mlpack
Public Member Functions | List of all members
GlorotInitializationType< Uniform > Class Template Reference

This class is used to initialize the weight matrix with the Glorot Initialization method. More...

Public Member Functions

 GlorotInitializationType ()
 Initialize the Glorot initialization object. More...

 
template
<
typename
eT
>
void Initialize (arma::Mat< eT > &W, const size_t rows, const size_t cols)
 Initialize the elements weight matrix with glorot initialization method. More...

 
template
<
typename
eT
>
void Initialize (arma::Mat< eT > &W)
 Initialize the elements weight matrix with glorot initialization method. More...

 
template
<
typename
eT
>
void Initialize (arma::Cube< eT > &W, const size_t rows, const size_t cols, const size_t slices)
 Initialize the elements of the specified weight 3rd order tensor with glorot initialization method. More...

 
template
<
typename
eT
>
void Initialize (arma::Cube< eT > &W)
 Initialize the elements of the specified weight 3rd order tensor with glorot initialization method. More...

 
template
<
>
void Initialize (arma::Mat< eT > &W, const size_t rows, const size_t cols)
 
template
<
>
void Initialize (arma::Mat< eT > &W)
 
template
<
>
void Initialize (arma::Mat< eT > &W, const size_t rows, const size_t cols)
 
template
<
>
void Initialize (arma::Mat< eT > &W)
 
template
<
typename
Archive
>
void serialize (Archive &, const uint32_t)
 Serialize the initialization. More...

 

Detailed Description


template<bool Uniform = true>
class mlpack::ann::GlorotInitializationType< Uniform >

This class is used to initialize the weight matrix with the Glorot Initialization method.

The method is defined by

\begin{eqnarray*} \mathrm{Var}[w_i] &=& \frac{2}{n_i + n_{i+1}} \\ w_i \sim \mathrm{U}[-\frac{\sqrt{6}}{\sqrt{n_i + n_{i+1}}}, \frac{\sqrt{6}}{\sqrt{n_i + n_{i+1}}}] \end{eqnarray*}

where $ n_{i+1} $ is the number of neurons in the outgoing layer, $ n_i $ represents the number of neurons in the ingoing layer. Here Normal Distribution may also be used if needed

For more information, see the following paper.

@inproceedings {pmlr-v9-glorot10a,
title = {Understanding the difficulty of training
deep feedforward neural networks},
author = {Xavier Glorot and Yoshua Bengio},
booktitle = {Proceedings of the Thirteenth International Conference
on Artificial Intelligence and Statistics},
year = {2010}
}

Definition at line 53 of file glorot_init.hpp.

Constructor & Destructor Documentation

◆ GlorotInitializationType()

GlorotInitializationType ( )
inline

Initialize the Glorot initialization object.

Definition at line 59 of file glorot_init.hpp.

References GlorotInitializationType< Uniform >::Initialize().

Member Function Documentation

◆ Initialize() [1/8]

void Initialize ( arma::Mat< eT > &  W,
const size_t  rows,
const size_t  cols 
)

Initialize the elements weight matrix with glorot initialization method.

Parameters
WWeight matrix to initialize.
rowsNumber of rows.
colsNumber of columns.

Referenced by GlorotInitializationType< Uniform >::GlorotInitializationType(), and GlorotInitializationType< Uniform >::Initialize().

◆ Initialize() [2/8]

void Initialize ( arma::Mat< eT > &  W)

Initialize the elements weight matrix with glorot initialization method.

Parameters
WWeight matrix to initialize.

◆ Initialize() [3/8]

void Initialize ( arma::Cube< eT > &  W,
const size_t  rows,
const size_t  cols,
const size_t  slices 
)
inline

Initialize the elements of the specified weight 3rd order tensor with glorot initialization method.

Parameters
WWeight matrix to initialize.
rowsNumber of rows.
colsNumber of columns.
slicesNumber of slices.

Definition at line 171 of file glorot_init.hpp.

References GlorotInitializationType< Uniform >::Initialize().

◆ Initialize() [4/8]

void Initialize ( arma::Cube< eT > &  W)
inline

Initialize the elements of the specified weight 3rd order tensor with glorot initialization method.

Parameters
WWeight matrix to initialize.

Definition at line 185 of file glorot_init.hpp.

References Log::Fatal, and GlorotInitializationType< Uniform >::Initialize().

◆ Initialize() [5/8]

void Initialize ( arma::Mat< eT > &  W,
const size_t  rows,
const size_t  cols 
)
inline

Definition at line 117 of file glorot_init.hpp.

References GaussianInitialization::Initialize().

◆ Initialize() [6/8]

void Initialize ( arma::Mat< eT > &  W)
inline

Definition at line 131 of file glorot_init.hpp.

References Log::Fatal, and GaussianInitialization::Initialize().

◆ Initialize() [7/8]

void Initialize ( arma::Mat< eT > &  W,
const size_t  rows,
const size_t  cols 
)
inline

Definition at line 143 of file glorot_init.hpp.

References RandomInitialization::Initialize().

◆ Initialize() [8/8]

void Initialize ( arma::Mat< eT > &  W)
inline

Definition at line 158 of file glorot_init.hpp.

References Log::Fatal, and RandomInitialization::Initialize().

◆ serialize()

void serialize ( Archive &  ,
const uint32_t   
)
inline

Serialize the initialization.

(Nothing to serialize for this one.)

Definition at line 112 of file glorot_init.hpp.

The documentation for this class was generated from the following file:
  • /home/jenkins-mlpack/mlpack.org/_src/mlpack-git/src/mlpack/methods/ann/init_rules/glorot_init.hpp