#include <onnxruntime_cxx_api.h>

Inheritance diagram for Ort::Value:

Public Types
using	Base = detail::ValueImpl< OrtValue >

using	OrtSparseValuesParam = detail::OrtSparseValuesParam

using	Shape = detail::Shape

Public Types inherited from Ort::detail::ValueImpl< OrtValue >
using	B = ConstValueImpl< OrtValue >

using	B = Base< OrtValue >

Public Types inherited from Ort::detail::ConstValueImpl< T >
using	B = Base< T >

Public Types inherited from Ort::detail::Base< T >
using	contained_type = T

Public Member Functions
	Value (std::nullptr_t)
	Create an empty Value object, must be assigned a valid one to be used.

	Value (OrtValue *p)
	Used for interop with the C API.

	Value (Value &&)=default

Value &	operator= (Value &&)=default

ConstValue	GetConst () const

UnownedValue	GetUnowned () const

Public Member Functions inherited from Ort::detail::ValueImpl< OrtValue >
R *	GetTensorMutableData ()
	Returns a non-const typed pointer to an OrtValue/Tensor contained buffer No type checking is performed, the caller must ensure the type matches the tensor type.

void *	GetTensorMutableRawData ()
	Returns a non-typed non-const pointer to a tensor contained data.

R &	At (const std::vector< int64_t > &location)

void	FillStringTensor (const char const s, size_t s_len)
	Set all strings at once in a string tensor.

void	FillStringTensorElement (const char *s, size_t index)
	Set a single string in a string tensor.

char *	GetResizedStringTensorElementBuffer (size_t index, size_t buffer_length)
	Allocate if necessary and obtain a pointer to a UTF-8 encoded string element buffer indexed by the flat element index, of the specified length.

void	UseCooIndices (int64_t *indices_data, size_t indices_num)
	Supplies COO format specific indices and marks the contained sparse tensor as being a COO format tensor. Values are supplied with a CreateSparseTensor() API. The supplied indices are not copied and the user allocated buffers lifespan must eclipse that of the OrtValue. The location of the indices is assumed to be the same as specified by OrtMemoryInfo argument at the creation time.

void	UseCsrIndices (int64_t inner_data, size_t inner_num, int64_t outer_data, size_t outer_num)
	Supplies CSR format specific indices and marks the contained sparse tensor as being a CSR format tensor. Values are supplied with a CreateSparseTensor() API. The supplied indices are not copied and the user allocated buffers lifespan must eclipse that of the OrtValue. The location of the indices is assumed to be the same as specified by OrtMemoryInfo argument at the creation time.

void	UseBlockSparseIndices (const Shape &indices_shape, int32_t *indices_data)
	Supplies BlockSparse format specific indices and marks the contained sparse tensor as being a BlockSparse format tensor. Values are supplied with a CreateSparseTensor() API. The supplied indices are not copied and the user allocated buffers lifespan must eclipse that of the OrtValue. The location of the indices is assumed to be the same as specified by OrtMemoryInfo argument at the creation time.

void	FillSparseTensorCoo (const OrtMemoryInfo data_mem_info, const OrtSparseValuesParam &values_param, const int64_t indices_data, size_t indices_num)
	The API will allocate memory using the allocator instance supplied to the CreateSparseTensor() API and copy the values and COO indices into it. If data_mem_info specifies that the data is located at difference device than the allocator, a X-device copy will be performed if possible.

void	FillSparseTensorCsr (const OrtMemoryInfo data_mem_info, const OrtSparseValuesParam &values, const int64_t inner_indices_data, size_t inner_indices_num, const int64_t *outer_indices_data, size_t outer_indices_num)
	The API will allocate memory using the allocator instance supplied to the CreateSparseTensor() API and copy the values and CSR indices into it. If data_mem_info specifies that the data is located at difference device than the allocator, a X-device copy will be performed if possible.

void	FillSparseTensorBlockSparse (const OrtMemoryInfo data_mem_info, const OrtSparseValuesParam &values, const Shape &indices_shape, const int32_t indices_data)
	The API will allocate memory using the allocator instance supplied to the CreateSparseTensor() API and copy the values and BlockSparse indices into it. If data_mem_info specifies that the data is located at difference device than the allocator, a X-device copy will be performed if possible.

Public Member Functions inherited from Ort::detail::ConstValueImpl< T >
template<typename R >
void	GetOpaqueData (const char domain, const char type_name, R &) const
	Obtains a pointer to a user defined data for experimental purposes.

bool	IsTensor () const
	Returns true if Value is a tensor, false for other types like map/sequence/etc.

bool	HasValue () const

size_t	GetCount () const
	< Return true if OrtValue contains data and returns false if the OrtValue is a None

Value	GetValue (int index, OrtAllocator *allocator) const

size_t	GetStringTensorDataLength () const
	This API returns a full length of string data contained within either a tensor or a sparse Tensor. For sparse tensor it returns a full length of stored non-empty strings (values). The API is useful for allocating necessary memory and calling GetStringTensorContent().

void	GetStringTensorContent (void buffer, size_t buffer_length, size_t offsets, size_t offsets_count) const
	The API copies all of the UTF-8 encoded string data contained within a tensor or a sparse tensor into a supplied buffer. Use GetStringTensorDataLength() to find out the length of the buffer to allocate. The user must also allocate offsets buffer with the number of entries equal to that of the contained strings.

template<typename R >
const R *	GetTensorData () const
	Returns a const typed pointer to the tensor contained data. No type checking is performed, the caller must ensure the type matches the tensor type.

const void *	GetTensorRawData () const
	Returns a non-typed pointer to a tensor contained data.

TypeInfo	GetTypeInfo () const
	The API returns type information for data contained in a tensor. For sparse tensors it returns type information for contained non-zero values. It returns dense shape for sparse tensors.

TensorTypeAndShapeInfo	GetTensorTypeAndShapeInfo () const
	The API returns type information for data contained in a tensor. For sparse tensors it returns type information for contained non-zero values. It returns dense shape for sparse tensors.

ConstMemoryInfo	GetTensorMemoryInfo () const
	This API returns information about the memory allocation used to hold data.

void	GetStringTensorElement (size_t buffer_length, size_t element_index, void *buffer) const
	The API copies UTF-8 encoded bytes for the requested string element contained within a tensor or a sparse tensor into a provided buffer. Use GetStringTensorElementLength() to obtain the length of the buffer to allocate.

std::string	GetStringTensorElement (size_t element_index) const
	Returns string tensor UTF-8 encoded string element. Use of this API is recommended over GetStringTensorElement() that takes void* buffer pointer.

size_t	GetStringTensorElementLength (size_t element_index) const
	The API returns a byte length of UTF-8 encoded string element contained in either a tensor or a spare tensor values.

OrtSparseFormat	GetSparseFormat () const
	The API returns the sparse data format this OrtValue holds in a sparse tensor. If the sparse tensor was not fully constructed, i.e. Use() or Fill() API were not used the value returned is ORT_SPARSE_UNDEFINED.

TensorTypeAndShapeInfo	GetSparseTensorValuesTypeAndShapeInfo () const
	The API returns type and shape information for stored non-zero values of the sparse tensor. Use GetSparseTensorValues() to obtain values buffer pointer.

TensorTypeAndShapeInfo	GetSparseTensorIndicesTypeShapeInfo (OrtSparseIndicesFormat format) const
	The API returns type and shape information for the specified indices. Each supported indices have their own enum values even if a give format has more than one kind of indices. Use GetSparseTensorIndicesData() to obtain pointer to indices buffer.

template<typename R >
const R *	GetSparseTensorIndicesData (OrtSparseIndicesFormat indices_format, size_t &num_indices) const
	The API retrieves a pointer to the internal indices buffer. The API merely performs a convenience data type casting on the return type pointer. Make sure you are requesting the right type, use GetSparseTensorIndicesTypeShapeInfo();.

bool	IsSparseTensor () const
	Returns true if the OrtValue contains a sparse tensor.

template<typename R >
const R *	GetSparseTensorValues () const
	The API returns a pointer to an internal buffer of the sparse tensor containing non-zero values. The API merely does casting. Make sure you are requesting the right data type by calling GetSparseTensorValuesTypeAndShapeInfo() first.

Public Member Functions inherited from Ort::detail::Base< T >
constexpr	Base ()=default

constexpr	Base (contained_type *p) noexcept

	~Base ()

	Base (const Base &)=delete

Base &	operator= (const Base &)=delete

	Base (Base &&v) noexcept

Base &	operator= (Base &&v) noexcept

constexpr	operator contained_type * () const noexcept

contained_type *	release ()
	Relinquishes ownership of the contained C object pointer The underlying object is not destroyed.

Static Public Member Functions
template<typename T >
static Value	CreateTensor (const OrtMemoryInfo info, T p_data, size_t p_data_element_count, const int64_t *shape, size_t shape_len)
	Creates a tensor with a user supplied buffer. Wraps OrtApi::CreateTensorWithDataAsOrtValue.

static Value	CreateTensor (const OrtMemoryInfo info, void p_data, size_t p_data_byte_count, const int64_t *shape, size_t shape_len, ONNXTensorElementDataType type)
	Creates a tensor with a user supplied buffer. Wraps OrtApi::CreateTensorWithDataAsOrtValue.

template<typename T >
static Value	CreateTensor (OrtAllocator allocator, const int64_t shape, size_t shape_len)
	Creates an OrtValue with a tensor using a supplied OrtAllocator. Wraps OrtApi::CreateTensorAsOrtValue. This overload will allocate the buffer for the tensor according to the supplied shape and data type. The allocated buffer will be owned by the returned OrtValue and will be freed when the OrtValue is released. The input data would need to be copied into the allocated buffer. This API is not suitable for strings.

static Value	CreateTensor (OrtAllocator allocator, const int64_t shape, size_t shape_len, ONNXTensorElementDataType type)
	Creates an OrtValue with a tensor using the supplied OrtAllocator. Wraps OrtApi::CreateTensorAsOrtValue. The allocated buffer will be owned by the returned OrtValue and will be freed when the OrtValue is released. The input data would need to be copied into the allocated buffer. This API is not suitable for strings.

static Value	CreateMap (const Value &keys, const Value &values)
	Creates an OrtValue with a Map Onnx type representation. The API would ref-count the supplied OrtValues and they will be released when the returned OrtValue is released. The caller may release keys and values after the call returns.

static Value	CreateSequence (const std::vector< Value > &values)
	Creates an OrtValue with a Sequence Onnx type representation. The API would ref-count the supplied OrtValues and they will be released when the returned OrtValue is released. The caller may release the values after the call returns.

template<typename T >
static Value	CreateOpaque (const char domain, const char type_name, const T &value)
	Creates an OrtValue wrapping an Opaque type. This is used for experimental support of non-tensor types.

template<typename T >
static Value	CreateSparseTensor (const OrtMemoryInfo info, T p_data, const Shape &dense_shape, const Shape &values_shape)
	This is a simple forwarding method to the other overload that helps deducing data type enum value from the type of the buffer.

static Value	CreateSparseTensor (const OrtMemoryInfo info, void p_data, const Shape &dense_shape, const Shape &values_shape, ONNXTensorElementDataType type)
	Creates an OrtValue instance containing SparseTensor. This constructs a sparse tensor that makes use of user allocated buffers. It does not make copies of the user provided data and does not modify it. The lifespan of user provided buffers should eclipse the life span of the resulting OrtValue. This call constructs an instance that only contain a pointer to non-zero values. To fully populate the sparse tensor call Use<Format>Indices() API below to supply a sparse format specific indices. This API is not suitable for string data. Use CreateSparseTensor() with allocator specified so strings can be properly copied into the allocated buffer.

template<typename T >
static Value	CreateSparseTensor (OrtAllocator *allocator, const Shape &dense_shape)
	This is a simple forwarding method to the below CreateSparseTensor. This helps to specify data type enum in terms of C++ data type. Use CreateSparseTensor<T>

static Value	CreateSparseTensor (OrtAllocator *allocator, const Shape &dense_shape, ONNXTensorElementDataType type)
	Creates an instance of OrtValue containing sparse tensor. The created instance has no data. The data must be supplied by on of the FillSparseTensor<Format>() methods that take both non-zero values and indices. The data will be copied into a buffer that would be allocated using the supplied allocator. Use this API to create OrtValues that contain sparse tensors with all supported data types including strings.

Additional Inherited Members
Protected Attributes inherited from Ort::detail::Base< T >
contained_type *	p_ {}

Detailed Description

Wrapper around OrtValue.

Member Typedef Documentation

◆ Base

using Ort::Value::Base = detail::ValueImpl<OrtValue>

◆ OrtSparseValuesParam

using Ort::Value::OrtSparseValuesParam = detail::OrtSparseValuesParam

◆ Shape

using Ort::Value::Shape = detail::Shape

Constructor & Destructor Documentation

◆ Value() [1/3]

Ort::Value::Value ( std::nullptr_t )

inlineexplicit

Create an empty Value object, must be assigned a valid one to be used.

◆ Value() [2/3]

Ort::Value::Value ( OrtValue * p )

inlineexplicit

Used for interop with the C API.

◆ Value() [3/3]

Ort::Value::Value ( Value && )

default

Member Function Documentation

◆ CreateMap()

static Value Ort::Value::CreateMap	(	const Value &	keys,
		const Value &	values
	)

static

Creates an OrtValue with a Map Onnx type representation. The API would ref-count the supplied OrtValues and they will be released when the returned OrtValue is released. The caller may release keys and values after the call returns.

Parameters

keys	an OrtValue containing a tensor with primitive data type keys.
values	an OrtValue that may contain a tensor. Ort currently supports only primitive data type values. Wraps OrtApi::CreateValue

◆ CreateOpaque()

template<typename T >

static Value Ort::Value::CreateOpaque	(	const char *	domain,
		const char *	type_name,
		const T &	value
	)

static

Creates an OrtValue wrapping an Opaque type. This is used for experimental support of non-tensor types.

Template Parameters

T	- the type of the value.

Parameters

domain	- zero terminated utf-8 string. Domain of the type.
type_name	- zero terminated utf-8 string. Name of the type.
value	- the value to be wrapped. Wraps OrtApi::CreateOpaqueValue

◆ CreateSequence()

static Value Ort::Value::CreateSequence ( const std::vector< Value > & values )

static

Creates an OrtValue with a Sequence Onnx type representation. The API would ref-count the supplied OrtValues and they will be released when the returned OrtValue is released. The caller may release the values after the call returns.

Parameters

values a vector of OrtValues that must have the same Onnx value type. Wraps OrtApi::CreateValue

◆ CreateSparseTensor() [1/4]

template<typename T >

static Value Ort::Value::CreateSparseTensor	(	const OrtMemoryInfo *	info,
		T *	p_data,
		const Shape &	dense_shape,
		const Shape &	values_shape
	)

static

This is a simple forwarding method to the other overload that helps deducing data type enum value from the type of the buffer.

Template Parameters

T	numeric datatype. This API is not suitable for strings.

Parameters

info	Memory description where the user buffers reside (CPU vs GPU etc)
p_data	pointer to the user supplied buffer, use nullptr for fully sparse tensors
dense_shape	a would be dense shape of the tensor
values_shape	non zero values shape. Use a single 0 shape for fully sparse tensors.

Returns

◆ CreateSparseTensor() [2/4]

static Value Ort::Value::CreateSparseTensor	(	const OrtMemoryInfo *	info,
		void *	p_data,
		const Shape &	dense_shape,
		const Shape &	values_shape,
		ONNXTensorElementDataType	type
	)

static

Creates an OrtValue instance containing SparseTensor. This constructs a sparse tensor that makes use of user allocated buffers. It does not make copies of the user provided data and does not modify it. The lifespan of user provided buffers should eclipse the life span of the resulting OrtValue. This call constructs an instance that only contain a pointer to non-zero values. To fully populate the sparse tensor call Use<Format>Indices() API below to supply a sparse format specific indices. This API is not suitable for string data. Use CreateSparseTensor() with allocator specified so strings can be properly copied into the allocated buffer.

Parameters

info	Memory description where the user buffers reside (CPU vs GPU etc)
p_data	pointer to the user supplied buffer, use nullptr for fully sparse tensors
dense_shape	a would be dense shape of the tensor
values_shape	non zero values shape. Use a single 0 shape for fully sparse tensors.
type	data type

Returns: Ort::Value instance containing SparseTensor

◆ CreateSparseTensor() [3/4]

template<typename T >

static Value Ort::Value::CreateSparseTensor	(	OrtAllocator *	allocator,
		const Shape &	dense_shape
	)

static

This is a simple forwarding method to the below CreateSparseTensor. This helps to specify data type enum in terms of C++ data type. Use CreateSparseTensor<T>

Template Parameters

T	numeric data type only. String data enum must be specified explicitly.

Parameters

allocator	allocator to use
dense_shape	a would be dense shape of the tensor

Returns: Ort::Value

◆ CreateSparseTensor() [4/4]

static Value Ort::Value::CreateSparseTensor	(	OrtAllocator *	allocator,
		const Shape &	dense_shape,
		ONNXTensorElementDataType	type
	)

static

Creates an instance of OrtValue containing sparse tensor. The created instance has no data. The data must be supplied by on of the FillSparseTensor<Format>() methods that take both non-zero values and indices. The data will be copied into a buffer that would be allocated using the supplied allocator. Use this API to create OrtValues that contain sparse tensors with all supported data types including strings.

Parameters

allocator	allocator to use. The allocator lifespan must eclipse that of the resulting OrtValue
dense_shape	a would be dense shape of the tensor
type	data type

Returns: an instance of Ort::Value

◆ CreateTensor() [1/4]

template<typename T >

static Value Ort::Value::CreateTensor	(	const OrtMemoryInfo *	info,
		T *	p_data,
		size_t	p_data_element_count,
		const int64_t *	shape,
		size_t	shape_len
	)

static

Creates a tensor with a user supplied buffer. Wraps OrtApi::CreateTensorWithDataAsOrtValue.

Template Parameters

T	The numeric datatype. This API is not suitable for strings.

Parameters

info	Memory description of where the p_data buffer resides (CPU vs GPU etc).
p_data	Pointer to the data buffer.
p_data_element_count	The number of elements in the data buffer.
shape	Pointer to the tensor shape dimensions.
shape_len	The number of tensor shape dimensions.

◆ CreateTensor() [2/4]

static Value Ort::Value::CreateTensor	(	const OrtMemoryInfo *	info,
		void *	p_data,
		size_t	p_data_byte_count,
		const int64_t *	shape,
		size_t	shape_len,
		ONNXTensorElementDataType	type
	)

static

Creates a tensor with a user supplied buffer. Wraps OrtApi::CreateTensorWithDataAsOrtValue.

Parameters

info	Memory description of where the p_data buffer resides (CPU vs GPU etc).
p_data	Pointer to the data buffer.
p_data_byte_count	The number of bytes in the data buffer.
shape	Pointer to the tensor shape dimensions.
shape_len	The number of tensor shape dimensions.
type	The data type.

◆ CreateTensor() [3/4]

template<typename T >

static Value Ort::Value::CreateTensor	(	OrtAllocator *	allocator,
		const int64_t *	shape,
		size_t	shape_len
	)

static

Creates an OrtValue with a tensor using a supplied OrtAllocator. Wraps OrtApi::CreateTensorAsOrtValue. This overload will allocate the buffer for the tensor according to the supplied shape and data type. The allocated buffer will be owned by the returned OrtValue and will be freed when the OrtValue is released. The input data would need to be copied into the allocated buffer. This API is not suitable for strings.

Template Parameters

T	The numeric datatype. This API is not suitable for strings.

Parameters

allocator	The allocator to use.
shape	Pointer to the tensor shape dimensions.
shape_len	The number of tensor shape dimensions.

◆ CreateTensor() [4/4]

static Value Ort::Value::CreateTensor	(	OrtAllocator *	allocator,
		const int64_t *	shape,
		size_t	shape_len,
		ONNXTensorElementDataType	type
	)

static

Creates an OrtValue with a tensor using the supplied OrtAllocator. Wraps OrtApi::CreateTensorAsOrtValue. The allocated buffer will be owned by the returned OrtValue and will be freed when the OrtValue is released. The input data would need to be copied into the allocated buffer. This API is not suitable for strings.

Parameters

allocator	The allocator to use.
shape	Pointer to the tensor shape dimensions.
shape_len	The number of tensor shape dimensions.
type	The data type.

◆ GetConst()

ConstValue Ort::Value::GetConst ( ) const

inline

◆ GetUnowned()

UnownedValue Ort::Value::GetUnowned ( ) const

inline

◆ operator=()

Value & Ort::Value::operator= ( Value && )

default

Public Types

Public Member Functions

Static Public Member Functions

Additional Inherited Members

Detailed Description

Member Typedef Documentation

◆ Base

◆ OrtSparseValuesParam

◆ Shape

Constructor & Destructor Documentation

◆ Value() [1/3]

◆ Value() [2/3]

◆ Value() [3/3]

Member Function Documentation

◆ CreateMap()

◆ CreateOpaque()

◆ CreateSequence()

◆ CreateSparseTensor() [1/4]

◆ CreateSparseTensor() [2/4]

◆ CreateSparseTensor() [3/4]

◆ CreateSparseTensor() [4/4]

◆ CreateTensor() [1/4]

◆ CreateTensor() [2/4]

◆ CreateTensor() [3/4]

◆ CreateTensor() [4/4]

◆ GetConst()

◆ GetUnowned()

◆ operator=()