Implementation of the Atrous Convolution class. More...

Inheritance diagram for AtrousConvolution< ForwardConvolutionRule, BackwardConvolutionRule, GradientConvolutionRule, InputType, OutputType >:

[legend]

Public Member Functions
	AtrousConvolution ()
	Create the AtrousConvolution object. More...

	AtrousConvolution (const size_t inSize, const size_t outSize, const size_t kernelWidth, const size_t kernelHeight, const size_t strideWidth=1, const size_t strideHeight=1, const size_t padW=0, const size_t padH=0, const size_t inputWidth=0, const size_t inputHeight=0, const size_t dilationWidth=1, const size_t dilationHeight=1, const std::string &paddingType="None")
	Create the AtrousConvolution object using the specified number of input maps, output maps, filter size, stride, dilation and padding parameter. More...

	AtrousConvolution (const size_t inSize, const size_t outSize, const size_t kernelWidth, const size_t kernelHeight, const size_t strideWidth, const size_t strideHeight, const std::tuple< size_t, size_t > &padW, const std::tuple< size_t, size_t > &padH, const size_t inputWidth=0, const size_t inputHeight=0, const size_t dilationWidth=1, const size_t dilationHeight=1, const std::string &paddingType="None")
	Create the AtrousConvolution object using the specified number of input maps, output maps, filter size, stride, dilation and padding parameter. More...

void	Backward (const InputType &, const OutputType &gy, OutputType &g)
	Ordinary feed backward pass of a neural network, calculating the function f(x) by propagating x backwards through f. More...

const OutputType &	Bias () const
	Get the bias of the layer. More...

OutputType &	Bias ()
	Modify the bias of the layer. More...

const size_t &	DilationHeight () const
	Get the dilation rate on the Y axis. More...

size_t &	DilationHeight ()
	Modify the dilation rate on the Y axis. More...

const size_t &	DilationWidth () const
	Get the dilation rate on the X axis. More...

size_t &	DilationWidth ()
	Modify the dilation rate on the X axis. More...

void	Forward (const InputType &input, OutputType &output)
	Ordinary feed forward pass of a neural network, evaluating the function f(x) by propagating the activity forward through f. More...

void	Gradient (const InputType &, const OutputType &error, OutputType &gradient)

const size_t &	InputHeight () const
	Get the input height. More...

size_t &	InputHeight ()
	Modify the input height. More...

size_t	InputShape () const
	Get the shape of the input. More...

const size_t &	InputWidth () const
	Get the input width. More...

size_t &	InputWidth ()
	Modify input the width. More...

const size_t &	KernelHeight () const
	Get the kernel height. More...

size_t &	KernelHeight ()
	Modify the kernel height. More...

const size_t &	KernelWidth () const
	Get the kernel width. More...

size_t &	KernelWidth ()
	Modify the kernel width. More...

const std::vector< size_t > &	OutputDimensions () const

const size_t &	OutputHeight () const
	Get the output height. More...

size_t &	OutputHeight ()
	Modify the output height. More...

const size_t &	OutputWidth () const
	Get the output width. More...

size_t &	OutputWidth ()
	Modify the output width. More...

PaddingType< InputType, OutputType > const &	Padding () const
	Get the internal Padding layer. More...

PaddingType< InputType, OutputType > &	Padding ()
	Modify the internal Padding layer. More...

OutputType const &	Parameters () const
	Get the parameters. More...

OutputType &	Parameters ()
	Modify the parameters. More...

void	Reset ()

template < typename Archive >
void	serialize (Archive &ar, const uint32_t)
	Serialize the layer. More...

const size_t &	StrideHeight () const
	Get the stride height. More...

size_t &	StrideHeight ()
	Modify the stride height. More...

const size_t &	StrideWidth () const
	Get the stride width. More...

size_t &	StrideWidth ()
	Modify the stride width. More...

const arma::Cube< typename OutputType::elem_type > &	Weight () const
	Get the weight of the layer. More...

arma::Cube< typename OutputType::elem_type > &	Weight ()
	Modify the weight of the layer. More...

size_t	WeightSize () const
	Get size of the weight matrix. More...

Public Member Functions inherited from Layer< InputType, OutputType >
	Layer ()
	Default constructor. More...

	Layer (const Layer &layer)
	Copy constructor. This is not responsible for copying weights! More...

	Layer (Layer &&layer)
	Move constructor. This is not responsible for moving weights! More...

virtual	~Layer ()
	Default deconstructor. More...

virtual void	Backward (const InputType &, const InputType &, InputType &)
	Performs a backpropagation step through the layer, with respect to the given input. More...

virtual Layer *	Clone () const=0
	Make a copy of the object. More...

virtual void	ComputeOutputDimensions ()
	Compute the output dimensions. More...

virtual void	Forward (const InputType &, InputType &)
	Takes an input object, and computes the corresponding output of the layer. More...

virtual void	Forward (const InputType &, const InputType &)
	Takes an input and output object, and computes the corresponding loss of the layer. More...

virtual void	Gradient (const InputType &, const InputType &, InputType &)
	Computing the gradient of the layer with respect to its own input. More...

const std::vector< size_t > &	InputDimensions () const
	Get the input dimensions. More...

std::vector< size_t > &	InputDimensions ()
	Modify the input dimensions. More...

virtual double	Loss ()
	Get the layer loss. More...

virtual Layer &	operator= (const Layer &layer)
	Copy assignment operator. This is not responsible for copying weights! More...

virtual Layer &	operator= (Layer &&layer)
	Move assignment operator. This is not responsible for moving weights! More...

const std::vector< size_t > &	OutputDimensions ()
	Get the output dimensions. More...

virtual size_t	OutputSize () final
	Get the number of elements in the output from this layer. More...

void	serialize (Archive &ar, const uint32_t)
	Serialize the layer. More...

virtual void	SetWeights (typename InputType ::elem_type *)
	Reset the layer parameter. More...

virtual bool const &	Training () const
	Get whether the layer is currently in training mode. More...

virtual bool &	Training ()
	Modify whether the layer is currently in training mode. More...

Additional Inherited Members
Protected Attributes inherited from Layer< InputType, OutputType >
std::vector< size_t >	inputDimensions
	Logical input dimensions of each point. More...

std::vector< size_t >	outputDimensions
	Logical output dimensions of each point. More...

bool	training
	If true, the layer is in training mode; otherwise, it is in testing mode. More...

bool	validOutputDimensions
	This is `true` if `ComputeOutputDimensions()` has been called, and `outputDimensions` can be considered to be up-to-date. More...

Detailed Description

template<typename ForwardConvolutionRule = NaiveConvolution<ValidConvolution>, typename BackwardConvolutionRule = NaiveConvolution<FullConvolution>, typename GradientConvolutionRule = NaiveConvolution<ValidConvolution>, typename InputType = arma::mat, typename OutputType = arma::mat>
class mlpack::ann::AtrousConvolution< ForwardConvolutionRule, BackwardConvolutionRule, GradientConvolutionRule, InputType, OutputType >

Implementation of the Atrous Convolution class.

The Atrous Convolution class represents a single layer of a neural network. Atrous (or Dilated) Convolutions are just simple convolutions applied to input with the defined, spaces included between the kernel cells, in order to capture a larger field of reception, without having to increase dicrete kernel sizes.

Template Parameters

ForwardConvolutionRule	Atrous Convolution to perform forward process.
BackwardConvolutionRule	Atrous Convolution to perform backward process.
GradientConvolutionRule	Atrous Convolution to calculate gradient.
InputDataType	Type of the input data (arma::colvec, arma::mat, arma::sp_mat or arma::cube).
OutputDataType	Type of the output data (arma::colvec, arma::mat, arma::sp_mat or arma::cube).

Definition at line 53 of file atrous_convolution.hpp.

Constructor & Destructor Documentation

◆ AtrousConvolution() [1/3]

AtrousConvolution ( )

Create the AtrousConvolution object.

◆ AtrousConvolution() [2/3]

AtrousConvolution	(	const size_t	inSize,
		const size_t	outSize,
		const size_t	kernelWidth,
		const size_t	kernelHeight,
		const size_t	strideWidth = `1`,
		const size_t	strideHeight = `1`,
		const size_t	padW = `0`,
		const size_t	padH = `0`,
		const size_t	inputWidth = `0`,
		const size_t	inputHeight = `0`,
		const size_t	dilationWidth = `1`,
		const size_t	dilationHeight = `1`,
		const std::string &	paddingType = `"None"`
	)

Create the AtrousConvolution object using the specified number of input maps, output maps, filter size, stride, dilation and padding parameter.

Parameters

inSize	The number of input maps.
outSize	The number of output maps.
kernelWidth	Width of the filter/kernel.
kernelHeight	Height of the filter/kernel.
strideWidth	Stride of filter application in the x direction.
strideHeight	Stride of filter application in the y direction.
padW	Padding width of the input.
padH	Padding height of the input.
inputWidth	The widht of the input data.
inputHeight	The height of the input data.
dilationWidth	The space between the cells of filters in x direction.
dilationHeight	The space between the cells of filters in y direction.
paddingType	The type of padding (Valid or Same). Defaults to None.

◆ AtrousConvolution() [3/3]

AtrousConvolution	(	const size_t	inSize,
		const size_t	outSize,
		const size_t	kernelWidth,
		const size_t	kernelHeight,
		const size_t	strideWidth,
		const size_t	strideHeight,
		const std::tuple< size_t, size_t > &	padW,
		const std::tuple< size_t, size_t > &	padH,
		const size_t	inputWidth = `0`,
		const size_t	inputHeight = `0`,
		const size_t	dilationWidth = `1`,
		const size_t	dilationHeight = `1`,
		const std::string &	paddingType = `"None"`
	)

Create the AtrousConvolution object using the specified number of input maps, output maps, filter size, stride, dilation and padding parameter.

Parameters

inSize	The number of input maps.
outSize	The number of output maps.
kernelWidth	Width of the filter/kernel.
kernelHeight	Height of the filter/kernel.
strideWidth	Stride of filter application in the x direction.
strideHeight	Stride of filter application in the y direction.
padW	A two-value tuple indicating padding widths of the input. First value is padding at left side. Second value is padding on right side.
padH	A two-value tuple indicating padding heights of the input. First value is padding at top. Second value is padding on bottom.
inputWidth	The widht of the input data.
inputHeight	The height of the input data.
dilationWidth	The space between the cells of filters in x direction.
dilationHeight	The space between the cells of filters in y direction.
paddingType	The type of padding (Valid/Same/None). Defaults to None.

Member Function Documentation

◆ Backward()

void Backward	(	const InputType &	,
		const OutputType &	gy,
		OutputType &	g
	)

Ordinary feed backward pass of a neural network, calculating the function f(x) by propagating x backwards through f.

Using the results from the feed forward pass.

Parameters

*	(input) The propagated input activation.
gy	The backpropagated error.
g	The calculated gradient.

◆ Bias() [1/2]

const OutputType& Bias ( ) const

inline

Get the bias of the layer.

Definition at line 191 of file atrous_convolution.hpp.

◆ Bias() [2/2]

OutputType& Bias ( )

inline

Modify the bias of the layer.

Definition at line 193 of file atrous_convolution.hpp.

◆ DilationHeight() [1/2]

const size_t& DilationHeight ( ) const

inline

Get the dilation rate on the Y axis.

Definition at line 241 of file atrous_convolution.hpp.

◆ DilationHeight() [2/2]

size_t& DilationHeight ( )

inline

Modify the dilation rate on the Y axis.

Definition at line 243 of file atrous_convolution.hpp.

◆ DilationWidth() [1/2]

const size_t& DilationWidth ( ) const

inline

Get the dilation rate on the X axis.

Definition at line 236 of file atrous_convolution.hpp.

◆ DilationWidth() [2/2]

size_t& DilationWidth ( )

inline

Modify the dilation rate on the X axis.

Definition at line 238 of file atrous_convolution.hpp.

◆ Forward()

void Forward	(	const InputType &	input,
		OutputType &	output
	)

Ordinary feed forward pass of a neural network, evaluating the function f(x) by propagating the activity forward through f.

Parameters

input	Input data used for evaluating the specified function.
output	Resulting output activation.

◆ Gradient()

void Gradient	(	const InputType &	,
		const OutputType &	error,
		OutputType &	gradient
	)

◆ InputHeight() [1/2]

const size_t& InputHeight ( ) const

inline

Get the input height.

Definition at line 201 of file atrous_convolution.hpp.

◆ InputHeight() [2/2]

size_t& InputHeight ( )

inline

Modify the input height.

Definition at line 203 of file atrous_convolution.hpp.

◆ InputShape()

size_t InputShape ( ) const

inline

Get the shape of the input.

Definition at line 257 of file atrous_convolution.hpp.

References AtrousConvolution< ForwardConvolutionRule, BackwardConvolutionRule, GradientConvolutionRule, InputType, OutputType >::serialize(), and core::v2::size().

◆ InputWidth() [1/2]

const size_t& InputWidth ( ) const

inline

Get the input width.

Definition at line 196 of file atrous_convolution.hpp.

◆ InputWidth() [2/2]

size_t& InputWidth ( )

inline

Modify input the width.

Definition at line 198 of file atrous_convolution.hpp.

◆ KernelHeight() [1/2]

const size_t& KernelHeight ( ) const

inline

Get the kernel height.

Definition at line 221 of file atrous_convolution.hpp.

◆ KernelHeight() [2/2]

size_t& KernelHeight ( )

inline

Modify the kernel height.

Definition at line 223 of file atrous_convolution.hpp.

◆ KernelWidth() [1/2]

const size_t& KernelWidth ( ) const

inline

Get the kernel width.

Definition at line 216 of file atrous_convolution.hpp.

◆ KernelWidth() [2/2]

size_t& KernelWidth ( )

inline

Modify the kernel width.

Definition at line 218 of file atrous_convolution.hpp.

◆ OutputDimensions()

const std::vector<size_t>& OutputDimensions ( ) const

inline

Definition at line 182 of file atrous_convolution.hpp.

References Layer< InputType, OutputType >::inputDimensions.

◆ OutputHeight() [1/2]

const size_t& OutputHeight ( ) const

inline

Get the output height.

Definition at line 211 of file atrous_convolution.hpp.

◆ OutputHeight() [2/2]

size_t& OutputHeight ( )

inline

Modify the output height.

Definition at line 213 of file atrous_convolution.hpp.

◆ OutputWidth() [1/2]

const size_t& OutputWidth ( ) const

inline

Get the output width.

Definition at line 206 of file atrous_convolution.hpp.

◆ OutputWidth() [2/2]

size_t& OutputWidth ( )

inline

Modify the output width.

Definition at line 208 of file atrous_convolution.hpp.

◆ Padding() [1/2]

PaddingType<InputType, OutputType> const& Padding ( ) const

inline

Get the internal Padding layer.

Definition at line 246 of file atrous_convolution.hpp.

◆ Padding() [2/2]

PaddingType<InputType, OutputType>& Padding ( )

inline

Modify the internal Padding layer.

Definition at line 248 of file atrous_convolution.hpp.

◆ Parameters() [1/2]

OutputType const& Parameters ( ) const

inlinevirtual

Get the parameters.

Reimplemented from Layer< InputType, OutputType >.

Definition at line 170 of file atrous_convolution.hpp.

◆ Parameters() [2/2]

OutputType& Parameters ( )

inlinevirtual

Modify the parameters.

Reimplemented from Layer< InputType, OutputType >.

Definition at line 172 of file atrous_convolution.hpp.

◆ Reset()

void Reset ( )

◆ serialize()

void serialize	(	Archive &	ar,
		const uint32_t
	)

Serialize the layer.

Referenced by AtrousConvolution< ForwardConvolutionRule, BackwardConvolutionRule, GradientConvolutionRule, InputType, OutputType >::InputShape().

◆ StrideHeight() [1/2]

const size_t& StrideHeight ( ) const

inline

Get the stride height.

Definition at line 231 of file atrous_convolution.hpp.

◆ StrideHeight() [2/2]

size_t& StrideHeight ( )

inline

Modify the stride height.

Definition at line 233 of file atrous_convolution.hpp.

◆ StrideWidth() [1/2]

const size_t& StrideWidth ( ) const

inline

Get the stride width.

Definition at line 226 of file atrous_convolution.hpp.

◆ StrideWidth() [2/2]

size_t& StrideWidth ( )

inline

Modify the stride width.

Definition at line 228 of file atrous_convolution.hpp.

◆ Weight() [1/2]

const arma::Cube<typename OutputType::elem_type>& Weight ( ) const

inline

Get the weight of the layer.

Definition at line 175 of file atrous_convolution.hpp.

◆ Weight() [2/2]

arma::Cube<typename OutputType::elem_type>& Weight ( )

inline

Modify the weight of the layer.

Definition at line 180 of file atrous_convolution.hpp.

◆ WeightSize()

size_t WeightSize ( ) const

inlinevirtual

Get size of the weight matrix.

Reimplemented from Layer< InputType, OutputType >.

Definition at line 251 of file atrous_convolution.hpp.

The documentation for this class was generated from the following file:

/home/jenkins-mlpack/mlpack.org/_src/mlpack-git/src/mlpack/methods/ann/layer/not_adapted/atrous_convolution.hpp

Public Member Functions

Additional Inherited Members

Detailed Description

Constructor & Destructor Documentation

◆ AtrousConvolution() [1/3]

◆ AtrousConvolution() [2/3]

◆ AtrousConvolution() [3/3]

Member Function Documentation

◆ Backward()

◆ Bias() [1/2]

◆ Bias() [2/2]

◆ DilationHeight() [1/2]

◆ DilationHeight() [2/2]

◆ DilationWidth() [1/2]

◆ DilationWidth() [2/2]

◆ Forward()

◆ Gradient()

◆ InputHeight() [1/2]

◆ InputHeight() [2/2]

◆ InputShape()

◆ InputWidth() [1/2]

◆ InputWidth() [2/2]

◆ KernelHeight() [1/2]

◆ KernelHeight() [2/2]

◆ KernelWidth() [1/2]

◆ KernelWidth() [2/2]

◆ OutputDimensions()

◆ OutputHeight() [1/2]

◆ OutputHeight() [2/2]

◆ OutputWidth() [1/2]

◆ OutputWidth() [2/2]

◆ Padding() [1/2]

◆ Padding() [2/2]

◆ Parameters() [1/2]

◆ Parameters() [2/2]

◆ Reset()

◆ serialize()

◆ StrideHeight() [1/2]

◆ StrideHeight() [2/2]

◆ StrideWidth() [1/2]

◆ StrideWidth() [2/2]

◆ Weight() [1/2]

◆ Weight() [2/2]

◆ WeightSize()