Struct RegularizationPathResult 

pub struct RegularizationPathResult {
    pub lambdas: Array1<Float>,
    pub coef_path: Array2<Float>,
    pub intercept_path: Array1<Float>,
    pub cv_scores: Array1<Float>,
    pub cv_scores_std: Array1<Float>,
    pub n_nonzero: Array1<usize>,
    pub active_features: Vec<Vec<usize>>,
    pub best_lambda: Float,
    pub best_lambda_idx: usize,
    pub lambda_1se: Float,
    pub lambda_1se_idx: usize,
}

Results of regularization path computation

Fields

lambdas: Array1<Float>

Lambda values used

coef_path: Array2<Float>

Coefficient paths (n_lambdas × n_features)

intercept_path: Array1<Float>

Intercept paths (n_lambdas)

cv_scores: Array1<Float>

Cross-validation scores (n_lambdas)

cv_scores_std: Array1<Float>

Standard errors of CV scores (n_lambdas)

n_nonzero: Array1<usize>

Number of non-zero coefficients at each lambda

active_features: Vec<Vec<usize>>

Indices of selected features at each lambda

best_lambda: Float

Best lambda value (based on CV)

best_lambda_idx: usize

Index of best lambda

lambda_1se: Float

Lambda at 1 standard error rule

lambda_1se_idx: usize

Index of lambda at 1 standard error rule

Implementations

impl RegularizationPathResult

pub fn coef_at_lambda(&self, lambda: Float) -> Option<ArrayView1<'_, Float>>

Get coefficient at a specific lambda value

pub fn feature_path(&self, feature_idx: usize) -> Option<ArrayView1<'_, Float>>

Get the coefficient path for a specific feature

pub fn sparse_model_1se(&self) -> (Float, ArrayView1<'_, Float>)

Find the sparsest model within 1 standard error of the best

pub fn summary(&self) -> HashMap<String, Float>

Get summary statistics

Trait Implementations

impl Clone for RegularizationPathResult

fn clone(&self) -> RegularizationPathResult

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for RegularizationPathResult

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

Formats the value using the given formatter.