Struct SelectKBest 

pub struct SelectKBest {
    pub k: usize,
    pub scoring_fn: ScoringFunction,
}

Parameters for SelectKBest feature selector (unfitted state).

Selects the top-k features according to a pluggable ScoringFunction. This is more flexible than MutualInformationSelector, which is hard-coded to mutual information scoring.

Example

use anofox_ml_preprocessing::SelectKBest;
use anofox_ml_preprocessing::select_k_best::ScoringFunction;
use anofox_ml_core::Transform;
use ndarray::array;

// Feature 0 perfectly separates the two classes; feature 1 is noise.
let x = array![
    [0.0, 0.5],
    [0.0, 0.8],
    [1.0, 0.3],
    [1.0, 0.7],
];
let y = array![0.0, 0.0, 1.0, 1.0];

let selector = SelectKBest::new(1, ScoringFunction::FClassif);
let fitted = selector.fit(&x, &y).unwrap();
let x_selected = fitted.transform(&x).unwrap();

assert_eq!(x_selected.ncols(), 1);

Fields

k: usize

Number of top features to select.

scoring_fn: ScoringFunction

Scoring function to rank features.

Implementations

impl SelectKBest

pub fn new(k: usize, scoring_fn: ScoringFunction) -> Self

Create a new SelectKBest selector that keeps the top k features ranked by the given scoring function.

pub fn fit<F: Float>( &self, x: &Array2<F>, y: &Array1<F>, ) -> Result<FittedSelectKBest<F>>

Fit the selector on the given data.

For ScoringFunction::FClassif and ScoringFunction::FRegression, y is used as the target variable. For ScoringFunction::Variance, y is ignored.

Trait Implementations

impl Clone for SelectKBest

fn clone(&self) -> SelectKBest

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for SelectKBest

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for SelectKBest

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for SelectKBest

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.