This class represents tensor - map-like storage for inference result information, such as output blob description (output layer dims, layout, rank, precision, etc.), inference result in a raw and interpreted forms. Tensor is based on GstStructure and, in general, can contain arbitrary (user-defined) fields of simplest data types, like integers, floats & strings. Tensor can contain raw inference result (such Tensor is produced by gvainference in Gstreamer pipeline), detection result (such Tensor is produced by gvadetect in Gstreamer pipeline and it's called detection Tensor), or both raw & interpreted inference results (such Tensor is produced by gvaclassify in Gstreamer pipeline). Tensors can be created and used on their own, or they can be created within RegionOfInterest or VideoFrame instances. Usually, in Gstreamer pipeline with GVA elements (gvadetect, gvainference, gvaclassify) Tensor objects will be available for access and modification from RegionOfInterest and VideoFrame instances. More...

#include <tensor.h>

Public Types
enum	Precision { Precision::UNSPECIFIED = GVA_PRECISION_UNSPECIFIED, Precision::FP32 = GVA_PRECISION_FP32, Precision::U8 = GVA_PRECISION_U8 }
	Describes tensor precision. More...

enum	Layout { Layout::ANY = GVA_LAYOUT_ANY, Layout::NCHW = GVA_LAYOUT_NCHW, Layout::NHWC = GVA_LAYOUT_NHWC, Layout::NC = GVA_LAYOUT_NC }
	Describes tensor layout. More...

Public Member Functions
template<class T >
const std::vector< T >	data () const
	Get raw inference output blob data. More...

std::vector< guint >	dims () const
	Get inference result blob dimensions info. More...

Precision	precision () const
	Get inference output blob precision. More...

Layout	layout () const
	Get inference result blob layout. More...

std::string	layer_name () const
	Get inference result blob layer name. More...

std::string	model_name () const
	Get model name which was used for inference. More...

std::string	format () const
	Get data format as specified in model pre/post-processing json configuration. More...

std::string	name () const
	Get tensor name as a string. More...

double	confidence () const
	Get confidence of detection or classification result extracted from the tensor. More...

std::string	label () const
	Get label. This label is set for Tensor instances produced by gvaclassify element. It will throw an exception if called for detection Tensor. To get detection class label, use RegionOfInterest::label. More...

std::vector< std::string >	fields () const
	Get vector of fields contained in Tensor instance. More...

bool	has_field (const std::string &field_name) const
	Check if Tensor instance has field. More...

std::string	get_string (const std::string &field_name, const std::string &default_value=std::string()) const
	Get string contained in value stored at field_name. More...

int	get_int (const std::string &field_name, int32_t default_value=0) const
	Get int contained in value stored at field_name. More...

double	get_double (const std::string &field_name, double default_value=0) const
	Get double contained in value stored at field_name. More...

void	set_string (const std::string &field_name, const std::string &value)
	Set field_name with string value. More...

void	set_int (const std::string &field_name, int value)
	Set field_name with int value. More...

void	set_double (const std::string &field_name, double value)
	Set field_name with double value. More...

void	set_name (const std::string &name)
	Set Tensor instance's name.

void	set_label (const std::string &label)
	Set label. It will throw an exception if called for detection Tensor. More...

std::string	precision_as_string () const
	Get inference result blob precision as a string. More...

std::string	layout_as_string () const
	Get inference result blob layout as a string. More...

std::string	element_id () const
	Get inference-id property value of GVA element from which this Tensor came. More...

int	label_id () const
	Get label id. More...

bool	is_detection () const
	Check if this Tensor is detection Tensor (contains detection results) More...

	Tensor (GstStructure *structure)
	Construct Tensor instance from GstStructure. Tensor does not own structure, so if you use this consrtuctor, free structure after Tensor's lifetime, if needed. More...

GstStructure *	gst_structure () const
	Get ptr to underlying GstStructure. More...

Protected Attributes
GstStructure *	_structure
	ptr to GstStructure that contains all tensor (inference results) data & info.

Friends
class	VideoFrame

class	AudioFrame

Detailed Description

This class represents tensor - map-like storage for inference result information, such as output blob description (output layer dims, layout, rank, precision, etc.), inference result in a raw and interpreted forms. Tensor is based on GstStructure and, in general, can contain arbitrary (user-defined) fields of simplest data types, like integers, floats & strings. Tensor can contain raw inference result (such Tensor is produced by gvainference in Gstreamer pipeline), detection result (such Tensor is produced by gvadetect in Gstreamer pipeline and it's called detection Tensor), or both raw & interpreted inference results (such Tensor is produced by gvaclassify in Gstreamer pipeline). Tensors can be created and used on their own, or they can be created within RegionOfInterest or VideoFrame instances. Usually, in Gstreamer pipeline with GVA elements (gvadetect, gvainference, gvaclassify) Tensor objects will be available for access and modification from RegionOfInterest and VideoFrame instances.

Member Enumeration Documentation

◆ Layout

enum GVA::Tensor::Layout

strong

Describes tensor layout.

Enumerator
ANY	unspecified layout
NCHW	NCWH layout
NHWC	NHWC layout
NC	NC layout

◆ Precision

enum GVA::Tensor::Precision

strong

Describes tensor precision.

Enumerator
UNSPECIFIED	default value
FP32	32bit floating point value
U8	unsignned 8bit integer value

Constructor & Destructor Documentation

◆ Tensor()

GVA::Tensor::Tensor ( GstStructure * structure )

inline

Construct Tensor instance from GstStructure. Tensor does not own structure, so if you use this consrtuctor, free structure after Tensor's lifetime, if needed.

Parameters

structure GstStructure to create Tensor instance from.

Member Function Documentation

◆ confidence()

double GVA::Tensor::confidence ( ) const

inline

Get confidence of detection or classification result extracted from the tensor.

Returns: confidence of inference result as a double, 0 if failed to get

◆ data()

template<class T >

const std::vector<T> GVA::Tensor::data ( ) const

inline

Get raw inference output blob data.

Template Parameters

T	type to interpret blob data

Returns: vector of values of type T representing raw inference data, empty vector if data can't be read

◆ dims()

std::vector<guint> GVA::Tensor::dims ( ) const

inline

Get inference result blob dimensions info.

Returns: vector of dimensions. Empty vector if dims are not set

◆ element_id()

std::string GVA::Tensor::element_id ( ) const

inline

Get inference-id property value of GVA element from which this Tensor came.

Returns: inference-id property value of GVA element from which this Tensor came, empty string if failed to get

◆ fields()

std::vector<std::string> GVA::Tensor::fields ( ) const

inline

Get vector of fields contained in Tensor instance.

Returns: vector of fields contained in Tensor instance

◆ format()

std::string GVA::Tensor::format ( ) const

inline

Get data format as specified in model pre/post-processing json configuration.

Returns: format as a string, empty string if failed to get

◆ get_double()

double GVA::Tensor::get_double	(	const std::string &	field_name,
		double	default_value = `0`
	)		const

inline

Get double contained in value stored at field_name.

Parameters

field_name	field name
default_value	default value

Returns: double value stored at field_name if field_name is found and contains an double, default_value otherwise

◆ get_int()

int GVA::Tensor::get_int	(	const std::string &	field_name,
		int32_t	default_value = `0`
	)		const

inline

Get int contained in value stored at field_name.

Parameters

field_name	field name
default_value	default value

Returns: int value stored at field_name if field_name is found and contains an int, default_value otherwise

◆ get_string()

std::string GVA::Tensor::get_string	(	const std::string &	field_name,
		const std::string &	default_value = `std::string()`
	)		const

inline

Get string contained in value stored at field_name.

Parameters

field_name	field name
default_value	default value

Returns: string value stored at field_name if field_name is found and contains a string, default_value string otherwise

◆ gst_structure()

GstStructure* GVA::Tensor::gst_structure ( ) const

inline

Get ptr to underlying GstStructure.

Returns: ptr to underlying GstStructure

◆ has_field()

bool GVA::Tensor::has_field ( const std::string & field_name ) const

inline

Check if Tensor instance has field.

Parameters

field_name field name

Returns: True if field with this name is found, False otherwise

◆ is_detection()

bool GVA::Tensor::is_detection ( ) const

inline

Check if this Tensor is detection Tensor (contains detection results)

Returns: True if tensor contains detection results, False otherwise

◆ label()

std::string GVA::Tensor::label ( ) const

inline

Get label. This label is set for Tensor instances produced by gvaclassify element. It will throw an exception if called for detection Tensor. To get detection class label, use RegionOfInterest::label.

Returns: label as a string, empty string if failed to get

◆ label_id()

int GVA::Tensor::label_id ( ) const

inline

Get label id.

Returns: label id as an int, 0 if failed to get

◆ layer_name()

std::string GVA::Tensor::layer_name ( ) const

inline

Get inference result blob layer name.

Returns: layer name as a string, empty string if failed to get

◆ layout()

Layout GVA::Tensor::layout ( ) const

inline

Get inference result blob layout.

Returns: Enum Layout, Layout::ANY if can't be read

◆ layout_as_string()

std::string GVA::Tensor::layout_as_string ( ) const

inline

Get inference result blob layout as a string.

Returns: layout as a string, "ANY" if can't be read

◆ model_name()

std::string GVA::Tensor::model_name ( ) const

inline

Get model name which was used for inference.

Returns: model name as a string, empty string if failed to get

◆ name()

std::string GVA::Tensor::name ( ) const

inline

Get tensor name as a string.

Returns: Tensor instance's name

◆ precision()

Precision GVA::Tensor::precision ( ) const

inline

Get inference output blob precision.

Returns: Enum Precision, Precision::UNSPECIFIED if can't be read

◆ precision_as_string()

std::string GVA::Tensor::precision_as_string ( ) const

inline

Get inference result blob precision as a string.

Returns: precision as a string, "ANY" if can't be read

◆ set_double()

void GVA::Tensor::set_double	(	const std::string &	field_name,
		double	value
	)

inline

Set field_name with double value.

Parameters

field_name	field name
value	value to set

◆ set_int()

void GVA::Tensor::set_int	(	const std::string &	field_name,
		int	value
	)

inline

Set field_name with int value.

Parameters

field_name	field name
value	value to set

◆ set_label()

void GVA::Tensor::set_label ( const std::string & label )

inline

Set label. It will throw an exception if called for detection Tensor.

Parameters

label label name as a string

◆ set_string()

void GVA::Tensor::set_string	(	const std::string &	field_name,
		const std::string &	value
	)

inline

Set field_name with string value.

Parameters

field_name	field name
value	value to set

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

/home/gst-libs/gst/videoanalytics/tensor.h

Public Types

Public Member Functions

Protected Attributes

Friends

Detailed Description

Member Enumeration Documentation

◆ Layout

◆ Precision

Constructor & Destructor Documentation

◆ Tensor()

Member Function Documentation

◆ confidence()

◆ data()

◆ dims()

◆ element_id()

◆ fields()

◆ format()

◆ get_double()

◆ get_int()

◆ get_string()

◆ gst_structure()

◆ has_field()

◆ is_detection()

◆ label()

◆ label_id()

◆ layer_name()

◆ layout()

◆ layout_as_string()

◆ model_name()

◆ name()

◆ precision()

◆ precision_as_string()

◆ set_double()

◆ set_int()

◆ set_label()

◆ set_string()