NEWS

---

recforest 1.0.2 (2026-01-13)

• Added id to globalVariables() to fix compatibility with upcoming dplyr release
• Fixed invalid file URI in README.md

recforest 1.0.1

• Updated test expectations following upstream correction to readmission dataset in frailtypack: 3 subjects previously had inconsistent death/event coding which has now been fixed.

recforest 1.0.0 (2024-12-05)

• Filling documentation for train_forest() and predict()
• Vignettes filled

recforest 0.2.0

• Adding 3 vignettes
• Setting pkgdown parameters
• Filling README
• train_forest() does not used X and Y anymore but data, id_var, covariates, event, time_vars, death_var to define data-related parameters

recforest 0.1.0

• Refactoring train_forest()
  • n_boostrap is set to NULL by default. If NULL, the number of bootstrap samples is set to 2/3 the number of individuals in the dataset.
  • params is not a function parameter anymore, each parameter must be explicitly provided in the function call

recforest 0.0.2

• Implementing s3 methods for "summary", "print" and "predict" ("plot" waiting to be implemented)
• Adding unit tests (code coverage: 94.21%)
• Adding documentation (README.md)
• Setting continuous integration on Gitlab CI
• Adding a demo dataset "bladder1_recforest"
• Setting pkgdown

recforest 0.0.1

• Refactoring + golden master