integer node id, from 1
predicted value at the node
current node impurity
whether the node is a leaf
split to calculate left and right nodes
left child
right child
information gain stats
integer node id, from 1
current node impurity
whether the node is a leaf
left child
predict value if node is not leaf
predict value if node is not leaf
feature value
predicted value
predicted value at the node
right child
split to calculate left and right nodes
information gain stats
build the left node and right nodes if not leaf
build the left node and right nodes if not leaf
array of nodes
(Since version 1.2.0) build should no longer be used since trees are constructed on-the-fly in training
:: DeveloperApi :: Node in a decision tree.
About node indexing: Nodes are indexed from 1. Node 1 is the root; nodes 2, 3 are the left, right children. Node index 0 is not used.