1 # Buildsheet autogenerated by ravenadm tool -- Do not edit.
7 SDESC[std]= Generalized Linear Models for Lasso, etc
8 HOMEPAGE= https://glmnet.stanford.edu
9 CONTACT= CRAN_Automaton[cran@ironwolf.systems]
12 SITES[main]= CRAN/src/contrib
13 https://loki.dragonflybsd.org/cranfiles/
14 DISTFILE[1]= glmnet_4.1-8.tar.gz:main
19 OPTIONS_AVAILABLE= none
20 OPTIONS_STANDARD= none
22 BUILDRUN_DEPENDS= R-foreach:single:std
25 R-RcppEigen:single:std
27 USES= cran gmake gettext:build
33 INSTALL_REQ_TOOLCHAIN= yes
35 [FILE:1104:descriptions/desc.single]
36 glmnet: Lasso and Elastic-Net Regularized Generalized Linear Models
38 Extremely efficient procedures for fitting the entire lasso or elastic-net
39 regularization path for linear regression, logistic and multinomial
40 regression models, Poisson regression, Cox model, multiple-response
41 Gaussian, and the grouped multinomial regression; see <<a
42 href="https://doi.org/10.18637%2Fjss.v033.i01"
43 target="_top">doi:10.18637/jss.v033.i01</a>> and <<a
44 href="https://doi.org/10.18637%2Fjss.v039.i05"
45 target="_top">doi:10.18637/jss.v039.i05</a>>. There are two new and
46 important additions. The family argument can be a GLM family object, which
47 opens the door to any programmed family (<<a
48 href="https://doi.org/10.18637%2Fjss.v106.i01"
49 target="_top">doi:10.18637/jss.v106.i01</a>>). This comes with a modest
50 computational cost, so when the built-in families suffice, they should be
51 used instead. The other novelty is the relax option, which refits each of
52 the active sets in the path unpenalized. The algorithm uses cyclical
53 coordinate descent in a path-wise fashion, as described in the papers
58 1ddbe5ce07076d1bdf58b0202ebd0ceac8eeb4796c5175681adb9e58c30ddcfe 2439515 CRAN/glmnet_4.1-8.tar.gz