secure_learning.models.secure_lasso module

Implementation of LASSO regression model.

class secure_learning.models.secure_lasso.Lasso(solver_type=SolverTypes.GD, alpha=1)

Bases: Linear

Solver for LASSO regression. Optimizes a model with objective function $$frac{1}{2n_{textrm{samples}}} times ||y - X_times_w||^2_2 + alpha_1 ||w||_1$$

__init__(solver_type=SolverTypes.GD, alpha=1)

Constructor method.

Parameters:

• solver_type (SolverTypes) – Solver type to use (e.g. Gradient Descent aka GD)

• alpha (float) – Regularization parameter

name = 'LASSO regression'