Treelite
Classes | Public Member Functions | Public Attributes | List of all members
Tree< ThresholdType, LeafOutputType >::Node Struct Reference

tree node More...

#include <tree.h>

Collaboration diagram for Tree< ThresholdType, LeafOutputType >::Node:
Collaboration graph
[legend]

Classes

union  Info
 store either leaf value or decision threshold More...
 

Public Member Functions

void Init ()
 Initialization method. Use this in lieu of constructor (POD types cannot have a non-trivial constructor)
 

Public Attributes

int32_t cleft_
 pointer to left and right children
 
int32_t cright_
 
uint32_t sindex_
 feature index used for the split highest bit indicates default direction for missing values
 
Info info_
 storage for leaf value or decision threshold
 
uint64_t data_count_
 number of data points whose traversal paths include this node. LightGBM models natively store this statistics.
 
double sum_hess_
 sum of hessian values for all data points whose traversal paths include this node. This value is generally correlated positively with the data count. XGBoost models natively store this statistics.
 
double gain_
 change in loss that is attributed to a particular split
 
SplitFeatureType split_type_
 feature split type
 
Operator cmp_
 operator to use for expression of form [fval] OP [threshold]. If the expression evaluates to true, take the left child; otherwise, take the right child.
 
bool data_count_present_
 whether data_count_ field is present
 
bool sum_hess_present_
 whether sum_hess_ field is present
 
bool gain_present_
 whether gain_present_ field is present
 
bool categories_list_right_child_
 

Detailed Description

template<typename ThresholdType, typename LeafOutputType>
struct treelite::Tree< ThresholdType, LeafOutputType >::Node

tree node

Definition at line 200 of file tree.h.


The documentation for this struct was generated from the following files: