Struct FittedMultiTaskLasso 

pub struct FittedMultiTaskLasso<F> { /* private fields */ }

Fitted multi-task Lasso regression model.

Stores the learned coefficient matrix (shape (n_tasks, n_features), the sklearn coef_ layout), the per-task intercept vector, and the number of block-coordinate-descent sweeps run. Implements Predict.

Implementations

impl<F: Float> FittedMultiTaskLasso<F>

pub fn coefficients(&self) -> &Array2<F>

Borrow the learned coefficient matrix, shape (n_tasks, n_features) (sklearn coef_ layout).

pub fn intercepts(&self) -> &Array1<F>

Borrow the per-task intercept vector, length n_tasks (sklearn intercept_).

pub fn n_iter(&self) -> usize

Number of block-coordinate-descent sweeps run by the solver (sklearn n_iter_).

pub fn dual_gap(&self) -> F

Duality gap at the returned solution, on the (1 / (2 * n_samples))-scaled objective.

Mirrors sklearn’s MultiTaskLasso.dual_gap_ attribute (_coordinate_descent.py:2636 — unpacked from enet_coordinate_descent_multi_task — then :2652 self.dual_gap_ /= n_samples, the final objective-scaling). This is the final block-CD duality gap (the value that decided convergence), scaled by 1 / n_samples so the exposed value matches sklearn’s dual_gap_.

Trait Implementations

impl<F: Clone> Clone for FittedMultiTaskLasso<F>

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

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<F: Debug> Debug for FittedMultiTaskLasso<F>

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

Formats the value using the given formatter.

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

fn predict(&self, x: &Array2<F>) -> Result<Array2<F>, FerroError>

Predict target values for the given feature matrix.

Computes pred[i, k] = sum_j X[i, j] * coef[k, j] + intercept[k], i.e. X.dot(coef^T) + intercepts (broadcast per task column), returning an (n_samples, n_tasks) array.

Errors

Returns FerroError::ShapeMismatch if the number of features does not match the fitted model.

type Output = ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>

The prediction output type (e.g., ndarray::Array1<F> or ndarray::Array1<usize>).

type Error = FerroError

The error type returned by predict.