iganet::GumbelSoftmax Class Reference

Gumbel-Softmax distribution activation function. More...

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

Inheritance diagram for iganet::GumbelSoftmax:

Public Member Functions
	GumbelSoftmax (double tau, int dim, bool hard)
	GumbelSoftmax (torch::nn::functional::GumbelSoftmaxFuncOptions options={})
	~GumbelSoftmax () override=default
torch::Tensor	apply (const torch::Tensor &input) const override
	Applies the activation function to the given input.
torch::nn::functional::GumbelSoftmaxFuncOptions &	options ()
	Returns non-constant reference to options.
const torch::nn::functional::GumbelSoftmaxFuncOptions &	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="gumbel_softmax") override
	Reads the activation function from a torch::serialize::InputArchive object.
torch::serialize::OutputArchive &	write (torch::serialize::OutputArchive &archive, const std::string &key="gumbel_softmax") 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::GumbelSoftmaxFuncOptions	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

Gumbel-Softmax distribution activation function.

Constructor & Destructor Documentation

GumbelSoftmax() [1/2]

iganet::GumbelSoftmax::GumbelSoftmax ( torch::nn::functional::GumbelSoftmaxFuncOptions  options = {} )

inlineexplicit

GumbelSoftmax() [2/2]

iganet::GumbelSoftmax::GumbelSoftmax ( double  tau, int  dim, bool  hard  )

inlineexplicit

~GumbelSoftmax()

iganet::GumbelSoftmax::~GumbelSoftmax ( )

overridedefault

Member Function Documentation

apply()

torch::Tensor iganet::GumbelSoftmax::apply ( const torch::Tensor &  input ) const

inlineoverridevirtual

Applies the activation function to the given input.

Implements iganet::ActivationFunction.

options() [1/2]

torch::nn::functional::GumbelSoftmaxFuncOptions & iganet::GumbelSoftmax::options ( )

inline

Returns non-constant reference to options.

options() [2/2]

const torch::nn::functional::GumbelSoftmaxFuncOptions & iganet::GumbelSoftmax::options ( ) const

inline

Returns constant reference to options.

pretty_print()

virtual void iganet::GumbelSoftmax::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::GumbelSoftmax::read ( torch::serialize::InputArchive &  archive, const std::string &  key = "gumbel_softmax"  )

inlineoverridevirtual

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

Implements iganet::ActivationFunction.

write()

torch::serialize::OutputArchive & iganet::GumbelSoftmax::write ( torch::serialize::OutputArchive &  archive, const std::string &  key = "gumbel_softmax"  ) const

inlineoverridevirtual

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

Implements iganet::ActivationFunction.

Member Data Documentation

options_

torch::nn::functional::GumbelSoftmaxFuncOptions iganet::GumbelSoftmax::options_

private

---

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

• include/layer.hpp