Struct ModelTreeNode 

pub struct ModelTreeNode {
    pub feature: Option<usize>,
    pub threshold: Float,
    pub left: Option<Box<ModelTreeNode>>,
    pub right: Option<Box<ModelTreeNode>>,
    pub coefficients: Option<Array1<Float>>,
    pub intercept: Option<Float>,
    pub n_samples: usize,
    pub std_dev: Float,
}

Node in a Model Tree

Fields

feature: Option<usize>

Feature index for splitting (None for leaf nodes)

threshold: Float

Split threshold value

left: Option<Box<ModelTreeNode>>

Left child node (samples <= threshold)

right: Option<Box<ModelTreeNode>>

Right child node (samples > threshold)

coefficients: Option<Array1<Float>>

Linear model coefficients for leaf node

intercept: Option<Float>

Intercept for linear model in leaf

n_samples: usize

Number of samples in this node

std_dev: Float

Standard deviation of targets in this node

Implementations

impl ModelTreeNode

pub fn new_leaf( coefficients: Array1<Float>, intercept: Float, n_samples: usize, std_dev: Float, ) -> Self

Create a new leaf node with a linear model

pub fn new_internal( feature: usize, threshold: Float, left: Self, right: Self, n_samples: usize, std_dev: Float, ) -> Self

Create a new internal split node

pub fn is_leaf(&self) -> bool

Check if this is a leaf node

pub fn predict_sample(&self, sample: &ArrayView1<'_, Float>) -> Float

Predict value for a single sample

Trait Implementations

impl Clone for ModelTreeNode

fn clone(&self) -> ModelTreeNode

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for ModelTreeNode

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.