Struct RFECV 

pub struct RFECV<F> { /* private fields */ }

Recursive Feature Elimination with Cross-Validation.

Like RFE, but uses cross-validation scores to determine the optimal number of features. The user supplies a vector of per-feature-count CV scores (e.g., from running RFE with different n_features_to_select values), and RFECV picks the number that maximises the score.

Examples

use ferrolearn_preprocess::rfe::RFECV;
use ferrolearn_core::traits::Transform;
use ndarray::array;

let importances = array![0.5, 0.3, 0.2];
// CV scores for selecting 1, 2, 3 features:
let cv_scores = vec![0.85, 0.95, 0.90];
let rfecv = RFECV::<f64>::new(&importances, &cv_scores, 1).unwrap();
// Best is 2 features (score 0.95)
assert_eq!(rfecv.n_features_selected(), 2);
let x = array![[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]];
let out = rfecv.transform(&x).unwrap();
assert_eq!(out.ncols(), 2);

Implementations

impl<F: Float + Send + Sync + 'static> RFECV<F>

pub fn new( importances: &Array1<F>, cv_scores: &[f64], step: usize, ) -> Result<Self, FerroError>

Create a new RFECV from pre-computed importances and CV scores.

Parameters

• importances — per-feature importance scores.
• cv_scores — CV score for each possible number of features (index 0 = 1 feature, index 1 = 2 features, …).
• step — features removed per iteration.

Errors

• FerroError::InvalidParameter if inputs are empty or mismatched.

pub fn cv_scores(&self) -> &[f64]

Return the CV scores.

pub fn optimal_n_features(&self) -> usize

Return the optimal number of features.

pub fn n_features_selected(&self) -> usize

Return the number of selected features.

pub fn ranking(&self) -> &[usize]

Return the feature ranking.

pub fn support(&self) -> &[bool]

Return the boolean support mask.

pub fn selected_indices(&self) -> &[usize]

Return the indices of the selected features.

Trait Implementations

impl<F: Clone> Clone for RFECV<F>

fn clone(&self) -> RFECV<F>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<F: Debug> Debug for RFECV<F>

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

Formats the value using the given formatter.

impl<F: Float + Send + Sync + 'static> Transform<ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>> for RFECV<F>

fn transform(&self, x: &Array2<F>) -> Result<Array2<F>, FerroError>

Return a matrix containing only the optimally selected features.

Errors

Returns FerroError::ShapeMismatch if column count does not match.

type Output = ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>

The transformed output type.

type Error = FerroError

The error type returned by transform.