Struct DeterministicPredictionConfig 

pub struct DeterministicPredictionConfig {
    pub warmup_steps: usize,
    pub history_window: usize,
    pub prediction_horizon: usize,
    pub history_decay: f32,
    pub residual_threshold: f32,
}

Configuration for deterministic gradient prediction.

Fields

warmup_steps: usize

Minimum steps of full training before prediction begins.

history_window: usize

Number of history steps to use for fitting.

prediction_horizon: usize

Steps to predict before correction.

history_decay: f32

Exponential decay for older history (1.0 = no decay).

residual_threshold: f32

Threshold for residual magnitude to trigger early correction.

Implementations

impl DeterministicPredictionConfig

pub const fn with_warmup_steps( self, steps: usize, ) -> DeterministicPredictionConfig

Builder: Set warmup steps.

pub const fn with_history_window( self, window: usize, ) -> DeterministicPredictionConfig

Builder: Set history window.

pub const fn with_prediction_horizon( self, horizon: usize, ) -> DeterministicPredictionConfig

Builder: Set prediction horizon.

pub const fn with_history_decay( self, decay: f32, ) -> DeterministicPredictionConfig

Builder: Set history decay.

Trait Implementations

impl Clone for DeterministicPredictionConfig

fn clone(&self) -> DeterministicPredictionConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for DeterministicPredictionConfig

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

Formats the value using the given formatter.

impl Default for DeterministicPredictionConfig

fn default() -> DeterministicPredictionConfig

Returns the “default value” for a type.