Struct VarianceThreshold

pub struct VarianceThreshold { /* private fields */ }

VarianceThreshold for removing low-variance features

Features with variance below the threshold are removed. This is useful for removing features that are mostly constant and don’t provide much information.

Implementations

impl VarianceThreshold

pub fn new(threshold: f64) -> Result<Self>

Creates a new VarianceThreshold selector

Arguments

• threshold - Features with variance below this threshold are removed (default: 0.0)

Returns

• A new VarianceThreshold instance

Examples

use scirs2_transform::selection::VarianceThreshold;

// Remove features with variance less than 0.1
let selector = VarianceThreshold::new(0.1);

pub fn with_defaults() -> Self

Creates a VarianceThreshold with default threshold (0.0)

This will only remove features that are completely constant.

pub fn fit<S>(&mut self, x: &ArrayBase<S, Ix2>) -> Result<()>

where S: Data, S::Elem: Float + NumCast,

Fits the VarianceThreshold to the input data

Arguments

• x - The input data, shape (n_samples, n_features)

Returns

• Result<()> - Ok if successful, Err otherwise

pub fn transform<S>(&self, x: &ArrayBase<S, Ix2>) -> Result<Array2<f64>>

where S: Data, S::Elem: Float + NumCast,

Transforms the input data by removing low-variance features

Arguments

• x - The input data, shape (n_samples, n_features)

Returns

• Result<Array2<f64>> - The transformed data with selected features only

pub fn fit_transform<S>(&mut self, x: &ArrayBase<S, Ix2>) -> Result<Array2<f64>>

where S: Data, S::Elem: Float + NumCast,

Fits the VarianceThreshold to the input data and transforms it

Arguments

• x - The input data, shape (n_samples, n_features)

Returns

• Result<Array2<f64>> - The transformed data with selected features only

pub fn variances(&self) -> Option<&Array1<f64>>

Returns the variances computed for each feature

Returns

• Option<&Array1<f64>> - The variances for each feature

pub fn get_support(&self) -> Option<&Vec<usize>>

Returns the indices of selected features

Returns

• Option<&Vec<usize>> - Indices of features that pass the variance threshold

pub fn get_support_mask(&self) -> Option<Array1<bool>>

Returns a boolean mask indicating which features are selected

Returns

• Option<Array1<bool>> - Boolean mask where true indicates selected features

pub fn n_features_selected(&self) -> Option<usize>

Returns the number of selected features

Returns

• Option<usize> - Number of features that pass the variance threshold

pub fn inverse_transform<S>( &self, _x: &ArrayBase<S, Ix2>, ) -> Result<Array2<f64>>

where S: Data, S::Elem: Float + NumCast,

Inverse transform - not applicable for feature selection

This method is not implemented for feature selection as it’s not possible to reconstruct removed features.