Package org.deeplearning4j.nn.conf.layers

Class SubsamplingLayer.Builder

java.lang.Object

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

org.deeplearning4j.nn.conf.layers.SubsamplingLayer.BaseSubsamplingBuilder<SubsamplingLayer.Builder>

org.deeplearning4j.nn.conf.layers.SubsamplingLayer.Builder

Enclosing class:

SubsamplingLayer

public static class SubsamplingLayer.Builder
extends SubsamplingLayer.BaseSubsamplingBuilder<SubsamplingLayer.Builder>

Field Summary

Fields inherited from class org.deeplearning4j.nn.conf.layers.SubsamplingLayer.BaseSubsamplingBuilder

avgPoolIncludePadInDivisor, cnn2DFormat, convolutionMode, cudnnAllowFallback, eps, kernelSize, padding, pnorm, poolingType, stride

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

allParamConstraints, biasConstraints, iDropout, layerName, weightConstraints

Constructor Summary

Constructors

Constructor	Description
Builder(int... kernelSize)
Builder(int[] kernelSize, int[] stride)
Builder(int[] kernelSize, int[] stride, int[] padding)
Builder(PoolingType poolingType)
Builder(PoolingType poolingType, int[] kernelSize)
Builder(PoolingType poolingType, int[] kernelSize, int[] stride, int[] padding)
Builder(SubsamplingLayer.PoolingType poolingType)
Builder(SubsamplingLayer.PoolingType poolingType, int[] kernelSize)
Builder(SubsamplingLayer.PoolingType poolingType, int[] kernelSize, int[] stride)
Builder(SubsamplingLayer.PoolingType poolingType, int[] kernelSize, int[] stride, int[] padding)

Method Summary

Modifier and Type	Method	Description
protected boolean	allowCausal()
SubsamplingLayer	build()
SubsamplingLayer.Builder	dilation(int... dilation)	Kernel dilation.
SubsamplingLayer.Builder	kernelSize(int... kernelSize)	Kernel size
SubsamplingLayer.Builder	padding(int... padding)	Padding
void	setDataFormat(CNN2DFormat format)
void	setDilation(int[] dilation)
void	setKernelSize(int... kernelSize)
void	setPadding(int... padding)
void	setStride(int... stride)
SubsamplingLayer.Builder	stride(int... stride)	Stride

Methods inherited from class org.deeplearning4j.nn.conf.layers.SubsamplingLayer.BaseSubsamplingBuilder

avgPoolIncludePadInDivisor, convolutionMode, cudnnAllowFallback, dataFormat, eps, helperAllowFallback, pnorm, poolingType, poolingType, setConvolutionMode, setEps, setPnorm

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(SubsamplingLayer.PoolingType poolingType,
               int[] kernelSize,
               int[] stride)

Builder

public Builder(SubsamplingLayer.PoolingType poolingType,
               int[] kernelSize)

Builder

public Builder(SubsamplingLayer.PoolingType poolingType,
               int[] kernelSize,
               int[] stride,
               int[] padding)

Builder

public Builder(PoolingType poolingType,
               int[] kernelSize)

Builder

public Builder(PoolingType poolingType,
               int[] kernelSize,
               int[] stride,
               int[] padding)

Builder

public Builder(int[] kernelSize,
               int[] stride,
               int[] padding)

Builder

public Builder(int[] kernelSize,
               int[] stride)

Builder

public Builder(int... kernelSize)

Builder

public Builder(SubsamplingLayer.PoolingType poolingType)

Builder

public Builder(PoolingType poolingType)

Method Detail

allowCausal

protected boolean allowCausal()

Specified by:

allowCausal in class SubsamplingLayer.BaseSubsamplingBuilder<SubsamplingLayer.Builder>

kernelSize

public SubsamplingLayer.Builder kernelSize(int... kernelSize)

Kernel size

Parameters:

kernelSize - kernel size in height and width dimensions

stride

public SubsamplingLayer.Builder stride(int... stride)

Stride

Parameters:

stride - stride in height and width dimensions

padding

public SubsamplingLayer.Builder padding(int... padding)

Padding

Parameters:

padding - padding in the height and width dimensions

dilation

public SubsamplingLayer.Builder dilation(int... dilation)

Kernel dilation. Default: {1, 1}, which is standard convolutions. Used for implementing dilated convolutions, which are also known as atrous convolutions.
NOTE: Kernel dilation is less common in practice for subsampling layers, compared to convolutional layers. For more details, see: Yu and Koltun (2014) and Chen et al. (2014), as well as http://deeplearning.net/software/theano/tutorial/conv_arithmetic.html#dilated-convolutions

Parameters:

dilation - Dilation for kernel

build

public SubsamplingLayer build()

Specified by:

build in class Layer.Builder<SubsamplingLayer.Builder>

setKernelSize

public void setKernelSize(int... kernelSize)

setStride

public void setStride(int... stride)

setPadding

public void setPadding(int... padding)

setDilation

public void setDilation(int[] dilation)

setDataFormat

public void setDataFormat(CNN2DFormat format)