Changelog - literanger

0.1.1 - 2024-09-13

Performance improved. R interface and C++ core have been separated.

Changed

• Set Depends to R >= 3.6.0.
• Training speed was increased by ~25% by reducing memory allocations 19e7c475 and inlining access to data. 233063e0.
• Source code underwent a minor re-organisation to separate the R-specific components from the C++ library.

Added

• DataVector class to read, write, and pass data without R.

0.1.0 - 2024-09-03

New feature! literanger can now serialize trained random forests using cereal.

The project has been moved to GitLab: https://gitlab.com/stephematician/literanger.

Changed

• The value-type returned by predict now matches the response type in training ea67c83e
• Bump cpp11 to 0.4.7.

Added

• Functions read_literanger and write_literanger for serialization.

Fixed

• Fixed bug in implementation of always-selected candidates for splitting, e.g. the names_of_always_draw argument 6d31d7f3
• Minor performance tweak 9a3b639a in particular for ‘maxstat’ 37580d9b

0.0.2 - 2023-07-11

Update to pass CRAN’s ASAN check

Changed

• Improve performance of node splitting (d3f6424)

Added

• Add re-entrant log gamma to speed up beta splitting rule (d7f058d)
• Minor fixes to documentation (91b6c6d, 0f62d02)

Fixed

• Fix potential illegal access and incorrect unweighted sampling without replacement (b6df5d9)

0.0.1 - 2023-06-25

First release

A refactoring and adaptation of the ranger package https://github.com/imbs-hl/ranger for random forests. Has faster prediction mode intended for embedding into the multiple imputation algorithm proposed by Doove et al in:

Doove, L. L., Van Buuren, S., & Dusseldorp, E. (2014). Recursive partitioning for missing data imputation in the presence of interaction effects. Computational statistics & data analysis, 72, 92-104.

Added

• Fit classification and regression trees
• Prediction via most frequent value or mean
• Get predictions as terminal node identifiers in each tree or as a random draw from inbag values in a random tree