features.msgl: Nonzero features

In msgl: Multinomial Sparse Group Lasso

Description

Extracts the nonzero features for each model.

Usage

## S3 method for class 'msgl'
features(object, ...)

Arguments

object	a msgl object
...	ignored

Value

a list of of length nmod(x) containing the nonzero features (that is nonzero colums of the beta matrices)

Author(s)

Martin Vincent

Examples

data(SimData)


lambda <- msgl::lambda(x, classes, alpha = .5, d = 50, lambda.min = 0.05)
fit <- msgl::fit(x, classes, alpha = .5, lambda = lambda)

# the nonzero features of model 1, 10 and 25
features(fit)[c(1,10,25)]

# count the number of nonzero features in each model
sapply(features(fit), length)

msgl documentation built on May 8, 2019, 9:03 a.m.