org.apache.spark.mllib.tree.impurity
Class for calculating the Gini impurity (http://en.wikipedia.org/wiki/Decision_tree_learning#Gini_impurity) during multiclass classification.
:: DeveloperApi :: variance calculation
number of instances
sum of labels
summation of squares of the labels
information value, or 0 if count = 0
:: DeveloperApi :: information calculation for multiclass classification
Array[Double] with counts for each label
sum of counts for all labels
information value, or 0 if totalCount = 0
Get this impurity instance.
Get this impurity instance. This is useful for passing impurity parameters to a Strategy in Java.
Class for calculating the Gini impurity (http://en.wikipedia.org/wiki/Decision_tree_learning#Gini_impurity) during multiclass classification.