Struct RFE 

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

Recursive Feature Elimination.

Starting from all features, repeatedly removes the step least-important features until n_features_to_select features remain. The ranking is determined by the importance vector supplied at construction.

Examples

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

// Feature importances: feature 0 is most important, feature 2 least
let importances = array![0.6, 0.3, 0.1];
let rfe = RFE::<f64>::new(&importances, 1, 1).unwrap();
assert_eq!(rfe.support(), &[true, false, false]);
assert_eq!(rfe.ranking(), &[1, 2, 3]);
let x = array![[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]];
let out = rfe.transform(&x).unwrap();
assert_eq!(out.ncols(), 1);

Implementations

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

pub fn new( importances: &Array1<F>, n_features_to_select: usize, step: usize, ) -> Result<Self, FerroError>

Create a new RFE from pre-computed feature importances.

Parameters

• importances — per-feature importance scores (higher = more important).
• n_features_to_select — number of features to keep.
• step — number of features to remove per iteration.

Errors

• FerroError::InvalidParameter if importances is empty, step is zero, or n_features_to_select exceeds the number of features.

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

Return the feature ranking (1 = best, higher = eliminated earlier).

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

Return the boolean support mask.

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

Return the indices of the selected features.

pub fn n_features_selected(&self) -> usize

Return the number of selected features.

Trait Implementations

impl<F: Clone> Clone for RFE<F>

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

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<F: Debug> Debug for RFE<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 RFE<F>

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

Return a matrix containing only the selected features.

Errors

Returns FerroError::ShapeMismatch if the number of columns differs from the number of features used at construction.

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

The transformed output type.

type Error = FerroError

The error type returned by transform.