MMind.DL.InferEngine Class Reference

This class defines the inference engine. More...

Public Member Functions
	InferEngine ()
	Initializes a new instance of the InferEngine class.
StatusCode	Create (string packPath, BackendType backendType=BackendType.GpuDefault, uint deviceId=0)
	Create an inference engine.
StatusCode	SetBatchSizeAndFloatPrecision (uint batchSize, FloatPrecisionType floatPrecision, uint moduleIdx)
	Set the batch size and floating-point precision of the inference engine.
StatusCode	SetUnsupSegParam (float defectThreshold, float nonDefectThreshold, uint moduleIdx)
	Set the defect threshold and non-defect threshold for the unsupervised segmentation.
List< DLAlgoType >	GetModuleTypes ()
	Get the list of model types.
StatusCode	Infer (List< MMindImage > images)
	Run image inference using the model package inference engine.
StatusCode	GetResults (out List< Result > results)
	Get the model inference result.
StatusCode	ResultVisualization (List< MMindImage > images)
	Draw all the model results onto the images.
StatusCode	ModuleResultVisualization (List< MMindImage > images, uint moduleIndex)
	Draw the model results of the specified index onto the images.
void	Release ()
	Release the memory of the inference engine.

Detailed Description

This class defines the inference engine.

Constructor & Destructor Documentation

InferEngine()

MMind.DL.InferEngine.InferEngine

(

)

Initializes a new instance of the InferEngine class.

Member Function Documentation

Create()

StatusCode MMind.DL.InferEngine.Create	(	string	packPath,
		BackendType	backendType = BackendType::GpuDefault,
		uint	deviceId = 0
	)

Create an inference engine.

Parameters

packPath	The path to the model package exported from Mech-DLK.
backendType	See BackendType for details.
deviceId	The index of the specified GPU during model inference.

Returns

See StatusCode for details.

GetModuleTypes()

List< DLAlgoType > MMind.DL.InferEngine.GetModuleTypes

(

)

Get the list of model types.

Returns

See DLAlgoType for details.

GetResults()

StatusCode MMind.DL.InferEngine.GetResults

(

out List< Result >

results

)

Get the model inference result.

Parameters

results

See Result for details.

Returns

See StatusCode for details.

Infer()

StatusCode MMind.DL.InferEngine.Infer

(

List< MMindImage >

images

)

Run image inference using the model package inference engine.

Parameters

images

See MMindImage for details.

Returns

See StatusCode for details.

ModuleResultVisualization()

StatusCode MMind.DL.InferEngine.ModuleResultVisualization	(	List< MMindImage >	images,
		uint	moduleIndex
	)

Draw the model results of the specified index onto the images.

Parameters

images	See MMindImage for details.
moduleIndex	Specified model index in the model package.

Returns

See StatusCode for details.

Release()

void MMind.DL.InferEngine.Release

(

)

Release the memory of the inference engine.

ResultVisualization()

StatusCode MMind.DL.InferEngine.ResultVisualization

(

List< MMindImage >

images

)

Draw all the model results onto the images.

Parameters

images

See MMindImage for details.

Returns

See StatusCode for details.

SetBatchSizeAndFloatPrecision()

StatusCode MMind.DL.InferEngine.SetBatchSizeAndFloatPrecision	(	uint	batchSize,
		FloatPrecisionType	floatPrecision,
		uint	moduleIdx
	)

Set the batch size and floating-point precision of the inference engine.

Parameters

batchSize	The batch size of the model package.
floatPrecision	See FloatPrecisionType for details.
moduleIdx	Specified model index in the model package.

Returns

See StatusCode for details.

SetUnsupSegParam()

StatusCode MMind.DL.InferEngine.SetUnsupSegParam	(	float	defectThreshold,
		float	nonDefectThreshold,
		uint	moduleIdx
	)

Set the defect threshold and non-defect threshold for the unsupervised segmentation.

Parameters

defectThreshold	The defect threshold.
nonDefectThreshold	The non-defect threshold.
moduleIdx	Specified model index in the model package.

Returns

See StatusCode for details.

---

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

• APIs/InferEngine.cs