Changes in version 5.0 o Fixed clang-13 issue with OpenMP. o Changed HTTP URLs to use HTTPS. Changes in version 4.0 o Internal PRNG switched to PCG32, with D. Lemire method for used for unbiased sampling. Hence, same random seed does not reproduce the same results as pre-4.0 versions. One can also expect slight speed regression. o Migrated to GitLab. o Fixed error in NA checking. Thanks to Patrick Schratz for spotting this. Changes in version 3.0 o Parallel training with OpenMP. o New organisation of the internal PRNG. o Error propagation reporting removed (arguments reportErrorEvery and saveErrorPropagation) since it brings too little value for the complexity it requires. And can be emulated with merging anyway. Changes in version 2.0.3 (2017-11-15) o Registration of native routines. o Documentation updates. Changes in version 2.0.2 (2016-07-22) o Added a wrapper for performing feature selection using shadow importance. Changes in version 2.0.1 (2016-05-20) o Fixed bug in merge.rFerns code, causing merged importance scores to be wrong, especially after merging very small ensembles. o Fixed memory protection bug triggering rare segmentation faults in the R's GC, often long after the rFerns model was completed. Changes in version 2.0.0 (2015-12-03) o Shadow importance. o Roxygen documentation. o Model merging for parallel, batch and deeply traceable training. o Fixed a minor error in leaf score calculation (basically redundant to the model performance). Thanks to Mateusz Fedoryszak for spotting this. o Fixed OOB predictions returned by predict(), which are now always data.frame, even in the multi-label case. Changes in version 1.1.0 (2014-11-30) o Code portability fixes. o predict method returns OOB results for no input data even when fern forest is not saved. o Fixed minor importance calculation bug. Changes in version 1.0.0 (2014-11-11) o Published in JSS. Changes in version 0.3.3 (2014-03-24) o Fixed uninitialised memory bug causing problems with LLVM. Changes in version 0.3.2 (2014-02-18) o Prototype of a multi-label classification. o Removed redundant constant from scores definition; score values will be different, but the importance scores and prediction results will be the same as generated by previous versions. Changes in version 0.3.1 (2013-04-23) o predict method for rFerns returns OOB votes/scores when no predictors are given. Changes in version 0.3 (2012-01-18) o Handling of unbalanced classes. o Performance tweaks. Changes in version 0.2 (2012-01-02) o Added support for integer and ordered-factor attributes, suggested by Bill Venables. Changes in version 0.1 (2011-12-30) o First public release.