CreateSvmSgdClassifier

Create a linear SVM classifier using stochastic gradient descent for training. The initial reference count is set to one if a new model is created and kept, otherwise. This SVM classifier is only applicable to binary classification problems. It learns a separating hyperplane between a class with label -1 and a class with label 1. These class labels are predefined. For training, any positive class labels are mapped to 1 and any negative class labels are mapped to -1. Models of this type neither support on-line training (sample by sample) nor retraining.

Syntax

Definition:

HRESULT CreateSvmSgdClassifier(
    HRESULT                         hrPrev,
    ITcVnMlModel*&                  ipMlModel,
    ETcVnSvmSgdClassifierType       eType = SSCT_ASGD,
    ETcVnSvmSgdClassifierMarginType eMarginType = SSCMT_SOFT_MARGIN,
    float                           fMarginRegularization = 0.00001f,
    float                           fInitialStepSize = 0.05f,
    float                           fStepDecreasingPower = 0.75f,
    ULONG                           nMaxIterations = 0,
    double                          fEpsilon = 0.0
)

Parameters

Name	Type	Default	Description
hrPrev	HRESULT		HRESULT indicating the result of previous operations (If SUCCEEDED(hrPrev) equals false, no operation is executed.)
ipMlModel	ITcVnMlModel*&		Returns the created model (Non-zero interface pointers are reused.)
eType	ETcVnSvmSgdClassifierType	SSCT_ASGD	Learning algorithm type (default: TCVN_SSCT_ASGD)
eMarginType	ETcVnSvmSgdClassifierMarginType	SSCMT_SOFT_MARGIN	Margin type (default: TCVN_SSCMT_SOFT_MARGIN)
fMarginRegularization	float	0.00001f	Margin regularization parameter (default: 0.00001)
fInitialStepSize	float	0.05f	Initial step size (default: 0.05)
fStepDecreasingPower	float	0.75f	Power parameter (default: 0.75)
nMaxIterations	ULONG	0	Maximum number of iterations (disabled if it equals 0 and fEpsilon is different from 0.0; triggers the usage of the default value of 100000 if nMaxIterations and fEpsilon equal 0)
fEpsilon	double	0.0	Maximum allowed difference of the error between two successive iterations (disabled if it equals 0.0 and nMaxIterations is different from 0; triggers the usage of the default value of 0.00001 if nMaxIterations and fEpsilon equal 0)

Return value

HRESULT

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