Type Members

1.  class FeaturePowerSetTerms extends AnyRef

   Weighted combinatorial factors corresponding to the power-set of features that are included

   Weighted combinatorial factors corresponding to the power-set of features that are included

   The set is extended as new features are encountered. Each element in weightBySubsetSize corresponds to all of the subsets of the features that are included vs excluded of a given size:

   • the 0th element corresponds to the term where all of the encountered features are excluded,
   • the 1st element sums the terms that have exactly one of the encountered feature included,
   • the 2nd element sums the terms that have exactly two of the encountered features included, and so on and so forth. The final element represents the single term where all of the features are "turned on".

   The calculation is performed efficiently using a dynamic programming technique that runs in quadratic time with the number of features. It is described in the extend and unwind methods.

2.  case class FeatureWeightFactor(weightWhenExcluded: Double, weightWhenIncluded: Double) extends Product with Serializable

   Description of a feature's effect on the weight ascribed to the node

   Description of a feature's effect on the weight ascribed to the node

   Decision trees can be thought of as weighted sums over their leaves. When there is knowledge of every feature, all of the weight goes into a single leaf. However, when feature are excluded from the prediction, then both paths of decisions that depend on those excluded features are assigned non-zero weights that are proportional to the share of the training data that followed that path.

   This class captures those weights for a single feature and a single node. The weightWhenIncluded gives the factor of the weight of the node due to this feature when the feature is included (i.e. known). It is always 0 or 1. The weightWhenExcluded gives the factor of the weight of the node due to this feature when the feature is excluded (i.e. unknown). It is always strictly greater than 0 and strictly less than 1.

   weightWhenExcluded

   factor of the weight of the leaf when the feature is excluded

   weightWhenIncluded

   factor of the weight of the leaf when the feature is included

3.  class InternalModelNode[T <: PredictionResult[Any]] extends ModelNode[T]

   Internal node in the decision tree

   Internal node in the decision tree

   T

   type of the output

4.  class ModelLeaf[T] extends ModelNode[PredictionResult[T]]
5.  trait ModelNode[+T <: PredictionResult[Any]] extends Serializable
6.  class TrainingLeaf[T] extends TrainingNode[AnyVal, T]

   Average the training data to make a leaf prediction

7.  abstract class TrainingNode[T <: AnyVal, S] extends Serializable

   Class to provide getNode interface for internal and leaf training nodes Created by maxhutch on 11/29/16.

   Class to provide getNode interface for internal and leaf training nodes Created by maxhutch on 11/29/16.

   T

   type of the input vector

   S

   type of the model output

8.  case class TreeMeta(depth: Int) extends Product with Serializable

   Created by maxhutch on 12/21/16.