mlr 2.17.0.9000 Unreleased

filters - bugfixes

  • Fixed an issue which caused the random forest minimal depth filter to only return NA values when using thresholding. NAs should only be returned for features below the given threshold. (@annette987, #2710)
  • Fixed problem which prevented passing filter options via argument more.args for simple filters (@annette987, #2709)

mlr 2.17.0.9001 Unreleased

mlr 2.17.0.9002 Unreleased

  • Bugfix: tuneThreshold() now accounts for the direction of the measure. Beforehand, the performance measure was always minimized (#2732).
  • pkgdown: Move changelog to Appendix
  • Account for {checkmate} v2.0.0 update (#2734)

mlr 2.17.0.9003 Unreleased

  • remove measure clValid::dunn and its tests (package orphaned) (#2742)
  • Refactor function calls from packages (<pkg::fun>) within ParamSets (#2730) to avoid errors in listLearners() if those pkgs are not installed
  • remove regr_slim learner due to pkg (flare) being orphaned on CRAN

mlr 2.18.0.9000 Unreleased

  • Internal changes.

mlr 2.18.0.9001 Unreleased

  • Warning if fix.factors.prediction = TRUE causes the generation of NAs for new factor levels in prediction (@jakob-r, #2794)
  • Clear error message if prediction of wrapped learner has not the same length as newdata (@jakob-r, #2794)

mlr 2.18.0.9002 Unreleased

  • Switch to testthat v3 (#2796)
  • Enable parallel tests (#2796)
  • replace package PMCMR by PMCMRplus (#2796)
  • Remove CoxBoost learner due to CRAN removal
  • Silence warnings in tests

mlr 2.18.0.9003 Unreleased

  • Add filter FSelectoRcpp::relief(). This C++ based implementation of the RelieF filter algorithm is way faster than the Java based one from the {FSelector} package (#2804)
  • Fix S3 print method for FilterWrapper objects
  • Make ibrier measure work with survival tasks (#2789)