Struct TrainConfig 

pub struct TrainConfig {

    pub algorithm: TrainAlgorithm,
    pub task: Task,
    pub tree_type: TreeType,
    pub criterion: Criterion,
    pub max_depth: Option<usize>,
    pub min_samples_split: Option<usize>,
    pub min_samples_leaf: Option<usize>,
    pub physical_cores: Option<usize>,
    pub n_trees: Option<usize>,
    pub max_features: MaxFeatures,
    pub seed: Option<u64>,
    pub compute_oob: bool,
    pub learning_rate: Option<f64>,
    pub bootstrap: bool,
    pub top_gradient_fraction: Option<f64>,
    pub other_gradient_fraction: Option<f64>,
}

Unified training configuration shared by the Rust and Python entry points.

The crate keeps one normalized config type so the binding layer only has to perform input validation and type conversion; all semantic decisions happen from this one structure downward.

Fields

algorithm: TrainAlgorithm

High-level training family.

task: Task

Regression or classification.

tree_type: TreeType

Tree learner used by the selected algorithm family.

criterion: Criterion

Split criterion. Criterion::Auto is resolved by the trainer.

max_depth: Option<usize>

Maximum tree depth.

min_samples_split: Option<usize>

Smallest node size that is still allowed to split.

min_samples_leaf: Option<usize>

Minimum child size after a split.

physical_cores: Option<usize>

Optional cap on training-side rayon threads.

n_trees: Option<usize>

Number of trees for ensemble algorithms.

max_features: MaxFeatures

Feature subsampling strategy.

seed: Option<u64>

Seed used for reproducible sampling and randomized splits.

compute_oob: bool

Whether random forests should compute out-of-bag metrics.

learning_rate: Option<f64>

Gradient boosting shrinkage factor.

bootstrap: bool

Whether gradient boosting should bootstrap rows before gradient sampling.

top_gradient_fraction: Option<f64>

Fraction of largest-gradient rows always kept by GOSS sampling.

other_gradient_fraction: Option<f64>

Fraction of the remaining rows randomly retained by GOSS sampling.

Trait Implementations

impl Clone for TrainConfig

fn clone(&self) -> TrainConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for TrainConfig

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

Formats the value using the given formatter.

impl Default for TrainConfig

fn default() -> Self

Returns the “default value” for a type.

impl Copy for TrainConfig