discrim
discrim settled into a thin engine shim after handing its model definitions to parsnip.
Comparison · Analytics
bonsai vs themis
A side-by-side editorial comparison of bonsai and themis — release velocity, themes, recent moves, and the top alternatives to consider.
Shared themes:tidymodels
bonsai vs themis: at a glance
What is bonsai?
bonsai keeps widening tidymodels' boosted-tree engine bench, catboost most recently
bonsai exists to attach non-core tree engines to parsnip's boost_tree() and rand_forest(), and the release history reads as a steady accumulation of them: partykit, aorsf, lightgbm, and now catboost. The 0.4.x line is spent making catboost behave like a full tidymodels citizen rather than adding anything new.
What is themis?
themis is back to adding real resampling algorithms after a documentation-heavy stretch.
themis supplies recipes steps for handling class imbalance in tidymodels. The 1.0.x line was consumed by documentation accuracy, message translation and internal consistency work. Version 1.1.0 returns to substance with two new under-sampling methods.
bonsai vs themis: editorial side-by-side
bonsai
0.0ANALYTICS
bonsai keeps widening tidymodels' boosted-tree engine bench, catboost most recently
◆ Current state
bonsai exists to attach non-core tree engines to parsnip's boost_tree() and rand_forest(), and the release history reads as a steady accumulation of them: partykit, aorsf, lightgbm, and now catboost. The 0.4.x line is spent making catboost behave like a full tidymodels citizen rather than adding anything new.
◆ Where it's heading
Each engine follows the same arc — land it, then close the gaps that keep it from tuning cleanly (parameter naming, multi_predict, threading, case weights). Recent work is squarely in that second phase for catboost, with dials supplying the matching parameter objects on its own release schedule. Bug-fix density is high relative to new surface.
◆ Prediction
Expect the catboost integration to keep filling in tuning and GPU-related arguments before any further engine is added; the entries give no signal about which engine would come next.
themis
2.5ANALYTICS
themis is back to adding real resampling algorithms after a documentation-heavy stretch.
◆ Current state
themis supplies recipes steps for handling class imbalance in tidymodels. The 1.0.x line was consumed by documentation accuracy, message translation and internal consistency work. Version 1.1.0 returns to substance with two new under-sampling methods.
◆ Where it's heading
The package grows by adding algorithms rather than restructuring itself. tomek() was rewritten to handle multiple classes and drop the unbalanced dependency, case weights arrived at 1.0.0, and cluster-centroid and condensed-nearest-neighbour under-sampling arrive now — each shipped as both a recipes step and a direct-implementation function.
◆ Prediction
Expect further under- and over-sampling methods in the same paired form, as the package fills out coverage of the standard class-imbalance literature.
Alternatives to bonsai and themis
Other Analytics products tracked by Sparkpulse, ranked by recent ship velocity. Each card links to a full editorial trajectory and lets you pivot into a head-to-head comparison with either bonsai or themis.
dbparser
dbparser shed its database and CSV writers to become just a DrugBank parser.
desirability2
desirability2 is making multi-metric model selection a first-class tidymodels step.
crosstalk
crosstalk is frozen infrastructure: four releases in five years, mostly CRAN upkeep.
cmdstanr
cmdstanr keeps adding fast approximations beside full HMC, and fighting Windows toolchains.
comtradr
comtradr's 1.0 line is a long tail of patches against a brittle UN trade API.
See all bonsai alternatives → · See all themis alternatives →
Recent activity from bonsai and themis
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
- 10d agothemisthemis 1.1.0 adds cluster-centroid and CNN under-sampling
- 2mo agobonsaicatboost gains multi_predict() and corrected tuning parameters
- 1y agobonsaicatboost engine added to boost_tree()
- 1y agobonsailightgbm accepts sparse matrices for fit and predict
- 1y agothemisthemis 1.0.3 corrects resampling direction in documentation
- 2y agobonsaiaorsf fit failure in multisession workers fixed
- 2y agobonsaiaorsf engine added; lightgbm gains dataset params and case weights
- 2y agothemisthemis 1.0.2 makes internal consistency and speed changes
- 3y agothemisthemis 1.0.1 fixes upsampling errors when none is needed
- 3y agobonsailightgbm num_leaves becomes tunable; alias arguments disallowed
- 4y agothemisthemis 1.0.0 adds case weights to up- and down-sampling
- 4y agothemisthemis 0.2.2 rewrites tomek() for multiclass, drops a dependency
Frequently asked questions
What is the difference between bonsai and themis?
Both compete on the same themes — tidymodels — within Analytics. themis is currently shipping more aggressively (velocity 2.5 vs 0.0), with 0 editorial sparks in the last 30 days against 0. See the at-a-glance table above for a side-by-side breakdown of velocity, recent sparks, and editorial themes.
Is bonsai better than themis?
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. themis is currently shipping more aggressively (velocity 2.5 vs 0.0), with 0 editorial sparks in the last 30 days against 0. For your specific use case, the alternatives sections above list other Analytics products to evaluate alongside.
What are the best alternatives to bonsai?
Top bonsai alternatives in Analytics are ranked by recent ship velocity. Browse the "bonsai alternatives" section above for the current picks, or visit /alternatives/bonsai-r for the full list with editorial commentary on each.
What are the best alternatives to themis?
Top themis alternatives in Analytics are ranked by recent ship velocity. Browse the "themis alternatives" section above for the current picks, or visit /alternatives/themis for the full list with editorial commentary on each.