Package org.elasticsearch.client.ml.inference.trainedmodel.tree

Class Tree.Builder

java.lang.Object

org.elasticsearch.client.ml.inference.trainedmodel.tree.Tree.Builder

Enclosing class:

Tree

public static class Tree.Builder
extends java.lang.Object

Constructor Summary

Constructors

Constructor	Description
Builder()

Method Summary

Modifier and Type	Method	Description
TreeNode.Builder	addJunction(int nodeIndex, int featureIndex, boolean isDefaultLeft, double decisionThreshold)	Add a decision node.
Tree.Builder	addLeaf(int nodeIndex, double value)	Sets the node at nodeIndex to a leaf node.
Tree.Builder	addNode(TreeNode.Builder node)
Tree	build()
Tree.Builder	setClassificationLabels(java.util.List<java.lang.String> classificationLabels)
Tree.Builder	setFeatureNames(java.util.List<java.lang.String> featureNames)
Tree.Builder	setNodes(java.util.List<TreeNode.Builder> nodes)
Tree.Builder	setNodes(TreeNode.Builder... nodes)
Tree.Builder	setTargetType(TargetType targetType)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

Builder

public Builder()

Method Detail

setFeatureNames

public Tree.Builder setFeatureNames(java.util.List<java.lang.String> featureNames)

addNode

public Tree.Builder addNode(TreeNode.Builder node)

setNodes

public Tree.Builder setNodes(java.util.List<TreeNode.Builder> nodes)

setNodes

public Tree.Builder setNodes(TreeNode.Builder... nodes)

setTargetType

public Tree.Builder setTargetType(TargetType targetType)

setClassificationLabels

public Tree.Builder setClassificationLabels(java.util.List<java.lang.String> classificationLabels)

addJunction

public TreeNode.Builder addJunction(int nodeIndex,
                                    int featureIndex,
                                    boolean isDefaultLeft,
                                    double decisionThreshold)

Add a decision node. Space for the child nodes is allocated

Parameters:

nodeIndex - Where to place the node. This is either 0 (root) or an existing child node index

featureIndex - The feature index the decision is made on

isDefaultLeft - Default left branch if the feature is missing

decisionThreshold - The decision threshold

Returns:

The created node

addLeaf

public Tree.Builder addLeaf(int nodeIndex,
                            double value)

Sets the node at nodeIndex to a leaf node.

Parameters:

nodeIndex - The index as allocated by a call to addJunction(int, int, boolean, double)

value - The prediction value

Returns:

this

build

public Tree build()