F_VN_InverseFeatureTransform

Syntax

Definition:

FUNCTION F_VN_InverseFeatureTransform : HRESULT
VAR_INPUT
    ipFeatureTransform : ITcVnMlModel;
    ipSamples          : ITcUnknown;
    pipTransformed     : Pointer To  ITcUnknown;
    hrPrev             : HRESULT;
END_VAR

Inputs

Name	Type	Description
ipFeatureTransform	ITcVnMlModel	Feature transform instance to be used
ipSamples	ITcUnknown	Container holding one or more of input sample(s) (ContainerType_Vector_REAL, ContainerType_Vector_LREAL, ContainerType_Vector_Vector_REAL, or ContainerType_Vector_Vector_LREAL)
pipTransformed	Pointer To ITcUnknown	Returns the transformed sample(s) using the same type like ipSamples
hrPrev	HRESULT	HRESULT indicating the result of previous operations (If SUCCEEDED(hrPrev) equals false, no operation is executed.)

Return value

HRESULT

Further information

The function F_VN_InverseFeatureTransform inverts the feature transformation of one or more samples using a transformation model. The direct transformation can be carried out with the same model using the function F_VN_FeatureTransform.

Parameter

Transformation model

A previously created model must be transferred to ipFeatureTransform for feature transformation.

Samples

A container with the samples to be transformed is transferred to ipSamples. The following container types are permitted:

ContainerType_Vector_REAL or ContainerType_Vector_LREAL for a single sample
ContainerType_Vector_Vector_REAL or ContainerType_Vector_Vector_LREAL for a container with several samples.

Transformed samples

A container with the transformed samples is returned via the pointer pipTransformed. The container type is taken from ipSamples.

Application

For example, an inverse feature transformation is carried out as follows:

hr := F_VN_InverseFeatureTransform(
    ipFeatureTransform  := ipMlModel,
    ipSamples           := ipSamplesTransformed,
    pipTransformed      := ADR(ipTransformed),
    hrPrev              := hr);

Related functions

F_VN_FeatureTransform

Required License

TC3 Vision Machine Learning

System Requirements

Development environment	Target platform	PLC libraries to include
TwinCAT V3.1.4024.54 or later	PC or CX (x64) with PL50, e.g. Intel 4-core Atom CPU	Tc3_Vision