Enum ScaleMode 

pub enum ScaleMode {
    Empirical,
    TreeChain,
}

Available on crate feature alloc only.

How DistributionalSGBT estimates uncertainty (σ).

• Empirical (default): tracks an EWMA of squared prediction errors. σ = sqrt(ewma_sq_err). Always calibrated, zero tuning, O(1) compute. Use this when σ drives learning-rate modulation (σ high → learn faster).

• TreeChain: trains a full second ensemble of Hoeffding trees to predict log(σ) from features (NGBoost-style dual chain). Gives feature-conditional uncertainty but requires strong signal in the scale gradients.

Variants

Empirical

EWMA of squared prediction errors — always calibrated, O(1).

TreeChain

Full Hoeffding-tree ensemble for feature-conditional log(σ) prediction.

Trait Implementations

impl Clone for ScaleMode

fn clone(&self) -> ScaleMode

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ScaleMode

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

Formats the value using the given formatter.

impl Default for ScaleMode

fn default() -> ScaleMode

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for ScaleMode

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl PartialEq for ScaleMode

fn eq(&self, other: &ScaleMode) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Serialize for ScaleMode

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

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl Copy for ScaleMode

impl Eq for ScaleMode

impl StructuralPartialEq for ScaleMode