InferenceEngine::Builder::DetectionOutputLayer Class Reference

The class represents a builder for ArgMax layer. More...

#include <ie_detection_output_layer.hpp>

Inheritance diagram for InferenceEngine::Builder::DetectionOutputLayer:

Collaboration diagram for InferenceEngine::Builder::DetectionOutputLayer:

Public Member Functions
	DetectionOutputLayer (const std::string &name="")
	The constructor creates a builder with the name. More...
	DetectionOutputLayer (const Layer::Ptr &layer)
	The constructor creates a builder from generic builder. More...
	DetectionOutputLayer (const Layer::CPtr &layer)
	The constructor creates a builder from generic builder. More...
DetectionOutputLayer &	setName (const std::string &name)
	Sets the name for the layer. More...
const Port &	getOutputPort () const
	Returns output port. More...
DetectionOutputLayer &	setOutputPort (const Port &port)
	Sets output port. More...
const std::vector< Port > &	getInputPorts () const
	Returns input ports. More...
DetectionOutputLayer &	setInputPorts (const std::vector< Port > &ports)
	Sets input ports. More...
size_t	getNumClasses () const
	Returns number of classes. More...
DetectionOutputLayer &	setNumClasses (size_t num)
	Sets number of classes to be predict. More...
int	getBackgroudLabelId () const
	Returns background label ID. More...
DetectionOutputLayer &	setBackgroudLabelId (int labelId)
	Sets background label ID. More...
int	getTopK () const
	Returns maximum number of results to be kept on NMS stage. More...
DetectionOutputLayer &	setTopK (int topK)
	Sets maximum number of results to be kept on NMS stage. More...
int	getKeepTopK () const
	Returns number of total boxes to be kept per image after NMS step. More...
DetectionOutputLayer &	setKeepTopK (int topK)
	Sets number of total boxes to be kept per image after NMS step. More...
int	getNumOrientClasses () const
	Returns number of oriented classes. More...
DetectionOutputLayer &	setNumOrientClasses (int numClasses)
	Sets number of oriented classes. More...
std::string	getCodeType () const
	Returns type of coding method for bounding boxes. More...
DetectionOutputLayer &	setCodeType (std::string type)
	Sets type of coding method for bounding boxes. More...
int	getInterpolateOrientation () const
	Returns interpolate orientation. More...
DetectionOutputLayer &	setInterpolateOrientation (int orient)
	Sets interpolate orientation. More...
float	getNMSThreshold () const
	Returns threshold to be used in NMS stage. More...
DetectionOutputLayer &	setNMSThreshold (float threshold)
	Sets threshold to be used in NMS stage. More...
float	getConfidenceThreshold () const
	Returns confidence threshold. More...
DetectionOutputLayer &	setConfidenceThreshold (float threshold)
	Sets confidence threshold. More...
bool	getShareLocation () const
	Returns share location. More...
DetectionOutputLayer &	setShareLocation (bool flag)
	Sets share location. More...
bool	getVariantEncodedInTarget () const
	Returns encoded settings. More...
DetectionOutputLayer &	setVariantEncodedInTarget (bool flag)
	Sets encoded settings. More...
Public Member Functions inherited from InferenceEngine::Builder::LayerDecorator
	LayerDecorator (const std::string &type, const std::string &name)
	The constructor creates layer builders with layer type and layer name. More...
	LayerDecorator (const Layer::Ptr &layer)
	The constructor creates layer builders from reference to generic layer builder. More...
	LayerDecorator (const Layer::CPtr &layer)
	The constructor creates layer builders from reference to generic layer builder. More...
	LayerDecorator (const LayerDecorator &rval)
	The copy constructor. More...
LayerDecorator &	operator= (const LayerDecorator &rval)
	Copy operator for LayerDecorator. More...
virtual	~LayerDecorator ()=default
	Virtual destructor.
virtual	operator Layer () const
	The operator creates generic builder. More...
virtual	operator Layer::Ptr ()
	The operator creates generic builder. More...
virtual	operator Layer::CPtr () const
	The operator creates generic builder. More...
const std::string &	getType () const
	Returns layer type. More...
const std::string &	getName () const
	Returns layer name. More...

Additional Inherited Members
Protected Member Functions inherited from InferenceEngine::Builder::LayerDecorator
Layer::Ptr &	getLayer ()
const Layer::CPtr	getLayer () const
void	checkType (const std::string &type) const
Protected Attributes inherited from InferenceEngine::Builder::LayerDecorator
Layer::CPtr	cLayer

Detailed Description

The class represents a builder for ArgMax layer.

Constructor & Destructor Documentation

DetectionOutputLayer() [1/3]

InferenceEngine::Builder::DetectionOutputLayer::DetectionOutputLayer ( const std::string &  name = "" )

explicit

The constructor creates a builder with the name.

Parameters

name	Layer name

DetectionOutputLayer() [2/3]

InferenceEngine::Builder::DetectionOutputLayer::DetectionOutputLayer ( const Layer::Ptr &  layer )

explicit

The constructor creates a builder from generic builder.

Parameters

layer

pointer to generic builder

DetectionOutputLayer() [3/3]

InferenceEngine::Builder::DetectionOutputLayer::DetectionOutputLayer ( const Layer::CPtr &  layer )

explicit

The constructor creates a builder from generic builder.

Parameters

layer

constant pointer to generic builder

Member Function Documentation

getBackgroudLabelId()

int InferenceEngine::Builder::DetectionOutputLayer::getBackgroudLabelId

(

)

const

Returns background label ID.

Returns

Background ID

getCodeType()

std::string InferenceEngine::Builder::DetectionOutputLayer::getCodeType

(

)

const

Returns type of coding method for bounding boxes.

Returns

String with code type

getConfidenceThreshold()

float InferenceEngine::Builder::DetectionOutputLayer::getConfidenceThreshold

(

)

const

Returns confidence threshold.

Returns

Threshold

getInputPorts()

const std::vector<Port>& InferenceEngine::Builder::DetectionOutputLayer::getInputPorts

(

)

const

Returns input ports.

Returns

Vector of input ports

getInterpolateOrientation()

int InferenceEngine::Builder::DetectionOutputLayer::getInterpolateOrientation

(

)

const

Returns interpolate orientation.

Returns

Interpolate orientation

getKeepTopK()

int InferenceEngine::Builder::DetectionOutputLayer::getKeepTopK

(

)

const

Returns number of total boxes to be kept per image after NMS step.

Returns

Keep top K

getNMSThreshold()

float InferenceEngine::Builder::DetectionOutputLayer::getNMSThreshold

(

)

const

Returns threshold to be used in NMS stage.

Returns

Threshold

getNumClasses()

size_t InferenceEngine::Builder::DetectionOutputLayer::getNumClasses

(

)

const

Returns number of classes.

Returns

Number of classes

getNumOrientClasses()

int InferenceEngine::Builder::DetectionOutputLayer::getNumOrientClasses

(

)

const

Returns number of oriented classes.

Returns

Number of oriented classes

getOutputPort()

const Port& InferenceEngine::Builder::DetectionOutputLayer::getOutputPort

(

)

const

Returns output port.

Returns

Output port

getShareLocation()

bool InferenceEngine::Builder::DetectionOutputLayer::getShareLocation

(

)

const

Returns share location.

Returns

true if bounding boxes are shared among different classes

getTopK()

int InferenceEngine::Builder::DetectionOutputLayer::getTopK

(

)

const

Returns maximum number of results to be kept on NMS stage.

Returns

Top K

getVariantEncodedInTarget()

bool InferenceEngine::Builder::DetectionOutputLayer::getVariantEncodedInTarget

(

)

const

Returns encoded settings.

Returns

true if variance is encoded in target

setBackgroudLabelId()

DetectionOutputLayer& InferenceEngine::Builder::DetectionOutputLayer::setBackgroudLabelId

(

int

labelId

)

Sets background label ID.

Parameters

labelId

Background ID if there is no background class, set it to -1.

Returns

reference to layer builder

setCodeType()

DetectionOutputLayer& InferenceEngine::Builder::DetectionOutputLayer::setCodeType

(

std::string

type

)

Sets type of coding method for bounding boxes.

Parameters

type

Type

Returns

reference to layer builder

setConfidenceThreshold()

DetectionOutputLayer& InferenceEngine::Builder::DetectionOutputLayer::setConfidenceThreshold

(

float

threshold

)

Sets confidence threshold.

Parameters

threshold

Threshold

Returns

reference to layer builder

setInputPorts()

DetectionOutputLayer& InferenceEngine::Builder::DetectionOutputLayer::setInputPorts

(

const std::vector< Port > &

ports

)

Sets input ports.

Parameters

ports

Vector of input ports

Returns

reference to layer builder

setInterpolateOrientation()

DetectionOutputLayer& InferenceEngine::Builder::DetectionOutputLayer::setInterpolateOrientation

(

int

orient

)

Sets interpolate orientation.

Parameters

orient

Orientation

Returns

reference to layer builder

setKeepTopK()

DetectionOutputLayer& InferenceEngine::Builder::DetectionOutputLayer::setKeepTopK

(

int

topK

)

Sets number of total boxes to be kept per image after NMS step.

Parameters

topK	Keep top K

Returns

reference to layer builder

setName()

DetectionOutputLayer& InferenceEngine::Builder::DetectionOutputLayer::setName

(

const std::string &

name

)

Sets the name for the layer.

Parameters

name	Layer name

Returns

reference to layer builder

setNMSThreshold()

DetectionOutputLayer& InferenceEngine::Builder::DetectionOutputLayer::setNMSThreshold

(

float

threshold

)

Sets threshold to be used in NMS stage.

Parameters

threshold

NMS threshold

Returns

reference to layer builder

setNumClasses()

DetectionOutputLayer& InferenceEngine::Builder::DetectionOutputLayer::setNumClasses

(

size_t

num

)

Sets number of classes to be predict.

Parameters

num	Number of classes

Returns

reference to layer builder

setNumOrientClasses()

DetectionOutputLayer& InferenceEngine::Builder::DetectionOutputLayer::setNumOrientClasses

(

int

numClasses

)

Sets number of oriented classes.

Parameters

numClasses

Number of classes

Returns

reference to layer builder

setOutputPort()

DetectionOutputLayer& InferenceEngine::Builder::DetectionOutputLayer::setOutputPort

(

const Port &

port

)

Sets output port.

Parameters

port	Output port

Returns

reference to layer builder

setShareLocation()

DetectionOutputLayer& InferenceEngine::Builder::DetectionOutputLayer::setShareLocation

(

bool

flag

)

Sets share location.

Parameters

flag	true if bounding boxes are shared among different classes

Returns

reference to layer builder

setTopK()

DetectionOutputLayer& InferenceEngine::Builder::DetectionOutputLayer::setTopK

(

int

topK

)

Sets maximum number of results to be kept on NMS stage.

Parameters

topK

Top K

Returns

reference to layer builder

setVariantEncodedInTarget()

DetectionOutputLayer& InferenceEngine::Builder::DetectionOutputLayer::setVariantEncodedInTarget

(

bool

flag

)

Sets encoded settings.

Parameters

flag	true if variance is encoded in target

Returns

reference to layer builder

---

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

• builders/ie_detection_output_layer.hpp