Struct EmbeddedMethods 

pub struct EmbeddedMethods;

Embedded feature selection methods

Implementations

impl EmbeddedMethods

pub fn lasso_selection( features: &Array2<f64>, target: &Array1<f64>, alpha: f64, max_iterations: usize, ) -> Result<FeatureSelectionResult>

LASSO-based feature selection

Uses LASSO regularization to select features.

Arguments

• features - Feature matrix (samples x features)
• target - Target variable
• alpha - Regularization parameter
• max_iterations - Maximum iterations for optimization

Returns

• Feature selection result

Example

use scirs2_core::ndarray::{Array1, Array2};
use scirs2_series::feature_selection::EmbeddedMethods;

let features = Array2::from_shape_vec((100, 10), (0..1000).map(|x| x as f64).collect()).expect("Operation failed");
let target = Array1::from_vec((0..100).map(|x| x as f64).collect());

let result = EmbeddedMethods::lasso_selection(&features, &target, 1.0, 1000).expect("Operation failed");
println!("Selected {} features", result.selected_features.len());

pub fn ridge_selection( features: &Array2<f64>, target: &Array1<f64>, alpha: f64, n_features: Option<usize>, ) -> Result<FeatureSelectionResult>

Ridge regression-based feature ranking

Uses Ridge regularization to rank features by importance.

Arguments

• features - Feature matrix (samples x features)
• target - Target variable
• alpha - Regularization parameter
• n_features - Number of top features to select

Returns

• Feature selection result

pub fn tree_based_selection( features: &Array2<f64>, target: &Array1<f64>, n_features: Option<usize>, ) -> Result<FeatureSelectionResult>

Tree-based feature importance (simplified decision tree)

Estimates feature importance using a simplified decision tree approach.

Arguments

• features - Feature matrix (samples x features)
• target - Target variable
• n_features - Number of top features to select

Returns

• Feature selection result