Speed optimizations for iregnet
- Mentors
- Toby Dylan Hocking, Anuj Khare
- Organization
- R project for statistical computing
The iregnet package is the first R package to support four types of censoring and elastic net (L1 + L2) regularization. Though it is already useful for making predictions in data sets with possibly censored observations, the current code is too slow for large data sets. After this GSOC project, its model fitting code will be even faster. The iregnet package will get better performance in large data sets.