F_VN_CreateNbcModel
Create a normal Bayes classifier of the specified type. The initial reference count is set to one if a new model is created and kept, otherwise. In order to train normal Bayes classifiers, a sufficiently high number of samples is required for each class. It depends on the number of features and the distribution of the data. Hence, it needs to be tested for each application. Models of this type do not support on-line training (sample by sample). For the retraining of such classifier models, the set of presented classes must be identical to the previous learning steps. Otherwise, an exception is raised.
Syntax
Definition:
FUNCTION F_VN_CreateNbcModel : HRESULT
VAR_INPUT
ipMlModel : Reference To ITcVnMlModel;
eNbcType : ETcVnNbc;
hrPrev : HRESULT;
END_VAR
Inputs
Name |
Type |
Description |
---|---|---|
ipMlModel |
Reference To ITcVnMlModel |
Returns the created model (Non-zero interface pointers are reused.) |
eNbcType |
Normal Bayes classifier type | |
hrPrev |
HRESULT indicating the result of previous operations (If SUCCEEDED(hrPrev) equals false, no operation is executed.) |
Further information
The function F_VN_CreateNbcModel
creates a Normal Bayes Classifier (NBC) model.
Normal Bayes Classifier models
The Normal Bayes Classifier creates a mixture model in which each class is represented by a Gaussian distribution. The parameters of these Gaussian functions are determined during training.
Parameter
Model
The created model is returned in the interface pointer ipMlModel
.
Model type
eNbcType
specifies whether NBC is used for classification (TCVN_NBC_CLASSIFIER
) or for anomaly detection (TCVN_NBC_NOVELTY_DETECTOR
).
Application
For example, an NBC model for classification can be created like this:
hr := F_VN_CreateNbcModel(
ipMlModel := ipMlModel,
eNbcType := TCVN_NBC_CLASSIFIER,
hrPrev := hr);
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 |