Mila::Dnn::LayerNormConfig Class Reference

Inheritance diagram for Mila::Dnn::LayerNormConfig:

Collaboration diagram for Mila::Dnn::LayerNormConfig:

Public Member Functions
	LayerNormConfig (int64_t axis)
	Construct in axis mode.
	LayerNormConfig (shape_t shape)
	Construct in shape mode.
void	fromMetadata (const SerializationMetadata &meta) override
	Populate configuration from provided metadata.
std::optional< int64_t >	getAxis () const noexcept
float	getEpsilon () const noexcept
const shape_t &	getNormalizedShape () const noexcept
bool	hasBias () const noexcept
bool	hasNormalizedShape () const noexcept
SerializationMetadata	toMetadata () const override
	Convert configuration into a SerializationMetadata object.
std::string	toString () const override
	Produce a short, human-readable summary of the configuration.
void	validate () const override
	Validate configuration parameters.
template<typename Self>
decltype(auto)	withBias (this Self &&self, bool has_bias)
	Enable or disable learnable bias.
template<typename Self>
decltype(auto)	withEpsilon (this Self &&self, float epsilon)
	Set epsilon for numerical stability.
Public Member Functions inherited from Mila::Dnn::ComponentConfig
virtual	~ComponentConfig ()=default
	Virtual destructor for polymorphic base.

Private Attributes
std::optional< dim_t >	axis_ { std::nullopt }
float	epsilon_ { 1e-5f }
bool	has_bias_ { true }
shape_t	normalized_shape_ {}

Constructor & Destructor Documentation

LayerNormConfig() [1/2]

Mila::Dnn::LayerNormConfig::LayerNormConfig ( shape_t shape )

inlineexplicit

Construct in shape mode.

Normalizes over the trailing dimensions described by shape.

Parameters

shape

Trailing dimensions to normalize over (e.g. shape_t{ model_dim }).

LayerNormConfig() [2/2]

Mila::Dnn::LayerNormConfig::LayerNormConfig ( int64_t axis )

inlineexplicit

Construct in axis mode.

Normalizes over a single axis.

Parameters

axis	Axis along which to normalize (negative indexing supported).

Member Function Documentation

fromMetadata()

void Mila::Dnn::LayerNormConfig::fromMetadata ( const SerializationMetadata & meta )

inlineoverridevirtual

Populate configuration from provided metadata.

Implementations should read available keys and leave missing keys at their current/default values to preserve forward/backward compatibility.

Parameters

meta	Metadata to read configuration values from.

Implements Mila::Dnn::ComponentConfig.

Here is the call graph for this function:

getAxis()

std::optional< int64_t > Mila::Dnn::LayerNormConfig::getAxis ( ) const

inlinenoexcept

getEpsilon()

float Mila::Dnn::LayerNormConfig::getEpsilon ( ) const

inlinenoexcept

getNormalizedShape()

const shape_t & Mila::Dnn::LayerNormConfig::getNormalizedShape ( ) const

inlinenoexcept

hasBias()

bool Mila::Dnn::LayerNormConfig::hasBias ( ) const

inlinenoexcept

hasNormalizedShape()

bool Mila::Dnn::LayerNormConfig::hasNormalizedShape ( ) const

inlinenoexcept

toMetadata()

SerializationMetadata Mila::Dnn::LayerNormConfig::toMetadata ( ) const

inlineoverridevirtual

Convert configuration into a SerializationMetadata object.

Implementations should include any fields required to fully reconstruct the configuration via fromMetadata.

Returns

SerializationMetadata Metadata representation of the config.

Implements Mila::Dnn::ComponentConfig.

Here is the call graph for this function:

toString()

std::string Mila::Dnn::LayerNormConfig::toString ( ) const

inlineoverridevirtual

Produce a short, human-readable summary of the configuration.

Implementations should return a compact, single-line description suitable for logging and debugging.

Returns

std::string Human-readable summary of the configuration.

Implements Mila::Dnn::ComponentConfig.

validate()

void Mila::Dnn::LayerNormConfig::validate ( ) const

inlineoverridevirtual

Validate configuration parameters.

Called by callers to ensure the configuration represents a valid, constructible component. Implementations must throw std::invalid_argument (or a derived exception) when validation fails.

Exceptions

std::invalid_argument

If the configuration is invalid.

Implements Mila::Dnn::ComponentConfig.

withBias()

template<typename Self>

decltype(auto) Mila::Dnn::LayerNormConfig::withBias ( this Self && self, bool has_bias )

inline

Enable or disable learnable bias.

Default: true.

withEpsilon()

template<typename Self>

decltype(auto) Mila::Dnn::LayerNormConfig::withEpsilon ( this Self && self, float epsilon )

inline

Set epsilon for numerical stability.

Default: 1e-5f.

Member Data Documentation

axis_

std::optional<dim_t> Mila::Dnn::LayerNormConfig::axis_ { std::nullopt }

private

epsilon_

float Mila::Dnn::LayerNormConfig::epsilon_ { 1e-5f }

private

has_bias_

bool Mila::Dnn::LayerNormConfig::has_bias_ { true }

private

normalized_shape_

shape_t Mila::Dnn::LayerNormConfig::normalized_shape_ {}

private

---

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

• /__w/Mila/Mila/Mila/Src/Dnn/Components/Normalization/LayerNorm/LayerNorm.Config.ixx