kumulant
com.eignex.kumulant/stat/regression/tree/ClassCountsResult

ClassCountsResult

@Serializable
@SerialName(value = "ClassCountsResult")
data class ClassCountsResult(val numClasses: Int, val counts: DoubleArray) : Result(source)

Snapshot of a K-bin weighted class-count vector. Used as the leaf aggregate for ClassificationTree: the per-leaf running tally of how many (weighted) observations of each class landed in this leaf. Exposes derived class probabilities and the argmax prediction.

Constructors

ClassCountsResult

constructor(numClasses: Int, counts: DoubleArray)(source)

Properties

counts

val counts: DoubleArray(source)

Per-class accumulated weight; length numClasses.

entropy

val entropy: Double(source)

Shannon entropy of the empirical class distribution, in nats. Zero on an empty leaf.

gini

val gini: Double(source)

Gini impurity 1 - Sum p_k^2. Zero on a pure leaf or an empty leaf.

numClasses

val numClasses: Int(source)

Number of classes.

totalWeights

val totalWeights: Double(source)

Total weight summed across all classes.

Functions

equals

open operator override fun equals(other: Any?): Boolean(source)

hashCode

open override fun hashCode(): Int(source)

predict

fun predict(): Int(source)

Argmax class index; ties resolve to the lowest index.

probabilities

fun probabilities(): DoubleArray(source)

Class probabilities; uniform when no observations have been seen yet.