NEWS.md

In PingYangChen/milr: Multiple-Instance Logistic Regression with LASSO Penalty

rfda v0.3.0

• Remove dependency: assertthat, purrr
• Performance tuning with rewriting some functions with RcppParallel.
• Fix the NULL return in predict.milr and predict.softmax.
• Add some tests for milr.
• Add vignette based on the published R Journal.

rfda v0.2.0

• New options for milr: numLambda, tolerance.
• Add option for output the labels of bags/instances into fitted and predict functions.
• Use the coefficients of ridge regression acquired by glmnet as initial values for p > n problems.

rfda v0.1.0

• First release on CRAN.

PingYangChen/milr documentation built on Oct. 30, 2020, 2:18 a.m.