skglm.penalties.LogSumPenalty

class skglm.penalties.LogSumPenalty(alpha, eps)

Log sum penalty.

The penalty value reads

`"value"(w) = sum_(j=1)^(n_"features") log(1 + abs(w_j) / epsilon)`

__init__(alpha, eps)

Methods

__init__(alpha, eps)
derivative(w)	Compute the element-wise derivative.
generalized_support(w)	Return a mask which is True for coefficients in the generalized support.
get_spec()	Specify the numba types of the class attributes.
is_penalized(n_features)	Return a binary mask with the penalized features.
params_to_dict()	Get the parameters to initialize an instance of the class.
prox_1d(value, stepsize, j)	Compute the proximal operator of the log-sum penalty.
subdiff_distance(w, grad, ws)	Compute distance of negative gradient to the subdifferential at w.
value(w)	Compute the value of the log-sum penalty at w.