pulsatrix 1.1.0
Loading...
Searching...
No Matches
pulsatrix::SurrogateNode Struct Reference

One node. A leaf has feature -1. More...

#include <surrogate_tree.hpp>

Public Member Functions

bool is_leaf () const
 

Public Attributes

int64_t feature = -1
 
double threshold = 0.0
 Inputs with x[feature] <= threshold go to left.
 
int left = -1
 
int right = -1
 
int64_t samples = 0
 Training inputs that reached the node.
 
double impurity = 0.0
 Gini impurity, or the variance of the target.
 
double value = 0.0
 The mean target, or the majority class (lowest index on a tie).
 
std::vector< int64_t > class_counts
 Classification: how many training inputs of each class reached the node.
 

Detailed Description

One node. A leaf has feature -1.

Member Function Documentation

◆ is_leaf()

bool pulsatrix::SurrogateNode::is_leaf ( ) const
inline

Member Data Documentation

◆ class_counts

std::vector<int64_t> pulsatrix::SurrogateNode::class_counts

Classification: how many training inputs of each class reached the node.

◆ feature

int64_t pulsatrix::SurrogateNode::feature = -1

◆ impurity

double pulsatrix::SurrogateNode::impurity = 0.0

Gini impurity, or the variance of the target.

◆ left

int pulsatrix::SurrogateNode::left = -1

◆ right

int pulsatrix::SurrogateNode::right = -1

◆ samples

int64_t pulsatrix::SurrogateNode::samples = 0

Training inputs that reached the node.

◆ threshold

double pulsatrix::SurrogateNode::threshold = 0.0

Inputs with x[feature] <= threshold go to left.

◆ value

double pulsatrix::SurrogateNode::value = 0.0

The mean target, or the majority class (lowest index on a tie).


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