NEWS.md

In optRF: Optimising Random Forest Stability by Determining the Optimal Number of Trees

optRF 1.2.1

Improvements

• Improved output of estimate_numtrees and estimate_stability functions

Bug Fixes

• Fixed a bug with the plot functions for the opt_importance_object and opt_prediction_object objects.

optRF 1.2.0

New Features

• Added function measure_stability to measure the stability of random forest for a certain data set with a certain number of trees.

Improvements

• Adding checks for validity of parameters alpha, number_repetitions and rec_thresh.

Other

• Updated CITATION file with published manuscript.

optRF 1.1.0

New Features

• Added support for selecting multiple classes in opt_prediction.
• Introduced the verbose parameter in opt_importance and opt_prediction to control message display.
• Added the importance parameter to opt_importance, allowing users to specify the importance measure used by ranger.
• Included code for generating the SNPdata data set.
• Added vignettes to guide users on using the package.

Improvements

• Replaced the evaluation of random forest using simulated data with out-of-bag (OOB) data.
• Removed and simplified redundant code to enhance efficiency.
• Reduced minimum version requirements for imported packages.

Bug Fixes

• Fixed a bug in the calculation of estimate_numtrees.

Other

• Added a NEWS.md file to track changes to the package.
• Added a CITATION file.
• Added a README.md file for the Github repository.

optRF 1.0.1

• Initial release on CRAN

optRF documentation built on June 8, 2025, 10:24 a.m.