hex.genmodel.easy.prediction

Class AnomalyDetectionPrediction

java.lang.Object

hex.genmodel.easy.prediction.AbstractPrediction

hex.genmodel.easy.prediction.AnomalyDetectionPrediction

All Implemented Interfaces:

java.io.Serializable

public class AnomalyDetectionPrediction
extends AbstractPrediction

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
java.lang.Boolean	isAnomaly Only available when MojoModel has contamination parameter defined otherwise is null.
int[]	leafNodeAssignmentIds Ditto, available in MOJO 1.3 and newer
java.lang.String[]	leafNodeAssignments Only valid for tree-based models, null for all other mojo models.
double	normalizedScore Higher number means more anomalous point, smaller number means more normal point.
double	score The raw number that an algorithm is using to count final anomaly score.
double[]	stageProbabilities Staged predictions of tree algorithms (prediction probabilities of trees per iteration).

Constructor Summary

Constructors

Constructor and Description
AnomalyDetectionPrediction()
AnomalyDetectionPrediction(double[] preds)

Method Summary

Modifier and Type	Method and Description
double[]	toPreds()

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

isAnomaly

public java.lang.Boolean isAnomaly

Only available when MojoModel has contamination parameter defined otherwise is null.

score

public double score

The raw number that an algorithm is using to count final anomaly score. E.g. for Isolation Forest this number is mean path length of data in the trees. Smaller number means more anomalous point, higher number means more normal point.

normalizedScore

public double normalizedScore

Higher number means more anomalous point, smaller number means more normal point.

leafNodeAssignments

public java.lang.String[] leafNodeAssignments

Only valid for tree-based models, null for all other mojo models.

leafNodeAssignmentIds

public int[] leafNodeAssignmentIds

Ditto, available in MOJO 1.3 and newer

stageProbabilities

public double[] stageProbabilities

Staged predictions of tree algorithms (prediction probabilities of trees per iteration). The output structure is for tree Tt and class Cc: Binomial models: [probability T1.C1, probability T2.C1, ..., Tt.C1] where Tt.C1 correspond to the the probability p0 Multinomial models: [probability T1.C1, probability T1.C2, ..., Tt.Cc]

Constructor Detail

AnomalyDetectionPrediction

public AnomalyDetectionPrediction()

AnomalyDetectionPrediction

public AnomalyDetectionPrediction(double[] preds)

Method Detail

toPreds

public double[] toPreds()