CPU implementation of the Layer Normalization operation for neural networks. More...

Inheritance diagram for Mila::Dnn::Compute::CpuLayerNormOp:

[legend]

Collaboration diagram for Mila::Dnn::Compute::CpuLayerNormOp:

[legend]

Public Types
using	MR = typename CpuDevice::MR

using	OperationBase = UnaryOperation< DeviceType::Cpu, float >

Public Types inherited from Mila::Dnn::Compute::UnaryOperation< DeviceType::Cpu, float >
using	MR = std::conditional_t< TDeviceType==DeviceType::Cuda, CudaMemoryResource, HostMemoryResource >
	Memory resource type based on device type.

Public Member Functions
	CpuLayerNormOp (const LayerNormConfig &config)
	Constructs a new CPU Layer Normalization operation with the default device context.

	CpuLayerNormOp (std::shared_ptr< DeviceContext > context, const LayerNormConfig &config)
	Constructs a new CPU Layer Normalization operation with a specific device context.

void	backward (float dinp, float dweight, float dbias, float dout, float inp, float weight, float mean, float rstd, int B, int T, int C)
	Performs the backward pass of the Layer Normalization operation.

void	forward (const Tensor< float, MR > &input, const std::vector< std::shared_ptr< Tensor< float, MR > > > &parameters, const OperationAttributes &attributes, Tensor< float, MR > &output, std::vector< std::shared_ptr< Tensor< float, MR > > > &output_state) const override
	Performs the forward pass of the Layer Normalization operation.

std::string	getName () const override
	Gets the name of this operation.

Public Member Functions inherited from Mila::Dnn::Compute::UnaryOperation< DeviceType::Cpu, float >
	UnaryOperation (OperationType operation_type)
	Constructs a UnaryOperation with the specified operation type.

	UnaryOperation (OperationType operation_type, std::shared_ptr< DeviceContext > context)
	Constructs a UnaryOperation with the specified operation type and device context.

virtual	~UnaryOperation ()=default
	Virtual destructor for proper cleanup of derived classes.

virtual void	backward (const Tensor< float, MR > &grad, const std::vector< std::shared_ptr< Tensor< float, MR > > > &parameters, std::vector< std::shared_ptr< Tensor< float, MR > > > &output_grads) const
	Executes the backward pass of a unary operation.

virtual void	backward (const Tensor< float, MR > &input, const Tensor< float, MR > &output_grad, const std::vector< std::shared_ptr< Tensor< float, MR > > > &parameters, std::vector< std::shared_ptr< Tensor< float, MR > > > &parameter_grads, Tensor< float, MR > &input_grad, const OperationAttributes &properties, const std::vector< std::shared_ptr< Tensor< float, MR > > > &output_state) const
	Executes the comprehensive backward pass of a unary operation.

virtual void	forward (const Tensor< float, MR > &input, const std::vector< std::shared_ptr< Tensor< float, MR > > > &parameters, const OperationAttributes &properties, Tensor< float, MR > &output, std::vector< std::shared_ptr< Tensor< float, MR > > > &output_state) const=0
	Executes the forward pass of a unary operation.

Public Member Functions inherited from Mila::Dnn::Compute::OperationBase< TDeviceType, TInput1, TInput2, TOutput >
	OperationBase (OperationType operation_type, std::shared_ptr< DeviceContext > context)
	Constructs an OperationBase object with a specific device context and compute precision.

virtual	~OperationBase ()=default
	Virtual destructor for the OperationBase class.

std::shared_ptr< DeviceContext >	getDeviceContext () const
	Gets the device context associated with this operation.

DeviceType	getDeviceType () const
	Gets the device type for this operation.

OperationType	getOperationType () const
	Gets the operation type enumeration value.

Private Attributes
LayerNormConfig	config_
	Configuration for the LayerNorm operation.

Detailed Description

CPU implementation of the Layer Normalization operation for neural networks.

This class provides a CPU-based implementation of the Layer Normalization operation, which normalizes inputs across the features dimension for each sample in a batch. Layer normalization helps stabilize training by reducing internal covariate shift and is commonly used in transformer architectures and other deep neural networks.

The operation normalizes each input vector independently, unlike batch normalization which normalizes across the batch dimension.

CPU operations always use full precision regardless of policy settings.

Template Parameters

TInput	The data type of the input tensor elements.
TDataType	The data type used for computation and output (defaults to the input type).

Member Typedef Documentation

◆ MR

using Mila::Dnn::Compute::CpuLayerNormOp::MR = typename CpuDevice::MR

◆ OperationBase

using Mila::Dnn::Compute::CpuLayerNormOp::OperationBase = UnaryOperation<DeviceType::Cpu, float>

Constructor & Destructor Documentation

◆ CpuLayerNormOp() [1/2]

Mila::Dnn::Compute::CpuLayerNormOp::CpuLayerNormOp ( const LayerNormConfig & config )

inline

Constructs a new CPU Layer Normalization operation with the default device context.

CPU operations always use full precision regardless of policy settings.

Parameters

precision_policy Ignored for CPU operations, as they always use full precision.

◆ CpuLayerNormOp() [2/2]

Mila::Dnn::Compute::CpuLayerNormOp::CpuLayerNormOp	(	std::shared_ptr< DeviceContext >	context,
		const LayerNormConfig &	config
	)

inline

Constructs a new CPU Layer Normalization operation with a specific device context.

CPU operations always use full precision regardless of policy settings.

Parameters

context	The device context to use for this operation.
precision_policy	Ignored for CPU operations, as they always use full precision.

Exceptions

std::runtime_error If the context is not for a CPU device.

Member Function Documentation

◆ backward()

void Mila::Dnn::Compute::CpuLayerNormOp::backward	(	float *	dinp,
		float *	dweight,
		float *	dbias,
		float *	dout,
		float *	inp,
		float *	weight,
		float *	mean,
		float *	rstd,
		int	B,
		int	T,
		int	C
	)

inline

Performs the backward pass of the Layer Normalization operation.

Computes gradients with respect to inputs, weights, and biases based on the output gradient and the forward pass results.

Parameters

dinp	Pointer to the gradient buffer for input.
dweight	Pointer to the gradient buffer for weight parameters.
dbias	Pointer to the gradient buffer for bias parameters.
dout	Pointer to the gradient buffer from the output.
inp	Pointer to the original input values.
weight	Pointer to the weight parameters.
mean	Pointer to the mean values computed during forward pass.
rstd	Pointer to the reciprocal standard deviation values computed during forward pass.
B	Batch size.
TDataType	Sequence length.
C	Number of features/channels.

Here is the call graph for this function:

◆ forward()

void Mila::Dnn::Compute::CpuLayerNormOp::forward	(	const Tensor< float, MR > &	input,
		const std::vector< std::shared_ptr< Tensor< float, MR > > > &	parameters,
		const OperationAttributes &	attributes,
		Tensor< float, MR > &	output,
		std::vector< std::shared_ptr< Tensor< float, MR > > > &	output_state
	)		const

inlineoverride

Performs the forward pass of the Layer Normalization operation.

Normalizes each input vector across the feature dimension, then applies a learnable scaling factor and bias.

Parameters

input	Input tensor of shape [B, TDataType, C] where B is batch size, TDataType is sequence length, and C is feature dimension.
parameters	Vector of parameter tensors [weight, bias] where weight and bias are of shape [C].
attributes	Additional attributes for the operation.
output	Output tensor of the same shape as input, containing the normalized values.
output_state	Cache for intermediate results [mean, rstd] used in the backward pass.

Here is the call graph for this function:

◆ getName()

std::string Mila::Dnn::Compute::CpuLayerNormOp::getName ( ) const

inlineoverridevirtual

Gets the name of this operation.

Returns: std::string The name of the operation ("Cpu::LayerNormOp").

Implements Mila::Dnn::Compute::OperationBase< TDeviceType, TInput1, TInput2, TOutput >.

Member Data Documentation

◆ config_

LayerNormConfig Mila::Dnn::Compute::CpuLayerNormOp::config_

private

Configuration for the LayerNorm operation.

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

/home/runner/work/Mila/Mila/Mila/Src/Dnn/Compute/Operations/Cpu/CpuLayerNormOp.ixx

Public Types

Public Member Functions

Private Attributes

Detailed Description

Member Typedef Documentation

◆ MR

◆ OperationBase

Constructor & Destructor Documentation

◆ CpuLayerNormOp() [1/2]

◆ CpuLayerNormOp() [2/2]

Member Function Documentation

◆ backward()

◆ forward()

◆ getName()

Member Data Documentation

◆ config_