F_VN_CreateRTreesModelExp2
Create a random trees model of the specified type. The initial reference count is set to one if a new model is created and kept, otherwise. Models of this type neither support on-line training (sample by sample) nor retraining. Predictions can only be scalar. (expert function)
Syntax
Definition:
FUNCTION F_VN_CreateRTreesModelExp2 : HRESULT
VAR_INPUT
ipMlModel : Reference To ITcVnMlModel;
eRTreesType : ETcVnRTrees;
nMaxDepth : UDINT;
nMinSamples : UDINT;
nActiveVariables : UDINT;
nMaxIterations : UDINT;
fEpsilon : LREAL;
fRegressionAccuracy : REAL;
ipClassPriors : ITcVnContainer;
hrPrev : HRESULT;
END_VAR
Inputs
|
Name |
Type |
Description |
|---|---|---|
|
ipMlModel |
Reference To ITcVnMlModel |
Returns the created model (Non-zero interface pointers are reused.) |
|
eRTreesType |
Random trees model type | |
|
nMaxDepth |
UDINT |
Maximum tree depth (default: 5) |
|
nMinSamples |
UDINT |
Minimum number of samples within a node required for splitting (default: 10) |
|
nActiveVariables |
UDINT |
Number of variables considered for splitting (0 means sqrt(total number of variables); default: 0) |
|
nMaxIterations |
UDINT |
Maximum number of iterations (disabled if it equals 0 and fEpsilon is different from 0.0; triggers the usage of the default value of 50 if nMaxIterations and fEpsilon equal 0) |
|
fEpsilon |
LREAL |
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 0.1 if nMaxIterations and fEpsilon equal 0) |
|
fRegressionAccuracy |
REAL |
Termination criterion for regressors (only for regressors; set to default if not allowed; default: 0.0) |
|
ipClassPriors |
Class priors (ContainerType_Vector_REAL or ContainerType_Vector_LREAL; only for classifiers; optional, set to 0 if not required or not allowed; default: 0) | |
|
hrPrev |
HRESULT indicating the result of previous operations (If SUCCEEDED(hrPrev) equals false, no operation is executed.) |
Return valueFurther information
The function F_VN_CreateRTReesModelExp2 is an expert variant of F_VN_CreateRTreesModel. It contains additional parameters.
Parameter
Model
The created model is returned in the interface pointer ipMlModel.
Model type
eRTreesType specifies whether the Random Forest model is used for classification (TCVN_RT_CLASSIFIER) or for regression (TCVN_RT_REGRESSOR).
Maximum tree depth
nMaxDepth indicates the maximum number of decision levels in a tree.
Minimum number of samples within a node
nMinSamples is the minimum number of samples that must remain in a node during tree formation. Splits that would fall below this number are not carried out.
Splitting features
nActiveVariables is the number of features used to split the tree. With 0, sqrt(N) features are used with the number N of all available features.
Maximum iterations
A maximum of as many iterations as specified in nMaxIterations are used for the optimization. If the value is 0, the respective default value is used.
Termination limit
The optimization is aborted as soon as the error between two iterations does not change more than specified in fEpsilon. If the value is 0, the respective default value is used.
Regression accuracy
fRegressionAccuracy is the abort criterion for regression optimization. For a classification, the value must be set to 0.
Class-Priors
ipClassPriors is a container that can be used to specify the a priori probabilities of the individual classes. For a regression, the value should be set to 0.
Application
For example, an RTrees model for regression can be created like this:
hr := F_VN_CreateRTreesExp2(
ipMlModel := ipMlModel,
eRTreesType := TCVN_RT_REGRESSOR,
nMaxDepth := 8,
nMinSamples := 4,
nActiveVariables := 0,
nMaxIterations := 0,
fEpsilon := 0,
fRegressionAccuracy := 0.3,
ipClassPriors := 0,
hrPrev := hr);Required License
TC3 Machine Learning Realtime Inference