Struct GradientBoostedTree 

pub struct GradientBoostedTree {
    pub schema_version: u32,
    pub trees: Vec<TreeNode>,
    pub base_score: f64,
    pub learning_rate: f64,
    pub feature_names: Vec<String>,
    pub quantile: Option<f64>,
    pub output_scale: f64,
    pub metadata: Option<Value>,
}

A gradient-boosted tree ensemble model.

Predictions are computed as: base_score + learning_rate * sum(tree_predictions). Models are JSON-serializable with schema versioning for forward compatibility.

Fields

schema_version: u32

Schema version for forward/backward compatibility.

trees: Vec<TreeNode>

The ensemble of decision trees.

base_score: f64

Initial prediction (bias / intercept).

learning_rate: f64

Learning rate (shrinkage factor applied to each tree).

feature_names: Vec<String>

Feature names in order matching feature_index.

quantile: Option<f64>

Target quantile: None for L2/mean, Some(q) for quantile regression.

output_scale: f64

Output scale factor applied to the final prediction.

metadata: Option<Value>

Optional metadata about the model.

Implementations

impl GradientBoostedTree

pub fn predict(&self, features: &[f64]) -> f64

Predict a single sample.

pub fn predict_batch(&self, feature_matrix: &[Vec<f64>]) -> Vec<f64>

Predict a batch of samples.

pub fn n_trees(&self) -> usize

Number of trees in the ensemble.

pub fn n_features(&self) -> usize

Number of features the model expects.

pub fn feature_importance(&self) -> HashMap<usize, f64>

Compute feature importance based on split frequency. Returns a map from feature index to normalized importance [0, 1].

pub fn feature_importance_named(&self) -> HashMap<String, f64>

Compute named feature importance.

pub fn to_json(&self) -> Result<String>

Serialize the model to JSON string.

pub fn from_json(json: &str) -> Result<Self>

Deserialize a model from JSON string.

pub fn to_bytes(&self) -> Result<Vec<u8>>

Serialize the model to bytes (UTF-8 JSON).

pub fn from_bytes(data: &[u8]) -> Result<Self>

Deserialize a model from bytes (UTF-8 JSON).

Trait Implementations

impl Clone for GradientBoostedTree

fn clone(&self) -> GradientBoostedTree

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for GradientBoostedTree

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for GradientBoostedTree

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for GradientBoostedTree

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.