.linalg.activations.Activation activation) Set the activation function for the layer, from an Activation enumeration value.
T	activation(org.nd4j.linalg.activations.IActivation activationFunction) Set the activation function for the layer.
T	biasInit(double biasInit) Bias initialization value, for layers with biases.
T	biasUpdater(org.nd4j.linalg.learning.config.IUpdater biasUpdater) Gradient updater configuration, for the biases only.
T	dist(Distribution dist) Distribution to sample initial weights from.
T	gradientNormalization(GradientNormalization gradientNormalization) Gradient normalization strategy.
T	gradientNormalizationThreshold(double threshold) Threshold for gradient normalization, only used for GradientNormalization.ClipL2PerLayer, GradientNormalization.ClipL2PerParamType, and GradientNormalization.ClipElementWiseAbsoluteValue Not used otherwise. L2 threshold for first two types of clipping, or absolute value threshold for last type of clipping.
T	l1(double l1) L1 regularization coefficient (weights only).
T	l1Bias(double l1Bias) L1 regularization coefficient for the bias.
T	l2(double l2) L2 regularization coefficient (weights only).
T	l2Bias(double l2Bias) L2 regularization coefficient for the bias.
T	updater(org.nd4j.linalg.learning.config.IUpdater updater) Gradient updater.
T	updater(Updater updater) Deprecated.
T	weightInit(Distribution distribution) Set weight initialization scheme to random sampling via the specified distribution.
T	weightInit(WeightInit weightInit) Weight initialization scheme to use, for initial weight values
T	weightNoise(IWeightNoise weightNoise) Set the weight noise (such as DropConnect and WeightNoise) for this layer

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

activationFn

protected org.nd4j.linalg.activations.IActivation activationFn

weightInit

protected WeightInit weightInit

biasInit

protected double biasInit

dist

protected Distribution dist

l1

protected double l1

l2

protected double l2

l1Bias

protected double l1Bias

l2Bias

protected double l2Bias

iupdater

protected org.nd4j.linalg.learning.config.IUpdater iupdater

biasUpdater

protected org.nd4j.linalg.learning.config.IUpdater biasUpdater

gradientNormalization

protected GradientNormalization gradientNormalization

gradientNormalizationThreshold

protected double gradientNormalizationThreshold

weightNoise

protected IWeightNoise weightNoise

Constructor Detail

Builder

public Builder()

Method Detail

activation

public T activation(org.nd4j.linalg.activations.IActivation activationFunction)

Set the activation function for the layer. This overload can be used for custom IActivation instances

Parameters:

activationFunction - Activation function to use for the layer

activation

public T activation(org.nd4j.linalg.activations.Activation activation)

Set the activation function for the layer, from an Activation enumeration value.

Parameters:

activation - Activation function to use for the layer

weightInit

public T weightInit(WeightInit weightInit)

Weight initialization scheme to use, for initial weight values

See Also:

WeightInit

weightInit

public T weightInit(Distribution distribution)

Set weight initialization scheme to random sampling via the specified distribution. Equivalent to: .weightInit(WeightInit.DISTRIBUTION).dist(distribution)

Parameters:

distribution - Distribution to use for weight initialization

biasInit

public T biasInit(double biasInit)

Bias initialization value, for layers with biases. Defaults to 0

Parameters:

biasInit - Value to use for initializing biases

dist

public T dist(Distribution dist)

Distribution to sample initial weights from. Used in conjunction with .weightInit(WeightInit.DISTRIBUTION).

l1

public T l1(double l1)

L1 regularization coefficient (weights only). Use l1Bias(double) to configure the l1 regularization coefficient for the bias.

l2

public T l2(double l2)

L2 regularization coefficient (weights only). Use l2Bias(double) to configure the l2 regularization coefficient for the bias.

l1Bias

public T l1Bias(double l1Bias)

L1 regularization coefficient for the bias. Default: 0. See also l1(double)

l2Bias

public T l2Bias(double l2Bias)

L2 regularization coefficient for the bias. Default: 0. See also l2(double)

updater

@Deprecated
public T updater(Updater updater)

Deprecated.

Gradient updater. For example, SGD for standard stochastic gradient descent, NESTEROV for Nesterov momentum, RSMPROP for RMSProp, etc.

See Also:

Updater

updater

public T updater(org.nd4j.linalg.learning.config.IUpdater updater)

Gradient updater. For example, Adam or Nesterovs

Parameters:

updater - Updater to use

biasUpdater

public T biasUpdater(org.nd4j.linalg.learning.config.IUpdater biasUpdater)

Gradient updater configuration, for the biases only. If not set, biases will use the updater as set by updater(IUpdater)

Parameters:

biasUpdater - Updater to use for bias parameters

gradientNormalization

public T gradientNormalization(GradientNormalization gradientNormalization)

Gradient normalization strategy. Used to specify gradient renormalization, gradient clipping etc.

Parameters:

gradientNormalization - Type of normalization to use. Defaults to None.

See Also:

GradientNormalization

gradientNormalizationThreshold

public T gradientNormalizationThreshold(double threshold)

Threshold for gradient normalization, only used for GradientNormalization.ClipL2PerLayer, GradientNormalization.ClipL2PerParamType, and GradientNormalization.ClipElementWiseAbsoluteValue
Not used otherwise.
L2 threshold for first two types of clipping, or absolute value threshold for last type of clipping.

weightNoise

public T weightNoise(IWeightNoise weightNoise)

Set the weight noise (such as DropConnect and WeightNoise) for this layer

Parameters:

weightNoise - Weight noise instance to use