decisiontree

Implements C4.5 decision trees for integral labels and attributes.

Main class to use is org.allenai.nlpstack.parse.poly.decisiontree.DecisionTree. Use the companion object to build the tree. Then use ) or ) to do prediction.

Linear Supertypes

AnyRef, Any

case class DecisionTree(child: IndexedSeq[Seq[(Int, Int)]], splittingAttribute: IndexedSeq[Option[Int]], categoryCounts: IndexedSeq[Seq[(Int, Int)]]) extends Product with Serializable

Immutable decision tree for integer-valued features and categories.
case class DenseVector(label: Option[Int], attributes: IndexedSeq[Int]) extends FeatureVector with Product with Serializable

Instance with arbitrary integral attributes
sealed trait FeatureVector extends AnyRef

A feature vector with integral features and label.
case class FeatureVectors(featureVectors: IndexedSeq[FeatureVector]) extends Product with Serializable

FeatureVectors is a convenience container for feature vectors.
case class SparseVector(label: Option[Int], numAttributes: Int, trueAttributes: Set[Int]) extends FeatureVector with Product with Serializable

Instance with sparse binary attributes