Struct MultiTaskLasso 

pub struct MultiTaskLasso<F> {
    pub alpha: F,
    pub fit_intercept: bool,
    pub max_iter: usize,
    pub tol: F,
}

Multi-task Lasso regression (joint multi-output L2,1-regularized least squares).

Fits all target columns jointly under a group-Lasso (L2,1) penalty, so each feature is either active for all tasks or zero for all of them. Mirrors sklearn.linear_model.MultiTaskLasso (_coordinate_descent.py:2663).

Type Parameters

• F: The floating-point type (f32 or f64).

Fields

alpha: F

Regularization strength on the L2,1 penalty. Larger values specify stronger regularization and zero out more whole feature rows.

fit_intercept: bool

Whether to fit a per-task intercept (bias) term.

max_iter: usize

Maximum number of block-coordinate-descent iterations.

tol: F

Convergence tolerance on the relative coefficient change / dual gap.

Implementations

impl<F: Float> MultiTaskLasso<F>

pub fn new() -> Self

Create a new MultiTaskLasso with default settings.

Defaults: alpha = 1.0, fit_intercept = true, max_iter = 1000, tol = 1e-4 — mirroring sklearn’s ctor defaults MultiTaskLasso(alpha=1.0, fit_intercept=True, max_iter=1000, tol=1e-4) (_coordinate_descent.py:2663).

pub fn with_alpha(self, alpha: F) -> Self

Set the regularization strength.

pub fn with_fit_intercept(self, fit_intercept: bool) -> Self

Set whether to fit per-task intercept terms.

pub fn with_max_iter(self, max_iter: usize) -> Self

Set the maximum number of iterations.

pub fn with_tol(self, tol: F) -> Self

Set the convergence tolerance.

Trait Implementations

impl<F: Clone> Clone for MultiTaskLasso<F>

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

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<F: Debug> Debug for MultiTaskLasso<F>

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

Formats the value using the given formatter.

impl<F: Float> Default for MultiTaskLasso<F>

fn default() -> Self

Returns the “default value” for a type.

impl<F: Float + Send + Sync + ScalarOperand + FromPrimitive + 'static> Fit<ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>, ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>> for MultiTaskLasso<F>

fn fit( &self, x: &Array2<F>, y: &Array2<F>, ) -> Result<FittedMultiTaskLasso<F>, FerroError>

Fit the multi-task Lasso model using block coordinate descent.

Ports sklearn’s enet_coordinate_descent_multi_task (_cd_fast.pyx:740) with l2_reg = 0 and l1_reg = alpha * n_samples.

Errors

Returns FerroError::ShapeMismatch if Y.nrows() differs from the number of samples in X. Returns FerroError::InvalidParameter if alpha is negative. Returns FerroError::InsufficientSamples if there are no samples. Returns FerroError::NumericalInstability if a required float constant or column mean cannot be formed.

type Fitted = FittedMultiTaskLasso<F>

The fitted model type returned by fit.

type Error = FerroError

The error type returned by fit.