Logo ROOT   6.16/01
Reference Guide
List of all members | Public Types | Public Member Functions | Private Attributes | List of all members
TMVA::DNN::CNN::TMaxPoolLayer< Architecture_t > Class Template Reference

template<typename Architecture_t>
class TMVA::DNN::CNN::TMaxPoolLayer< Architecture_t >

Generic Max Pooling Layer class.

This generic Max Pooling Layer Class represents a pooling layer of a CNN. It inherits all of the properties of the convolutional layer TConvLayer, but it overrides the propagation methods. In a sense, max pooling can be seen as non-linear convolution: a filter slides over the input and produces one element as a function of the the elements within the receptive field. In addition to that, it contains a matrix of winning units.

The height and width of the weights and biases is set to 0, since this layer does not contain any weights.

Definition at line 57 of file MaxPoolLayer.h.

Public Types

using Matrix_t = typename Architecture_t::Matrix_t
 
using Scalar_t = typename Architecture_t::Scalar_t
 
- Public Types inherited from TMVA::DNN::CNN::TConvLayer< Architecture_t >
using Matrix_t = typename Architecture_t::Matrix_t
 
using Scalar_t = typename Architecture_t::Scalar_t
 

Public Member Functions

 TMaxPoolLayer (const TMaxPoolLayer &)
 Copy constructor. More...
 
 TMaxPoolLayer (size_t BatchSize, size_t InputDepth, size_t InputHeight, size_t InputWidth, size_t FilterHeight, size_t FilterWidth, size_t StrideRows, size_t StrideCols, Scalar_t DropoutProbability)
 Constructor. More...
 
 TMaxPoolLayer (TMaxPoolLayer< Architecture_t > *layer)
 Copy the max pooling layer provided as a pointer. More...
 
 ~TMaxPoolLayer ()
 Destructor. More...
 
virtual void AddWeightsXMLTo (void *parent)
 Writes the information and the weights about the layer in an XML node. More...
 
void Backward (std::vector< Matrix_t > &gradients_backward, const std::vector< Matrix_t > &activations_backward, std::vector< Matrix_t > &inp1, std::vector< Matrix_t > &inp2)
 Depending on the winning units determined during the Forward pass, it only forwards the derivatives to the right units in the previous layer. More...
 
void Forward (std::vector< Matrix_t > &input, bool applyDropout=false)
 Computes activation of the layer for the given input. More...
 
std::vector< Matrix_t > & GetIndexMatrix ()
 
const std::vector< Matrix_t > & GetIndexMatrix () const
 Getters. More...
 
void Print () const
 Prints the info about the layer. More...
 
virtual void ReadWeightsFromXML (void *parent)
 Read the information and the weights about the layer from XML node. More...
 
- Public Member Functions inherited from TMVA::DNN::CNN::TConvLayer< Architecture_t >
 TConvLayer (const TConvLayer &)
 Copy constructor. More...
 
 TConvLayer (size_t BatchSize, size_t InputDepth, size_t InputHeight, size_t InputWidth, size_t Depth, EInitialization Init, size_t FilterHeight, size_t FilterWidth, size_t StrideRows, size_t StrideCols, size_t PaddingHeight, size_t PaddingWidth, Scalar_t DropoutProbability, EActivationFunction f, ERegularization Reg, Scalar_t WeightDecay)
 Constructor. More...
 
 TConvLayer (TConvLayer< Architecture_t > *layer)
 Copy the conv layer provided as a pointer. More...
 
 ~TConvLayer ()
 Destructor. More...
 
virtual void AddWeightsXMLTo (void *parent)
 Writes the information and the weights about the layer in an XML node. More...
 
void Backward (std::vector< Matrix_t > &gradients_backward, const std::vector< Matrix_t > &activations_backward, std::vector< Matrix_t > &inp1, std::vector< Matrix_t > &inp2)
 Compute weight, bias and activation gradients. More...
 
void Forward (std::vector< Matrix_t > &input, bool applyDropout=false)
 Computes activation of the layer for the given input. More...
 
EActivationFunction GetActivationFunction () const
 
std::vector< Matrix_t > & GetDerivatives ()
 
const std::vector< Matrix_t > & GetDerivatives () const
 
Matrix_tGetDerivativesAt (size_t i)
 
const Matrix_tGetDerivativesAt (size_t i) const
 
Scalar_t GetDropoutProbability () const
 
size_t GetFilterDepth () const
 Getters. More...
 
size_t GetFilterHeight () const
 
size_t GetFilterWidth () const
 
std::vector< Matrix_t > & GetForwardMatrices ()
 
const std::vector< Matrix_t > & GetForwardMatrices () const
 
size_t GetNLocalViewPixels () const
 
size_t GetNLocalViews () const
 
size_t GetPaddingHeight () const
 
size_t GetPaddingWidth () const
 
ERegularization GetRegularization () const
 
size_t GetStrideCols () const
 
size_t GetStrideRows () const
 
Scalar_t GetWeightDecay () const
 
void Print () const
 Prints the info about the layer. More...
 
virtual void ReadWeightsFromXML (void *parent)
 Read the information and the weights about the layer from XML node. More...
 
- Public Member Functions inherited from TMVA::DNN::VGeneralLayer< Architecture_t >
 VGeneralLayer (const VGeneralLayer &)
 Copy Constructor. More...
 
 VGeneralLayer (size_t BatchSize, size_t InputDepth, size_t InputHeight, size_t InputWidth, size_t Depth, size_t Height, size_t Width, size_t WeightsNSlices, size_t WeightsNRows, size_t WeightsNCols, size_t BiasesNSlices, size_t BiasesNRows, size_t BiasesNCols, size_t OutputNSlices, size_t OutputNRows, size_t OutputNCols, EInitialization Init)
 Constructor. More...
 
 VGeneralLayer (size_t BatchSize, size_t InputDepth, size_t InputHeight, size_t InputWidth, size_t Depth, size_t Height, size_t Width, size_t WeightsNSlices, std::vector< size_t > WeightsNRows, std::vector< size_t > WeightsNCols, size_t BiasesNSlices, std::vector< size_t > BiasesNRows, std::vector< size_t > BiasesNCols, size_t OutputNSlices, size_t OutputNRows, size_t OutputNCols, EInitialization Init)
 General Constructor with different weights dimension. More...
 
 VGeneralLayer (VGeneralLayer< Architecture_t > *layer)
 Copy the layer provided as a pointer. More...
 
virtual ~VGeneralLayer ()
 Virtual Destructor. More...
 
virtual void AddWeightsXMLTo (void *parent)=0
 Writes the information and the weights about the layer in an XML node. More...
 
virtual void Backward (std::vector< Matrix_t > &gradients_backward, const std::vector< Matrix_t > &activations_backward, std::vector< Matrix_t > &inp1, std::vector< Matrix_t > &inp2)=0
 Backpropagates the error. More...
 
void CopyBiases (const std::vector< Matrix_t > &otherBiases)
 Copies the biases provided as an input. More...
 
void CopyWeights (const std::vector< Matrix_t > &otherWeights)
 Copies the weights provided as an input. More...
 
virtual void Forward (std::vector< Matrix_t > &input, bool applyDropout=false)=0
 Computes activation of the layer for the given input. More...
 
std::vector< Matrix_t > & GetActivationGradients ()
 
const std::vector< Matrix_t > & GetActivationGradients () const
 
Matrix_tGetActivationGradientsAt (size_t i)
 
const Matrix_tGetActivationGradientsAt (size_t i) const
 
size_t GetBatchSize () const
 Getters. More...
 
std::vector< Matrix_t > & GetBiases ()
 
const std::vector< Matrix_t > & GetBiases () const
 
Matrix_tGetBiasesAt (size_t i)
 
const Matrix_tGetBiasesAt (size_t i) const
 
std::vector< Matrix_t > & GetBiasGradients ()
 
const std::vector< Matrix_t > & GetBiasGradients () const
 
Matrix_tGetBiasGradientsAt (size_t i)
 
const Matrix_tGetBiasGradientsAt (size_t i) const
 
size_t GetDepth () const
 
size_t GetHeight () const
 
EInitialization GetInitialization () const
 
size_t GetInputDepth () const
 
size_t GetInputHeight () const
 
size_t GetInputWidth () const
 
std::vector< Matrix_t > & GetOutput ()
 
const std::vector< Matrix_t > & GetOutput () const
 
Matrix_tGetOutputAt (size_t i)
 
const Matrix_tGetOutputAt (size_t i) const
 
std::vector< Matrix_t > & GetWeightGradients ()
 
const std::vector< Matrix_t > & GetWeightGradients () const
 
Matrix_tGetWeightGradientsAt (size_t i)
 
const Matrix_tGetWeightGradientsAt (size_t i) const
 
std::vector< Matrix_t > & GetWeights ()
 
const std::vector< Matrix_t > & GetWeights () const
 
Matrix_tGetWeightsAt (size_t i)
 
const Matrix_tGetWeightsAt (size_t i) const
 
size_t GetWidth () const
 
void Initialize ()
 Initialize the weights and biases according to the given initialization method. More...
 
bool IsTraining () const
 
virtual void Print () const =0
 Prints the info about the layer. More...
 
void ReadMatrixXML (void *node, const char *name, Matrix_t &matrix)
 
virtual void ReadWeightsFromXML (void *parent)=0
 Read the information and the weights about the layer from XML node. More...
 
void SetBatchSize (size_t batchSize)
 Setters. More...
 
void SetDepth (size_t depth)
 
void SetHeight (size_t height)
 
void SetInputDepth (size_t inputDepth)
 
void SetInputHeight (size_t inputHeight)
 
void SetInputWidth (size_t inputWidth)
 
void SetIsTraining (bool isTraining)
 
void SetWidth (size_t width)
 
void Update (const Scalar_t learningRate)
 Updates the weights and biases, given the learning rate. More...
 
void UpdateBiases (const std::vector< Matrix_t > &biasGradients, const Scalar_t learningRate)
 Updates the biases, given the gradients and the learning rate. More...
 
void UpdateBiasGradients (const std::vector< Matrix_t > &biasGradients, const Scalar_t learningRate)
 Updates the bias gradients, given some other weight gradients and learning rate. More...
 
void UpdateWeightGradients (const std::vector< Matrix_t > &weightGradients, const Scalar_t learningRate)
 Updates the weight gradients, given some other weight gradients and learning rate. More...
 
void UpdateWeights (const std::vector< Matrix_t > &weightGradients, const Scalar_t learningRate)
 Updates the weights, given the gradients and the learning rate,. More...
 
void WriteMatrixToXML (void *node, const char *name, const Matrix_t &matrix)
 
void WriteTensorToXML (void *node, const char *name, const std::vector< Matrix_t > &tensor)
 helper functions for XML More...
 

Private Attributes

std::vector< Matrix_tindexMatrix
 Matrix of indices for the backward pass. More...
 

Additional Inherited Members

- Protected Attributes inherited from TMVA::DNN::CNN::TConvLayer< Architecture_t >
Scalar_t fDropoutProbability
 Probability that an input is active. More...
 
size_t fFilterDepth
 The depth of the filter. More...
 
size_t fFilterHeight
 The height of the filter. More...
 
size_t fFilterWidth
 The width of the filter. More...
 
size_t fNLocalViewPixels
 The number of pixels in one local image view. More...
 
size_t fNLocalViews
 The number of local views in one image. More...
 
size_t fStrideCols
 The number of column pixels to slid the filter each step. More...
 
size_t fStrideRows
 The number of row pixels to slid the filter each step. More...
 
- Protected Attributes inherited from TMVA::DNN::VGeneralLayer< Architecture_t >
std::vector< Matrix_tfActivationGradients
 Gradients w.r.t. the activations of this layer. More...
 
size_t fBatchSize
 Batch size used for training and evaluation. More...
 
std::vector< Matrix_tfBiases
 The biases associated to the layer. More...
 
std::vector< Matrix_tfBiasGradients
 Gradients w.r.t. the bias values of the layer. More...
 
size_t fDepth
 The depth of the layer. More...
 
size_t fHeight
 The height of the layer. More...
 
EInitialization fInit
 The initialization method. More...
 
size_t fInputDepth
 The depth of the previous layer or input. More...
 
size_t fInputHeight
 The height of the previous layer or input. More...
 
size_t fInputWidth
 The width of the previous layer or input. More...
 
bool fIsTraining
 Flag indicatig the mode. More...
 
std::vector< Matrix_tfOutput
 Activations of this layer. More...
 
std::vector< Matrix_tfWeightGradients
 Gradients w.r.t. the weights of the layer. More...
 
std::vector< Matrix_tfWeights
 The weights associated to the layer. More...
 
size_t fWidth
 The width of this layer. More...
 

#include <TMVA/DNN/CNN/MaxPoolLayer.h>

Inheritance diagram for TMVA::DNN::CNN::TMaxPoolLayer< Architecture_t >:
[legend]

Member Typedef Documentation

◆ Matrix_t

template<typename Architecture_t >
using TMVA::DNN::CNN::TMaxPoolLayer< Architecture_t >::Matrix_t = typename Architecture_t::Matrix_t

Definition at line 60 of file MaxPoolLayer.h.

◆ Scalar_t

template<typename Architecture_t >
using TMVA::DNN::CNN::TMaxPoolLayer< Architecture_t >::Scalar_t = typename Architecture_t::Scalar_t

Definition at line 61 of file MaxPoolLayer.h.

Constructor & Destructor Documentation

◆ TMaxPoolLayer() [1/3]

template<typename Architecture_t >
TMVA::DNN::CNN::TMaxPoolLayer< Architecture_t >::TMaxPoolLayer ( size_t  BatchSize,
size_t  InputDepth,
size_t  InputHeight,
size_t  InputWidth,
size_t  FilterHeight,
size_t  FilterWidth,
size_t  StrideRows,
size_t  StrideCols,
Scalar_t  DropoutProbability 
)

Constructor.

Definition at line 110 of file MaxPoolLayer.h.

◆ TMaxPoolLayer() [2/3]

template<typename Architecture_t >
TMVA::DNN::CNN::TMaxPoolLayer< Architecture_t >::TMaxPoolLayer ( TMaxPoolLayer< Architecture_t > *  layer)

Copy the max pooling layer provided as a pointer.

Definition at line 126 of file MaxPoolLayer.h.

◆ TMaxPoolLayer() [3/3]

template<typename Architecture_t >
TMVA::DNN::CNN::TMaxPoolLayer< Architecture_t >::TMaxPoolLayer ( const TMaxPoolLayer< Architecture_t > &  layer)

Copy constructor.

Definition at line 136 of file MaxPoolLayer.h.

◆ ~TMaxPoolLayer()

template<typename Architecture_t >
TMVA::DNN::CNN::TMaxPoolLayer< Architecture_t >::~TMaxPoolLayer

Destructor.

Definition at line 146 of file MaxPoolLayer.h.

Member Function Documentation

◆ AddWeightsXMLTo()

template<typename Architecture_t >
void TMVA::DNN::CNN::TMaxPoolLayer< Architecture_t >::AddWeightsXMLTo ( void parent)
virtual

Writes the information and the weights about the layer in an XML node.

Reimplemented from TMVA::DNN::CNN::TConvLayer< Architecture_t >.

Definition at line 202 of file MaxPoolLayer.h.

◆ Backward()

template<typename Architecture_t >
auto TMVA::DNN::CNN::TMaxPoolLayer< Architecture_t >::Backward ( std::vector< Matrix_t > &  gradients_backward,
const std::vector< Matrix_t > &  activations_backward,
std::vector< Matrix_t > &  inp1,
std::vector< Matrix_t > &  inp2 
)
virtual

Depending on the winning units determined during the Forward pass, it only forwards the derivatives to the right units in the previous layer.

Must only be called directly at the corresponding call to Forward(...).

Reimplemented from TMVA::DNN::CNN::TConvLayer< Architecture_t >.

Definition at line 168 of file MaxPoolLayer.h.

◆ Forward()

template<typename Architecture_t >
auto TMVA::DNN::CNN::TMaxPoolLayer< Architecture_t >::Forward ( std::vector< Matrix_t > &  input,
bool  applyDropout = false 
)
virtual

Computes activation of the layer for the given input.

The input must be in 3D tensor form with the different matrices corresponding to different events in the batch. It spatially downsamples the input matrices.

Reimplemented from TMVA::DNN::CNN::TConvLayer< Architecture_t >.

Definition at line 152 of file MaxPoolLayer.h.

◆ GetIndexMatrix() [1/2]

template<typename Architecture_t >
std::vector< Matrix_t > & TMVA::DNN::CNN::TMaxPoolLayer< Architecture_t >::GetIndexMatrix ( )
inline

Definition at line 104 of file MaxPoolLayer.h.

◆ GetIndexMatrix() [2/2]

template<typename Architecture_t >
const std::vector< Matrix_t > & TMVA::DNN::CNN::TMaxPoolLayer< Architecture_t >::GetIndexMatrix ( ) const
inline

Getters.

Definition at line 103 of file MaxPoolLayer.h.

◆ Print()

template<typename Architecture_t >
auto TMVA::DNN::CNN::TMaxPoolLayer< Architecture_t >::Print
virtual

Prints the info about the layer.

Reimplemented from TMVA::DNN::CNN::TConvLayer< Architecture_t >.

Definition at line 184 of file MaxPoolLayer.h.

◆ ReadWeightsFromXML()

template<typename Architecture_t >
void TMVA::DNN::CNN::TMaxPoolLayer< Architecture_t >::ReadWeightsFromXML ( void parent)
virtual

Read the information and the weights about the layer from XML node.

Reimplemented from TMVA::DNN::CNN::TConvLayer< Architecture_t >.

Definition at line 216 of file MaxPoolLayer.h.

Member Data Documentation

◆ indexMatrix

template<typename Architecture_t >
std::vector<Matrix_t> TMVA::DNN::CNN::TMaxPoolLayer< Architecture_t >::indexMatrix
private

Matrix of indices for the backward pass.

Definition at line 64 of file MaxPoolLayer.h.

Libraries for TMVA::DNN::CNN::TMaxPoolLayer< Architecture_t >:
[legend]

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