R version 4.6.0 beta (2026-04-16 r89899 ucrt) -- "Because it was There"
Copyright (C) 2026 The R Foundation for Statistical Computing
Platform: x86_64-w64-mingw32/x64

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.

> library(testthat)
> library(sarb)

Attaching package: 'sarb'

The following object is masked from 'package:testthat':

    compare

> 
> test_check("sarb")

-- Model Comparison ------------------------------------------------------------

m1: RMSE = 0.3836, MAE = 0.2705
m2: RMSE = 0.8264, MAE = 0.5681


-- Pairwise Wilcoxon Signed-Rank Tests --

m1 vs m2: p = 0 (better: m1)

-- Which Ensemble Method? ------------------------------------------------------

v Recommended: sarb() \u2014 Signal-Adaptive Residual Boosting
Best average rank across 907 datasets (rank 3.38 of 8).
Use: sarb(y ~ ., data = df)

Runners-up:
2. boost_trees(method = 'xgboost') \u2014 rank 4.13
3. boost_trees(method = 'catboost') \u2014 rank 4.21

-- Which Ensemble Method? ------------------------------------------------------

v Recommended: boost_trees(method = 'lightgbm')
Fastest boosting implementation (histogram-based, C++).

Alternatives:
\u2022 boost_trees(method = 'xgboost') \u2014 also fast, more features
\u2022 forest_trees(method = 'extratrees') \u2014 fastest forest (no threshold
search)
[ FAIL 0 | WARN 0 | SKIP 0 | PASS 87 ]
> 
> proc.time()
   user  system elapsed 
   9.96    0.70   10.65