Struct TimeSeriesSplit 

pub struct TimeSeriesSplit { /* private fields */ }

Time Series Split for temporal data Provides train/test indices to split time series data while preserving temporal order

Implementations

impl TimeSeriesSplit

pub fn new( n_samples: usize, n_splits: usize, max_train_size: Option<usize>, test_size: Option<usize>, gap: usize, ) -> Result<Self>

Create a new Time Series Split

Arguments

• n_samples - Total number of samples
• n_splits - Number of splits (must be at least 2)
• max_train_size - Maximum size of training set (None for no limit)
• test_size - Fixed test set size (None for equal splits)
• gap - Number of samples to exclude between train and test

pub fn n_splits(&self) -> usize

Get the number of splits

pub fn get_fold(&self, fold: usize) -> Result<(Vec<usize>, Vec<usize>)>

Get train and test indices for a specific fold

pub fn split( &self, features: &Array2<f64>, labels: &Array1<usize>, fold: usize, ) -> Result<(Array2<f64>, Array1<usize>, Array2<f64>, Array1<usize>)>

Split features and labels for a specific fold

pub fn split_regression( &self, features: &Array2<f64>, labels: &Array1<f64>, fold: usize, ) -> Result<(Array2<f64>, Array1<f64>, Array2<f64>, Array1<f64>)>

Split regression features and labels for a specific fold

Trait Implementations

impl Clone for TimeSeriesSplit

fn clone(&self) -> TimeSeriesSplit

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for TimeSeriesSplit

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

Formats the value using the given formatter.