IgANet
IgANets - Isogeometric Analysis Networks
Loading...
Searching...
No Matches
iganet::LeakyReLU Class Reference

Leaky ReLU activation function. More...

#include </home/runner/work/iganet/iganet/include/layer.hpp>

Inheritance diagram for iganet::LeakyReLU:
iganet::ActivationFunction iganet::utils::FullQualifiedName

Public Member Functions

 LeakyReLU (double negative_slope, bool inplace=false)
 
 LeakyReLU (torch::nn::functional::LeakyReLUFuncOptions options={})
 
 ~LeakyReLU () override=default
 
torch::Tensor apply (const torch::Tensor &input) const override
 Applies the activation function to the given input.
 
torch::nn::functional::LeakyReLUFuncOptions & options ()
 Returns non-constant reference to options.
 
const torch::nn::functional::LeakyReLUFuncOptions & options () const
 Returns constant reference to options.
 
virtual void pretty_print (std::ostream &os=Log(log::info)) const noexcept override
 Returns a string representation of the activation function.
 
torch::serialize::InputArchive & read (torch::serialize::InputArchive &archive, const std::string &key="leaky_relu") override
 Reads the activation function from a torch::serialize::InputArchive object.
 
torch::serialize::OutputArchive & write (torch::serialize::OutputArchive &archive, const std::string &key="leaky_relu") const override
 Writes the activation function into a torch::serialize::OutputArchive object.
 
- Public Member Functions inherited from iganet::ActivationFunction
virtual ~ActivationFunction ()=default
 

Private Attributes

torch::nn::functional::LeakyReLUFuncOptions options_
 

Additional Inherited Members

- Protected Member Functions inherited from iganet::utils::FullQualifiedName
virtual const std::string & name () const noexcept
 Returns the full qualified name of the object.
 
- Protected Attributes inherited from iganet::utils::FullQualifiedName
at::optional< std::string > name_
 String storing the full qualified name of the object.
 

Detailed Description

Leaky ReLU activation function.

\[ \text{LeakyReLU}(x) = \begin{cases} x & \text{ if } x \ge 0 \text{negative_slope} * x & \text{ otherwise } \end{cases} \]

Constructor & Destructor Documentation

◆ LeakyReLU() [1/2]

iganet::LeakyReLU::LeakyReLU ( torch::nn::functional::LeakyReLUFuncOptions  options = {})
inlineexplicit

◆ LeakyReLU() [2/2]

iganet::LeakyReLU::LeakyReLU ( double  negative_slope,
bool  inplace = false 
)
inlineexplicit

◆ ~LeakyReLU()

iganet::LeakyReLU::~LeakyReLU ( )
overridedefault

Member Function Documentation

◆ apply()

torch::Tensor iganet::LeakyReLU::apply ( const torch::Tensor &  input) const
inlineoverridevirtual

Applies the activation function to the given input.

Implements iganet::ActivationFunction.

◆ options() [1/2]

torch::nn::functional::LeakyReLUFuncOptions & iganet::LeakyReLU::options ( )
inline

Returns non-constant reference to options.

◆ options() [2/2]

const torch::nn::functional::LeakyReLUFuncOptions & iganet::LeakyReLU::options ( ) const
inline

Returns constant reference to options.

◆ pretty_print()

virtual void iganet::LeakyReLU::pretty_print ( std::ostream &  os = Log(log::info)) const
inlineoverridevirtualnoexcept

Returns a string representation of the activation function.

Implements iganet::ActivationFunction.

◆ read()

torch::serialize::InputArchive & iganet::LeakyReLU::read ( torch::serialize::InputArchive &  archive,
const std::string &  key = "leaky_relu" 
)
inlineoverridevirtual

Reads the activation function from a torch::serialize::InputArchive object.

Implements iganet::ActivationFunction.

◆ write()

torch::serialize::OutputArchive & iganet::LeakyReLU::write ( torch::serialize::OutputArchive &  archive,
const std::string &  key = "leaky_relu" 
) const
inlineoverridevirtual

Writes the activation function into a torch::serialize::OutputArchive object.

Implements iganet::ActivationFunction.

Member Data Documentation

◆ options_

torch::nn::functional::LeakyReLUFuncOptions iganet::LeakyReLU::options_
private

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