org.deeplearning4j.nn.conf.layers

Class Layer.Builder<T extends Layer.Builder<T>>

java.lang.Object

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

Direct Known Subclasses:

AbstractSameDiffLayer.Builder, ActivationLayer.Builder, BaseLayer.Builder, BaseUpsamplingLayer.UpsamplingBuilder, Bidirectional.Builder, Cropping1D.Builder, Cropping2D.Builder, Cropping3D.Builder, FrozenLayer.Builder, GlobalPoolingLayer.Builder, LocalResponseNormalization.Builder, SpaceToBatchLayer.Builder, SpaceToDepthLayer.Builder, Subsampling3DLayer.BaseSubsamplingBuilder, SubsamplingLayer.BaseSubsamplingBuilder, Yolo2OutputLayer.Builder, ZeroPadding1DLayer.Builder, ZeroPadding3DLayer.Builder, ZeroPaddingLayer.Builder

Enclosing class:

Layer

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

Field Summary

Fields

Modifier and Type	Field and Description
protected List<LayerConstraint>	allParamConstraints
protected List<LayerConstraint>	biasConstraints
protected IDropout	iDropout
protected String	layerName
protected List<LayerConstraint>	weightConstraints

Constructor Summary

Constructors

Constructor and Description
Builder()

Method Summary

Modifier and Type	Method and Description
abstract <E extends Layer> E	build()
T	constrainAllParameters(LayerConstraint... constraints) Set constraints to be applied to this layer.
T	constrainBias(LayerConstraint... constraints) Set constraints to be applied to bias parameters of this layer.
T	constrainWeights(LayerConstraint... constraints) Set constraints to be applied to the weight parameters of this layer.
T	dropOut(double inputRetainProbability) Dropout probability.
T	dropOut(IDropout dropout) Set the dropout for all layers in this network
T	name(String layerName) Layer name assigns layer string name.

Methods inherited from class java.lang.Object

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

Field Detail

layerName

protected String layerName

allParamConstraints

protected List<LayerConstraint> allParamConstraints

weightConstraints

protected List<LayerConstraint> weightConstraints

biasConstraints

protected List<LayerConstraint> biasConstraints

iDropout

protected IDropout iDropout

Constructor Detail

Builder

public Builder()

Method Detail

name

public T name(String layerName)

Layer name assigns layer string name. Allows easier differentiation between layers.

dropOut

public T dropOut(double inputRetainProbability)

Dropout probability. This is the probability of retaining each input activation value for a layer. dropOut(x) will keep an input activation with probability x, and set to 0 with probability 1-x.
dropOut(0.0) is a special value / special case - when set to 0.0., dropout is disabled (not applied). Note that a dropout value of 1.0 is functionally equivalent to no dropout: i.e., 100% probability of retaining each input activation.
When useDropConnect(boolean) is set to true (false by default), this method sets the drop connect probability instead.

Note 1: Dropout is applied at training time only - and is automatically not applied at test time (for evaluation, etc)
Note 2: This sets the probability per-layer. Care should be taken when setting lower values for complex networks (too much information may be lost with aggressive (very low) dropout values).
Note 3: Frequently, dropout is not applied to (or, has higher retain probability for) input (first layer) layers. Dropout is also often not applied to output layers. This needs to be handled MANUALLY by the user - set .dropout(0) on those layers when using global dropout setting.
Note 4: Implementation detail (most users can ignore): DL4J uses inverted dropout, as described here: http://cs231n.github.io/neural-networks-2/

Parameters:

inputRetainProbability - Dropout probability (probability of retaining each input activation value for a layer)

See Also:

dropOut(IDropout)

dropOut

public T dropOut(IDropout dropout)

Set the dropout for all layers in this network

Parameters:

dropout - Dropout, such as Dropout, GaussianDropout, GaussianNoise etc

constrainAllParameters

public T constrainAllParameters(LayerConstraint... constraints)

Set constraints to be applied to this layer. Default: no constraints.
Constraints can be used to enforce certain conditions (non-negativity of parameters, max-norm regularization, etc). These constraints are applied at each iteration, after the parameters have been updated.

Parameters:

constraints - Constraints to apply to all parameters of this layer

constrainBias

public T constrainBias(LayerConstraint... constraints)

Set constraints to be applied to bias parameters of this layer. Default: no constraints.
Constraints can be used to enforce certain conditions (non-negativity of parameters, max-norm regularization, etc). These constraints are applied at each iteration, after the parameters have been updated.

Parameters:

constraints - Constraints to apply to all bias parameters of this layer

constrainWeights

public T constrainWeights(LayerConstraint... constraints)

Set constraints to be applied to the weight parameters of this layer. Default: no constraints.
Constraints can be used to enforce certain conditions (non-negativity of parameters, max-norm regularization, etc). These constraints are applied at each iteration, after the parameters have been updated.

Parameters:

constraints - Constraints to apply to all weight parameters of this layer

build

public abstract <E extends Layer> E build()