r-cran-glmnet 3.0-2-1 (amd64 binary) in ubuntu focal

 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.
 The algorithm uses cyclical coordinate descent in a path-wise fashion, as
 described in the paper Introduction to Glmnet.