Struct LeaveOneOut 

pub struct LeaveOneOut;

Leave-one-out cross-validation (LOO).

Each sample serves as the sole test point exactly once, with every other sample forming the training set — equivalent to KFold with n_splits == n_samples. Included for API completeness.

Cost warning: LOO fits the model n_samples times. On anything but small datasets this is very expensive; prefer a modest KFold (5 or 10) unless you specifically need LOO’s near-unbiased (but high-variance) estimate.

use model_selection_rs::splitters::{CvSplitter, LeaveOneOut};

let loo = LeaveOneOut;
let splits = loo.split(4).unwrap();
assert_eq!(splits.len(), 4);
assert_eq!(splits[0], (vec![1, 2, 3], vec![0]));

Trait Implementations

impl Clone for LeaveOneOut

fn clone(&self) -> LeaveOneOut

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copy for LeaveOneOut

impl CvSplitter for LeaveOneOut

fn split(&self, n_samples: usize) -> Result<Vec<(Vec<usize>, Vec<usize>)>>

Produce every (train_indices, test_indices) pair for n_samples rows.

fn n_splits(&self) -> usize

Number of splits this strategy yields.

impl Debug for LeaveOneOut

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

Formats the value using the given formatter.

impl Default for LeaveOneOut

fn default() -> LeaveOneOut

Returns the “default value” for a type.