org.deeplearning4j.nn.conf

Class MultiLayerConfiguration.Builder

java.lang.Object

org.deeplearning4j.nn.conf.MultiLayerConfiguration.Builder

Direct Known Subclasses:

NeuralNetConfiguration.ListBuilder

Enclosing class:

MultiLayerConfiguration

public static class MultiLayerConfiguration.Builder
extends Object

Field Summary

Fields

Modifier and Type	Field and Description
protected boolean	backward
protected Map<Integer,ConfOverride>	confOverrides
protected List<NeuralNetConfiguration>	confs
protected double	dampingFactor
protected int[]	hiddenLayerSizes
protected Map<Integer,InputPreProcessor>	inputPreProcessor
protected Map<Integer,OutputPreProcessor>	preProcessors
protected boolean	pretrain
protected boolean	useDropConnect
protected boolean	useRBMPropUpAsActivations

Constructor Summary

Constructors

Constructor and Description
Builder()

Method Summary

Modifier and Type	Method and Description
MultiLayerConfiguration.Builder	backward(boolean backward) Whether to do back prop or not
MultiLayerConfiguration	build()
MultiLayerConfiguration.Builder	confs(List<NeuralNetConfiguration> confs)
MultiLayerConfiguration.Builder	dampingFactor(double dampingFactor)
boolean	equals(Object o)
int	hashCode()
MultiLayerConfiguration.Builder	hiddenLayerSizes(int... hiddenLayerSizes) Specify the hidden layer sizes.
MultiLayerConfiguration.Builder	inputPreProcessor(Integer layer, InputPreProcessor preProcessor)
MultiLayerConfiguration.Builder	inputPreProcessors(Map<Integer,InputPreProcessor> inputPreProcessor) Specify the input pre processors.
MultiLayerConfiguration.Builder	override(ConfOverride override)
MultiLayerConfiguration.Builder	override(int layer, ConfOverride override)
MultiLayerConfiguration.Builder	preProcessor(Integer layer, OutputPreProcessor preProcessor)
MultiLayerConfiguration.Builder	preProcessors(Map<Integer,OutputPreProcessor> preProcessors)
MultiLayerConfiguration.Builder	pretrain(boolean pretrain) Whether to do pre train or not
String	toString()
MultiLayerConfiguration.Builder	useDropConnect(boolean useDropConnect) Whether to use drop connect or not
MultiLayerConfiguration.Builder	useRBMPropUpAsActivations(boolean useRBMPropUpAsActivations)

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, wait, wait, wait

Field Detail

confs

protected List<NeuralNetConfiguration> confs

hiddenLayerSizes

protected int[] hiddenLayerSizes

useDropConnect

protected boolean useDropConnect

pretrain

protected boolean pretrain

useRBMPropUpAsActivations

protected boolean useRBMPropUpAsActivations

dampingFactor

protected double dampingFactor

preProcessors

protected Map<Integer,OutputPreProcessor> preProcessors

inputPreProcessor

protected Map<Integer,InputPreProcessor> inputPreProcessor

backward

protected boolean backward

confOverrides

protected Map<Integer,ConfOverride> confOverrides

Constructor Detail

Builder

public Builder()

Method Detail

inputPreProcessors

public MultiLayerConfiguration.Builder inputPreProcessors(Map<Integer,InputPreProcessor> inputPreProcessor)

Specify the input pre processors. These are used at each layer for doing things like normalization and shaping of input.

Parameters:

inputPreProcessor - the input pre processor to use.

Returns:

builder pattern

backward

public MultiLayerConfiguration.Builder backward(boolean backward)

Whether to do back prop or not

Parameters:

backward - whether to do back prop or not

Returns:

inputPreProcessor

public MultiLayerConfiguration.Builder inputPreProcessor(Integer layer,
                                                         InputPreProcessor preProcessor)

preProcessor

public MultiLayerConfiguration.Builder preProcessor(Integer layer,
                                                    OutputPreProcessor preProcessor)

preProcessors

public MultiLayerConfiguration.Builder preProcessors(Map<Integer,OutputPreProcessor> preProcessors)

dampingFactor

public MultiLayerConfiguration.Builder dampingFactor(double dampingFactor)

useRBMPropUpAsActivations

public MultiLayerConfiguration.Builder useRBMPropUpAsActivations(boolean useRBMPropUpAsActivations)

pretrain

public MultiLayerConfiguration.Builder pretrain(boolean pretrain)

Whether to do pre train or not

Parameters:

pretrain - whether to do pre train or not

Returns:

builder pattern

useDropConnect

public MultiLayerConfiguration.Builder useDropConnect(boolean useDropConnect)

Whether to use drop connect or not

Parameters:

useDropConnect - true if drop connect should applied or not

Returns:

builder pattern

confs

public MultiLayerConfiguration.Builder confs(List<NeuralNetConfiguration> confs)

hiddenLayerSizes

public MultiLayerConfiguration.Builder hiddenLayerSizes(int... hiddenLayerSizes)

Specify the hidden layer sizes. Note that you can specify the layer sizes in continuous order. Whereever number of inputs and outputs are used this will be set for intermediate layers

Parameters:

hiddenLayerSizes - the hidden layer sizes to use

Returns:

build

public MultiLayerConfiguration build()

toString

public String toString()

Overrides:

toString in class Object

equals

public boolean equals(Object o)

Overrides:

equals in class Object

hashCode

public int hashCode()

Overrides:

hashCode in class Object

override

public MultiLayerConfiguration.Builder override(ConfOverride override)

override

public MultiLayerConfiguration.Builder override(int layer,
                                                ConfOverride override)