Package com.tibco.patterns.learn.rlink
Creation, loading and saving of RLink models; model training and evaluation
using individual feature vectors. This package contains the classes needed
to create the RLink JNI library and the confidence classes that use information
about internal constants used in RLink JNI.
The RLink class performs communication with rlink_jni library using JNI. It can manage several models at the same time.
The RLinkOut class is a prediction result that is returned from the model.
The Conf... classes define the concrete kinds of confidence measures that can be specified when creating PredictOptions class and used to get confidence for model predictions.
-
Interface Summary Interface Description Confidence Defines a confidence measure and its options. -
Class Summary Class Description ConfDefault Settings for the default model confidence measure.ConfFeature Feature based confidence measure.ConfNone Turns off calculation of a confidence measure.ConfRatio Weighted true/false ratio confidence measure.ConfRatioLabel Label based confidence measure.ConfStats Confidence measure based on submodel statistics.PredictOptions Holds all options for calculating prediction using a Learn model.RLink Performs communication with rlink_jni library using JNI to create, load, save, train and evaluate RLink models.RLinkOut Stores the result of RLink model prediction.ThermoCsvDump Dumps thermometers to CSV files -
Enum Summary Enum Description RLink.SubsetTrainMode Defines methods used to generate training examples for subsets of given training exampleRLink.ThermometerType Type of thermometers used in RLink model.