This primitive executes a custom kernel provided by the application. More...

#include <custom_gpu_primitive.hpp>

Inheritance diagram for cldnn::custom_gpu_primitive:

Collaboration diagram for cldnn::custom_gpu_primitive:

[legend]

Public Types
typedef cldnn_custom_gpu_primitive_desc	dto

Public Member Functions
	custom_gpu_primitive (const primitive_id &id, const std::vector< primitive_id > &input, const std::vector< std::string > &kernels_code, const std::string &kernel_entry_point, const std::vector< cldnn_arg > &kernel_arguments, const std::string &build_options, const layout &output_layout, const std::vector< size_t > &gws={}, const std::vector< size_t > &lws={})
	Constructs custom_gpu_primitive primitive. More...

	custom_gpu_primitive (const dto *dto)
	Constructs a copy from basic C API cldnn_custom_gpu_primitive_desc.

Public Member Functions inherited from cldnn::primitive_base< custom_gpu_primitive, cldnn_custom_gpu_primitive_desc >
const cldnn_primitive_desc *	get_dto () const override
	Returns pointer to a C API primitive descriptor casted to cldnn_primitive_desc.

Public Member Functions inherited from cldnn::primitive
	primitive (const primitive_type_id &type, const primitive_id &id, const std::vector< primitive_id > &input, const padding &output_padding=padding())

	primitive (const cldnn_primitive_desc *dto)
	Constructs a copy from basic C API cldnn_primitive_desc.

std::vector< std::reference_wrapper< primitive_id > >	dependecies ()
	Returns references to all primitive ids on which this primitive depends - inputs, weights, biases, etc.

std::vector< primitive_id >	dependecies () const
	Returns copy of all primitive ids on which this primitive depends - inputs, weights, biases, etc.

	operator primitive_id () const
	Implicit conversion to primiitive id.

Static Public Member Functions
static primitive_type_id	type_id ()

Public Attributes
fixed_size_vector_ref	kernels_code
	Source code for the kernel.

const std::string	kernel_entry_point
	The name of the entry point function in the kernel.

const std::vector< cldnn_arg >	kernel_arguments
	Argument bindings for the entry point function.

const std::string	build_options
	The kernel's build options.

const layout	output_layout
	The output layout declared by the primitive.

const std::vector< size_t >	gws
	The global working sizes.

const std::vector< size_t >	lws
	The local working sizes.

Public Attributes inherited from cldnn::primitive
const primitive_type_id	type
	Primitive's type id.

const primitive_id	id
	Primitive's id.

fixed_size_vector_ref	input
	List of ids of input primitives.

padding	output_padding
	Requested output padding.

Protected Member Functions
void	update_dto (dto &dto) const override

Protected Member Functions inherited from cldnn::primitive_base< custom_gpu_primitive, cldnn_custom_gpu_primitive_desc >
	primitive_base (const primitive_id &id, const std::vector< primitive_id > &input, const padding &output_padding=padding())

	primitive_base (const cldnn_custom_gpu_primitive_desc *dto)

Protected Member Functions inherited from cldnn::primitive
virtual std::vector< std::reference_wrapper< const primitive_id > >	get_dependencies () const

Protected Attributes
primitive_id_arr	_kernels_code

Protected Attributes inherited from cldnn::primitive
primitive_id_arr	_input

Detailed Description

This primitive executes a custom kernel provided by the application.

The application is required to provide all relevant details for executing the custom kernel such as: sources, entry point, work sizes and parameter bindings.

Definition at line 35 of file custom_gpu_primitive.hpp.

Constructor & Destructor Documentation

◆ custom_gpu_primitive()

cldnn::custom_gpu_primitive::custom_gpu_primitive	(	const primitive_id &	id,
		const std::vector< primitive_id > &	input,
		const std::vector< std::string > &	kernels_code,
		const std::string &	kernel_entry_point,
		const std::vector< cldnn_arg > &	kernel_arguments,
		const std::string &	build_options,
		const layout &	output_layout,
		const std::vector< size_t > &	gws = `{}`,
		const std::vector< size_t > &	lws = `{}`
	)

inline

Constructs custom_gpu_primitive primitive.

Parameters

id	This primitive id.
input	Input primitive ids.
kernels_code	Source code for the kernel
kernel_entry_point	The name of the entry point function in the kernel
kernel_arguments	Argument bindings for the entry point function
build_options	Build options/flags used during the compilation of the custom kernel
output_layout	Output layout declared by the primitive
gws	Global work sizes
lws	Local work sizes

Definition at line 49 of file custom_gpu_primitive.hpp.

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

C:/ML/OpenSource/clDNN/api/CPP/custom_gpu_primitive.hpp

Public Types

Public Member Functions

Static Public Member Functions

Public Attributes

Protected Member Functions

Protected Attributes

Detailed Description

Constructor & Destructor Documentation

◆ custom_gpu_primitive()