RMVMR
RMVMR is being tidied in lockstep with MVMR, the package it wraps
Comparison · Analytics
filtro vs treeshap
A side-by-side editorial comparison of filtro and treeshap — release velocity, themes, recent moves, and the top alternatives to consider.
Shared themes:r package
filtro vs treeshap: at a glance
What is filtro?
filtro moves to S7 and multiplies its feature-scoring methods in a single release.
filtro supplies feature-selection filter scores for the tidymodels stack. Version 0.2.0 adds five scoring methods — correlation, random forest importance, information gain, ROC AUC and cross tabulation — and moves the package from S3 to S7. It also gains a ranking layer: show_best_score_* and rank_best_score_* helpers for a single score, plus desirability-function helpers for optimising across several scores at once.
What is treeshap?
treeshap keeps widening its tree-model coverage while the SHAP math stays put.
treeshap computes exact SHAP values for tree ensembles in R, reaching each modelling framework through a per-framework unify() adapter. Since returning to CRAN in 2023 it has added GPBoost, ranger survival forests and multi-output models to that adapter layer. Four releases in three years, each dominated by adapter work contributed by users of one specific framework.
filtro vs treeshap: editorial side-by-side
filtro
0.0ANALYTICS
filtro moves to S7 and multiplies its feature-scoring methods in a single release.
◆ Current state
filtro supplies feature-selection filter scores for the tidymodels stack. Version 0.2.0 adds five scoring methods — correlation, random forest importance, information gain, ROC AUC and cross tabulation — and moves the package from S3 to S7. It also gains a ranking layer: show_best_score_* and rank_best_score_* helpers for a single score, plus desirability-function helpers for optimising across several scores at once.
◆ Where it's heading
The package is being built out on two axes at once — the catalogue of scores, and the machinery for choosing between them. The desirability functions are the more telling half, since they assume users will filter on several criteria rather than one. Adopting S7 while still pre-1.0 suggests the object model is being settled before the API is frozen.
◆ Prediction
Expect more scoring methods on the same S7 interface and a 1.0 release once the score and ranking APIs stop moving; the ranking helpers' naming is the most likely thing to change first.
treeshap
0.0ANALYTICS
treeshap keeps widening its tree-model coverage while the SHAP math stays put.
◆ Current state
treeshap computes exact SHAP values for tree ensembles in R, reaching each modelling framework through a per-framework unify() adapter. Since returning to CRAN in 2023 it has added GPBoost, ranger survival forests and multi-output models to that adapter layer. Four releases in three years, each dominated by adapter work contributed by users of one specific framework.
◆ Where it's heading
The direction is breadth of model support rather than new explanation methods: every release since the first CRAN submission adds or repairs a unify() backend. Maintenance is community-driven, with named contributors fixing the framework they personally use. Nothing in these entries points at work on the SHAP algorithms themselves.
◆ Prediction
Expect the next release to add or repair another unify() adapter as a contributor brings their own framework, rather than to change how explanations are computed.
Alternatives to filtro and treeshap
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 filtro or treeshap.
geoarrow-r
geoarrow tracks the GeoArrow spec and otherwise just keeps compiling
n2khab
n2khab keeps retracting interpretations of habitat data it can't actually support
tidypolars
tidypolars is grinding toward complete dplyr coverage, one supported function at a time
OneSampleMR
OneSampleMR found that argument order in a formula was silently changing its estimates
bpbounds
bpbounds found the same swapped-cell bug twice and clamped its bounds back into range
See all filtro alternatives → · See all treeshap alternatives →
Recent activity from filtro and treeshap
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
- 3mo agotreeshapGPBoost support lands; xgboost adapter repaired
- 11mo agofiltroFive new filter scores and the move to S7 ⚡
- 11mo agofiltroDevelopment build: score transformation handling
- 2y agotreeshapFixes broken lightgbm.unify examples
- 2y agotreeshapMulti-output model explanations added
- 2y agotreeshapFirst CRAN release consolidates the unify() adapters ⚡
Frequently asked questions
What is the difference between filtro and treeshap?
Both compete on the same themes — r package — within Analytics. filtro and treeshap are shipping at a similar cadence (velocity 0.0 vs 0.0, both within Sparkpulse's "active" band). See the at-a-glance table above for a side-by-side breakdown of velocity, recent sparks, and editorial themes.
Is filtro better than treeshap?
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. filtro and treeshap are shipping at a similar cadence (velocity 0.0 vs 0.0, both within Sparkpulse's "active" band). For your specific use case, the alternatives sections above list other Analytics products to evaluate alongside.
What are the best alternatives to filtro?
Top filtro alternatives in Analytics are ranked by recent ship velocity. Browse the "filtro alternatives" section above for the current picks, or visit /alternatives/filtro-r for the full list with editorial commentary on each.
What are the best alternatives to treeshap?
Top treeshap alternatives in Analytics are ranked by recent ship velocity. Browse the "treeshap alternatives" section above for the current picks, or visit /alternatives/treeshap for the full list with editorial commentary on each.