mlpack
Public Member Functions | List of all members
mlpack::ann::HardTanH< InputDataType, OutputDataType > Class Template Reference

The Hard Tanh activation function, defined by. More...

#include <hard_tanh.hpp>

Public Member Functions

 HardTanH (const double maxValue=1, const double minValue=-1)
 Create the HardTanH object using the specified parameters. More...
 
template<typename InputType , typename OutputType >
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...
 
template<typename DataType >
void Backward (const DataType &input, const DataType &gy, DataType &g)
 Ordinary feed backward pass of a neural network, calculating the function f(x) by propagating x backwards through f. More...
 
OutputDataType const & OutputParameter () const
 Get the output parameter.
 
OutputDataType & OutputParameter ()
 Modify the output parameter.
 
OutputDataType const & Delta () const
 Get the delta.
 
OutputDataType & Delta ()
 Modify the delta.
 
double const & MaxValue () const
 Get the maximum value.
 
double & MaxValue ()
 Modify the maximum value.
 
double const & MinValue () const
 Get the minimum value.
 
double & MinValue ()
 Modify the minimum value.
 
template<typename Archive >
void serialize (Archive &ar, const uint32_t)
 Serialize the layer.
 

Detailed Description

template<typename InputDataType = arma::mat, typename OutputDataType = arma::mat>
class mlpack::ann::HardTanH< InputDataType, OutputDataType >

The Hard Tanh activation function, defined by.

\begin{eqnarray*} f(x) &=& \left\{ \begin{array}{lr} max & : x > maxValue \\ min & : x \le minValue \\ x & : otherwise \end{array} \right. \\ f'(x) &=& \left\{ \begin{array}{lr} 0 & : x > maxValue \\ 0 & : x \le minValue \\ 1 & : otherwise \end{array} \right. \end{eqnarray*}

Template Parameters
InputDataTypeType of the input data (arma::colvec, arma::mat, arma::sp_mat or arma::cube).
OutputDataTypeType of the output data (arma::colvec, arma::mat, arma::sp_mat or arma::cube).

Constructor & Destructor Documentation

◆ HardTanH()

template<typename InputDataType , typename OutputDataType >
mlpack::ann::HardTanH< InputDataType, OutputDataType >::HardTanH ( const double  maxValue = 1,
const double  minValue = -1 
)

Create the HardTanH object using the specified parameters.

The range of the linear region can be adjusted by specifying the maxValue and minValue. Default (maxValue = 1, minValue = -1).

Parameters
maxValueRange of the linear region maximum value.
minValueRange of the linear region minimum value.

Member Function Documentation

◆ Backward()

template<typename InputDataType , typename OutputDataType >
template<typename DataType >
void mlpack::ann::HardTanH< InputDataType, OutputDataType >::Backward ( const DataType &  input,
const DataType &  gy,
DataType &  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
inputThe propagated input activation.
gyThe backpropagated error.
gThe calculated gradient.

◆ Forward()

template<typename InputDataType , typename OutputDataType >
template<typename InputType , typename OutputType >
void mlpack::ann::HardTanH< InputDataType, OutputDataType >::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
inputInput data used for evaluating the specified function.
outputResulting output activation.

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