LassoLars

• Modelo Lasso ajustado con el algoritmo LARS.

• Para la estimación de los parámetros se minimiza la función objetivo:

\min_w \frac{1}{2n} ||Xw -y||_2^2 + \alpha * ||w||_1

from sklearn.datasets import load_diabetes

X, y = load_diabetes(return_X_y=True)

from sklearn.linear_model import LassoLars

lassoLars = LassoLars(
    # --------------------------------------------------------------------------
    # Constant that multiplies the L1 term, controlling regularization strength.
    # alpha must be a non-negative float i.e. in [0, inf). When alpha = 0, the
    # objective is equivalent to ordinary least squares, solved by the
    # LinearRegression object.
    alpha=1.0,
    # ---------------------------------------------------------------------
    # Whether to fit the intercept for this model.
    fit_intercept=True,
    # ---------------------------------------------------------------------
    # Maximum number of iterations.
    max_iter=1000,
    # ---------------------------------------------------------------------
    # If True the full path is stored in the coef_path_ attribute. If you
    # compute the solution for a large problem or many targets, setting
    # fit_path to False will lead to a speedup, especially with a small
    # alpha.
    fit_path=True,
    # --------------------------------------------------------------------------
    # When set to True, forces the coefficients to be positive.
    positive=False,
    # ---------------------------------------------------------------------
    # Upper bound on a uniform noise parameter to be added to the y values,
    # to satisfy the model’s assumption of one-at-a-time computations.
    # Might help with stability.
    jitter=None,
    # --------------------------------------------------------------------------
    # Determines random number generation for jittering. Pass an int for
    # reproducible output across multiple function calls. Ignored if jitter
    # is None.
    random_state=None,
)

lassoLars.fit(X, y)

lassoLars.score(X, y)

0.3573805394842746

lassoLars.coef_

array([  0.        ,   0.        , 367.70162582,   6.30970264,
         0.        ,   0.        ,   0.        ,   0.        ,
       307.60214746,   0.        ])

lassoLars.intercept_

152.133484162896