org.apache.spark.mllib.tree.impurity
Interface Impurity

All Superinterfaces:

java.io.Serializable

All Known Implementing Classes:

Entropy, Gini, Variance

public interface Impurity
extends scala.Serializable

:: Experimental :: Trait for calculating information gain. This trait is used for (a) setting the impurity parameter in Strategy (b) calculating impurity values from sufficient statistics.

Method Summary

double	calculate(double[] counts, double totalCount) :: DeveloperApi :: information calculation for multiclass classification
double	calculate(double count, double sum, double sumSquares) :: DeveloperApi :: information calculation for regression

Method Detail

calculate

double calculate(double[] counts,
                 double totalCount)

:: DeveloperApi :: information calculation for multiclass classification

Parameters:

counts - Array[Double] with counts for each label

totalCount - sum of counts for all labels

Returns:

information value, or 0 if totalCount = 0

calculate

double calculate(double count,
                 double sum,
                 double sumSquares)

:: DeveloperApi :: information calculation for regression

Parameters:

count - number of instances

sum - sum of labels

sumSquares - summation of squares of the labels

Returns:

information value, or 0 if count = 0