Inheritance diagram for ngraph::runtime::HostTensor:

Collaboration diagram for ngraph::runtime::HostTensor:

Public Member Functions
	HostTensor (const element::Type &element_type, const Shape &shape, void *memory_pointer, const std::string &name="")

	HostTensor (const element::Type &element_type, const Shape &shape, const std::string &name="")

	HostTensor (const element::Type &element_type, const PartialShape &partial_shape, const std::string &name="")

	HostTensor (const std::string &name="")

	HostTensor (const Output< Node > &)

	HostTensor (const std::shared_ptr< op::v0::Constant > &constant)

void	initialize (const std::shared_ptr< op::v0::Constant > &constant)

void *	get_data_ptr ()

const void *	get_data_ptr () const

template<typename T >
T *	get_data_ptr ()

template<typename T >
const T *	get_data_ptr () const

template<element::Type_t ET>
element_type_traits< ET >::value_type *	get_data_ptr ()

template<element::Type_t ET>
const element_type_traits< ET >::value_type *	get_data_ptr () const

void	write (const void *p, size_t n) override
	Write bytes directly into the tensor. More...

void	read (void *p, size_t n) const override
	Read bytes directly from the tensor. More...

bool	get_is_allocated () const

void	set_element_type (const element::Type &element_type)
	Set the element type. Must be compatible with the current element type. More...

void	set_shape (const Shape &shape)
	Set the actual shape of the tensor compatibly with the partial shape. More...

void	set_unary (const HostTensorPtr &arg)
	Set the shape of a node from an input. More...

void	set_broadcast (const op::AutoBroadcastSpec &autob, const HostTensorPtr &arg0, const HostTensorPtr &arg1)
	Set the shape of the tensor using broadcast rules. More...

void	set_broadcast (const op::AutoBroadcastSpec &autob, const HostTensorPtr &arg0, const HostTensorPtr &arg1, const element::Type &element_type)
	Set the shape of the tensor using broadcast rules. More...

Public Member Functions inherited from ngraph::runtime::Tensor
Tensor &	operator= (const Tensor &)=default

virtual const ngraph::Shape &	get_shape () const
	Get tensor shape. More...

const ngraph::PartialShape &	get_partial_shape () const
	Get tensor partial shape. More...

virtual const element::Type &	get_element_type () const
	Get tensor element type. More...

virtual size_t	get_element_count () const
	Get number of elements in the tensor. More...

virtual size_t	get_size_in_bytes () const
	Get the size in bytes of the tensor. More...

const std::string &	get_name () const
	Get tensor's unique name. More...

bool	get_stale () const
	Get the stale value of the tensor. A tensor is stale if its data is changed. More...

void	set_stale (bool val)
	Set the stale value of the tensor. A tensor is stale if its data is changed.

virtual void	wait_for_read_ready ()
	check tensor for new data, call may block. backends may use this to ensure tensor is updated (eg: lazy eval).

virtual void	wait_for_write_ready ()
	notify tensor of new data, call may block. backends may use this as indication of new data in tensor.

Additional Inherited Members
Protected Member Functions inherited from ngraph::runtime::Tensor
	Tensor (const std::shared_ptr< ngraph::descriptor::Tensor > &descriptor)

Protected Attributes inherited from ngraph::runtime::Tensor
std::shared_ptr< ngraph::descriptor::Tensor >	m_descriptor

bool	m_stale

Member Function Documentation

◆ read()

void ngraph::runtime::HostTensor::read	(	void *	p,
		size_t	n
	)		const

overridevirtual

Read bytes directly from the tensor.

Parameters

p	Pointer to destination for data
n	Number of bytes to read, must be integral number of elements.

Implements ngraph::runtime::Tensor.

◆ set_broadcast() [1/2]

void ngraph::runtime::HostTensor::set_broadcast	(	const op::AutoBroadcastSpec &	autob,
		const HostTensorPtr &	arg0,
		const HostTensorPtr &	arg1
	)

Set the shape of the tensor using broadcast rules.

Parameters

autob	The broadcast mode
arg0	The first argument
arg1	The second argument

◆ set_broadcast() [2/2]

void ngraph::runtime::HostTensor::set_broadcast	(	const op::AutoBroadcastSpec &	autob,
		const HostTensorPtr &	arg0,
		const HostTensorPtr &	arg1,
		const element::Type &	element_type
	)

Set the shape of the tensor using broadcast rules.

Parameters

autob	The broadcast mode
arg0	The first argument
arg1	The second argument
element_type	The output element type

◆ set_element_type()

void ngraph::runtime::HostTensor::set_element_type ( const element::Type & element_type )

Set the element type. Must be compatible with the current element type.

Parameters

element_type The element type

◆ set_shape()

void ngraph::runtime::HostTensor::set_shape ( const Shape & shape )

Set the actual shape of the tensor compatibly with the partial shape.

Parameters

shape The shape being set

◆ set_unary()

void ngraph::runtime::HostTensor::set_unary ( const HostTensorPtr & arg )

Set the shape of a node from an input.

Parameters

arg	The input argument

◆ write()

void ngraph::runtime::HostTensor::write	(	const void *	p,
		size_t	n
	)

overridevirtual

Write bytes directly into the tensor.

Parameters

p	Pointer to source of data
n	Number of bytes to write, must be integral number of elements.

Implements ngraph::runtime::Tensor.

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

core/include/ngraph/runtime/host_tensor.hpp

Public Member Functions

Additional Inherited Members

Member Function Documentation

◆ read()

◆ set_broadcast() [1/2]

◆ set_broadcast() [2/2]

◆ set_element_type()

◆ set_shape()

◆ set_unary()

◆ write()