ML.GLMnet: ML.GLMnet

In frbl/OnlineSuperLearner: Online SuperLearner package

Description

Wrapper for a glmnet estimator. From their discription: Extremely efficient procedures for fitting the entire lasso or elastic-net regularization path for linear regression, logistic and multinomial regression models, Poisson regression and the Cox model. Two recent additions are the multiple-response Gaussian, and the grouped multinomial regression. The algorithm uses cyclical coordinate descent in a path-wise fashion, as described in http://www.jstatsoft.org/v33/i01.

Usage

ML.GLMnet

Format

An object of class R6ClassGenerator of length 24.

Methods

initialize(family = "binomial", alpha = 0.5, nlambda = 100, verbose = FALSE)

Creates a new GLMnet model

@param family string (default = "binomial") the family to use for the estimator. Generally this should be binomial (since we are running various logistic regressions). However, if you have a good reason to use a different specification, change it here.

@param alpha double (default = 0.5) the elasticnet mixing parameter, with $0 <= alpha <= 1$. $alpha=1$ is the lasso penalty, and $alpha=0$ the ridge penalty.

@param nlambda integer (default = 100) the number of lambda values.

@param verbose (default = FALSE) the verbosity of the fitting procedure

get_validity

Active method. Checks whether the specified configuration is valid.

@return boolean, true if it is a valid configuration.

frbl/OnlineSuperLearner documentation built on Feb. 9, 2020, 9:28 p.m.