Struct CrossValidator 

pub struct CrossValidator {
    pub mode: CrossValidationMode,
    pub min_train_size: usize,
    pub horizon: usize,
    pub n_folds: usize,
}

Cross-validation configuration.

Fields

mode: CrossValidationMode

Mode: expanding or sliding window.

min_train_size: usize

Minimum number of training observations.

horizon: usize

Number of steps to predict (test window).

n_folds: usize

Number of folds.

Implementations

impl CrossValidator

pub fn new( mode: CrossValidationMode, min_train_size: usize, horizon: usize, n_folds: usize, ) -> Self

Create a new cross-validator.

pub fn evaluate<F>( &self, timestamps: &[i64], values: &[f64], model_factory: F, ) -> Result<CrossValidationResult, ForecastError>

where F: Fn() -> Box<dyn ForecastModel>,

Evaluate a forecast model using walk-forward cross-validation.

model_factory is called once per fold to create a fresh model instance — this avoids carry-over state between folds.

Errors

Returns an error if the data is too short for the requested configuration or if any fold’s fit/predict fails.

Trait Implementations

impl Clone for CrossValidator

fn clone(&self) -> CrossValidator

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for CrossValidator

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

Formats the value using the given formatter.