F_VN_GetSampleNovelty

Get the degree of novelty of a single sample.
Can use available TwinCAT Job Tasks for executing parallel code regions.

Syntax

Definition:

FUNCTION F_VN_GetSampleNovelty : HRESULT
VAR_INPUT
    ipNoveltyDetector : ITcVnMlModel;
    ipSample          : ITcUnknown;
END_VAR
VAR_IN_OUT
    fNovelty          : REAL;
END_VAR
VAR_INPUT
    hrPrev            : HRESULT;
END_VAR

Inputs

Name	Type	Description
ipNoveltyDetector	ITcVnMlModel	Novelty detector to be used
ipSample	ITcUnknown	Container holding a single input sample (ContainerType_Vector_REAL or ContainerType_Vector_LREAL)
hrPrev	HRESULT	HRESULT indicating the result of previous operations (If SUCCEEDED(hrPrev) equals false, no operation is executed.)

In/Outputs

Name	Type	Description
fNovelty	REAL	Returns the degree of novelty (0.0 if a sample is completely known; > 0.0 otherwise) of the presented sample

Return value

HRESULT

Further information

The function F_VN_GetSampleNovelty determines the degree of novelty for a sample in a container.

Parameter

Anomaly model

To calculate the degree of novelty, the previously trained model must be transferred to ipNoveltyDetector.

Sample

A container with a sample is transferred to ipSample

Degree of novelty

The degree of novelty is returned via the reference fNovelty. The degree of novelty indicates how much a sample differs from the learned data. A degree of novelty of 0 indicates that a sample is known to the model.

Application

The degree of novelty of a sample can be calculated as follows:

hr := F_VN_GetSampleNovelty(
    ipNoveltyDetector   := ipNoveltyDetector,
    ipSample            := ipSample,
    fNovelty            := fNovelty,
    hrPrev              := hr);

Related functions

F_VN_GetBatchNovelties

Required License

TC3 Machine Learning Realtime Inference

System Requirements

Development environment	Target platform	PLC libraries to include
TwinCAT V3.1.4024.59 or later	PC or CX (x64) with min. PL50, e.g. Intel 4-core Atom CPU	Tc3_Vision