CrossValidator

sklears_model_selection

Trait CrossValidator 

Source 
pub trait CrossValidator: Send + Sync {
    // Required methods
    fn n_splits(&self) -> usize;
    fn split(
        &self,
        n_samples: usize,
        y: Option<&Array1<i32>>,
    ) -> Vec<(Vec<usize>, Vec<usize>)>;
}
Expand description

Trait for cross-validation iterators

Required Methods§

Source

fn n_splits(&self) -> usize

Returns the number of splits

Source

fn split( &self, n_samples: usize, y: Option<&Array1<i32>>, ) -> Vec<(Vec<usize>, Vec<usize>)>

Generate train/test indices for cross-validation

For cross-validators that don’t need y (like KFold), pass None. For stratified cross-validators, y should contain integer class labels.

Implementors§