Class InferenceEngine::Blob

class Blob

This class represents a universal container in the Inference Engine.

Note

Each Blob implementation must be derived from this Blob class directly or indirectly

Subclassed by InferenceEngine::CompoundBlob, InferenceEngine::MemoryBlob

Public Types

using Ptr = std::shared_ptr<Blob>

A smart pointer containing Blob object.

using CPtr = std::shared_ptr<const Blob>

A smart pointer to the const Blob object.

Public Functions

virtual ~Blob()

Blob virtual destructor.

template<typename T, typename std::enable_if<!std::is_pointer<T>::value && !std::is_reference<T>::value, int>::type = 0, typename std::enable_if<std::is_base_of<Blob, T>::value, int>::type = 0>
inline bool is() noexcept

Checks if the Blob object can be cast to the type T*.

Template Parameters

T – Type to be checked. Must represent a class derived from the Blob

Returns

true if this object can be dynamically cast to the type T*. Otherwise, false

template<typename T, typename std::enable_if<!std::is_pointer<T>::value && !std::is_reference<T>::value, int>::type = 0, typename std::enable_if<std::is_base_of<Blob, T>::value, int>::type = 0>
inline bool is() const noexcept

Checks if the Blob object can be cast to the type const T*.

Template Parameters

T – Type to be checked. Must represent a class derived from the Blob

Returns

true if this object can be dynamically cast to the type const T*. Otherwise, false

template<typename T, typename std::enable_if<!std::is_pointer<T>::value && !std::is_reference<T>::value, int>::type = 0, typename std::enable_if<std::is_base_of<Blob, T>::value && !std::is_same<RemoteBlob, T>::value, int>::type = 0>
inline T *as() noexcept

Casts this Blob object to the type T*.

Use InferenceEngine::as() to operate with shared Blob objects instead of raw pointers

Template Parameters

T – Type to cast to. Must represent a class derived from the Blob

Returns

Raw pointer to the object of the type T or nullptr on error

template<typename T, typename std::enable_if<!std::is_pointer<T>::value && !std::is_reference<T>::value, int>::type = 0, typename std::enable_if<std::is_same<RemoteBlob, T>::value, int>::type = 0>
inline T *as() noexcept

Casts this Blob object to the type RemoteBlob.

Use InferenceEngine::as() to operate with shared Blob objects instead of raw pointers

Template Parameters

T – Type to cast to. Must represent a class derived from the Blob

Returns

Raw pointer to the object of the type T or nullptr on error

template<typename T, typename std::enable_if<!std::is_pointer<T>::value && !std::is_reference<T>::value, int>::type = 0, typename std::enable_if<std::is_base_of<Blob, T>::value && !std::is_same<RemoteBlob, T>::value, int>::type = 0>
inline const T *as() const noexcept

Casts this Blob object to the type const T*.

Use InferenceEngine::as() to operate with shared Blob objects instead of raw pointers

Template Parameters

T – Type to cast to. Must represent a class derived from the Blob

Returns

Raw pointer to the object of the type const T or nullptr on error

template<typename T, typename std::enable_if<!std::is_pointer<T>::value && !std::is_reference<T>::value, int>::type = 0, typename std::enable_if<std::is_same<RemoteBlob, T>::value, int>::type = 0>
inline const T *as() const noexcept

Casts this Blob object to the type RemoteBlob.

Use InferenceEngine::as() to operate with shared Blob objects instead of raw pointers

Template Parameters

T – Type to cast to. Must represent a class derived from the Blob

Returns

Raw pointer to the object of the type T or nullptr on error

inline explicit Blob(const TensorDesc &tensorDesc)

Constructor. Creates an empty Blob object with the specified precision.

Parameters

tensorDesc – Defines the layout and dims of the blob

inline virtual const TensorDesc &getTensorDesc() const noexcept

Returns the tensor description.

Returns

A const reference to a tensor descriptor

inline virtual TensorDesc &getTensorDesc() noexcept

Returns the tensor description.

Returns

A reference to a tensor descriptor

inline virtual size_t size() const noexcept

By default, returns the total number of elements (a product of all the dims or 1 for scalar)

Return value and its interpretation heavily depend on the blob type

Returns

The total number of elements

inline virtual size_t byteSize() const

Returns the size of the current Blob in bytes.

Returns

Blob’s size in bytes

virtual size_t element_size() const = 0

Provides the number of bytes per element.

Deprecated:

Cast to MemoryBlob and use its API instead. Blob class can represent compound blob, which do not refer to the only solid memory.

The overall Blob capacity is size() * element_size(). Abstract method.

Returns

Returns the number of bytes per element

virtual void allocate() noexcept = 0

Allocates memory to store the data.

Abstract method.

virtual bool deallocate() noexcept = 0

Releases previously allocated data.

Abstract method.

Returns

True if deallocation happens successfully, false otherwise.

virtual void setShape(const SizeVector &dims)

Set new shape for blob, deallocate/allocate if new total size is bigger than previous one.

Parameters

dims – new shape

virtual Blob::Ptr createROI(const ROI &roi) const

Creates a blob describing given ROI object based on the current blob with memory sharing.

Note: default implementation throws “not implemented” exception.

Parameters

roi – A ROI object inside of the current blob.

Returns

A shared pointer to the newly created ROI blob.

virtual Blob::Ptr createROI(const std::vector<std::size_t> &begin, const std::vector<std::size_t> &end) const

Creates a blob describing given ROI object based on the current blob with memory sharing.

Note: default implementation may throws “not implemented” exception.

Parameters

• begin – A ROI start coordinate

• end – A ROI end coordinate

Returns

A shared pointer to the newly created ROI blob.

Public Static Functions

static Ptr CreateFromData(const DataPtr &data)

Creates a TBlob<> object from a Data node.

Parameters

data – A reference to a smart pointer of the Data node

Returns

Smart pointer to TBlob<> with the relevant C type to the precision of the data node