edu.berkeley.guir.lib.satin.recognizer.rubine
Class RubineClassifier
java.lang.Object
|
+--edu.berkeley.guir.lib.gesture.Classifier
|
+--edu.berkeley.guir.lib.satin.recognizer.rubine.RubineClassifier
- public class RubineClassifier
- extends Classifier
Fields inherited from class edu.berkeley.guir.lib.gesture.Classifier |
ccm, ccmInv, dotCategory, enabledCategories, EPSILON, featureClasses, featuresUsed, gestureSet, meanFeatureValues, MISRECOGNIZED_PROP, myCollectionListener, myPropChangeListener, propChangeSupport, trained, TRAINED_PROP, weights |
Methods inherited from class edu.berkeley.guir.lib.gesture.Classifier |
addPropertyChangeListener, addPropertyChangeListener, categoryDistance, categoryDistance, checkForInterrupt, classify, classifyWithoutTraining, computeCovarianceMatrix, distanceToCategory, distanceToCategory, dump, dumpMFV, dumpRelativeVariance, findPrincipleComponent, findPrincipleFeature, fixClassifier, getDefaultFeatureClasses, getDistancesByFeature, getFeature, getFeature, getFeatureClasses, getFeaturesUsed, getGestureSet, getNormalizedDistancesByFeature, getNormalizedDistancesByFeature, getNormalizedDistancesByFeature, getTrainingCategories, getWeights, isTrained, MahalanobisDistance, MahalanobisDistance, removePropertyChangeListener, removePropertyChangeListener, setFeatureClasses, setGestureSet, setTrained, testRecognition, testRecognition, testRecognition, testRecognition, train |
Methods inherited from class java.lang.Object |
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait |
DOT
public static final String DOT
RubineClassifier
public RubineClassifier()
RubineClassifier
public RubineClassifier(GestureSet gs)
- Does not automatically train.
classify_satin
protected SortedValueNumMap classify_satin(Gesture gesture)
throws Exception
- An internal version of classify for use with Satin.
I know that copying and pasting code is generally a bad practice, but
it will take a significant amount of work to unify and separate the two.
calculateAccuracy
protected double calculateAccuracy(double[] disc,
int j)
- Parameters:
disc
- is the array of dicriminants.j
- is the entry in disc whose accuracy we want