mrbayes
mrbayes spent 2026 auditing its own Bayesian MR estimators for coding errors.
Comparison · Analytics
fastplyr vs kernelshap
A side-by-side editorial comparison of fastplyr and kernelshap — release velocity, themes, recent moves, and the top alternatives to consider.
fastplyr vs kernelshap: at a glance
What is fastplyr?
A fast dplyr stand-in that keeps finding new places to skip work entirely.
fastplyr reimplements the dplyr verbs on a faster backend, exposing f_summarise, f_mutate, f_reframe and a set of group metadata helpers alongside optimized joins and quantiles. The most recent release removes non-API C functions and raises the floor to R 4.5.0, a steep requirement that follows the C++17 requirement introduced a release earlier. The verb surface itself has been stable since 0.9.0.
What is kernelshap?
kernelshap makes permutation SHAP practical past eight features, then fixes the kernel weights it had wrong.
kernelshap computes model-agnostic SHAP values in R through Kernel SHAP, permutation SHAP and an exact additive explainer. Version 0.8.0 added a sampling permutation-SHAP algorithm with standard errors and early stopping, lifting the practical feature ceiling past what the exact method allows. Version 0.9.0 then corrected a bug in how kernel weights were computed — exact Kernel SHAP now agrees with exact permutation SHAP — and moved parallelism from foreach to doFuture.
fastplyr vs kernelshap: editorial side-by-side
fastplyr
0.0ANALYTICS
A fast dplyr stand-in that keeps finding new places to skip work entirely.
◆ Current state
fastplyr reimplements the dplyr verbs on a faster backend, exposing f_summarise, f_mutate, f_reframe and a set of group metadata helpers alongside optimized joins and quantiles. The most recent release removes non-API C functions and raises the floor to R 4.5.0, a steep requirement that follows the C++17 requirement introduced a release earlier. The verb surface itself has been stable since 0.9.0.
◆ Where it's heading
The optimization strategy has shifted from making individual functions fast to reasoning about expressions before evaluating them — 0.9.9 began marking simple operators as group-unaware so expressions built only from them are evaluated across the whole data frame rather than per group. That is a structural bet: the package increasingly inspects what you wrote to decide how much work is actually needed. Running alongside it is a steady tightening of build requirements, with C++17, R 4.5.0 and CRAN's C API rules all landing within a year.
◆ Prediction
Expect the group-unaware classification to widen to more functions, since each addition compounds across every grouped expression, and expect the dependency floors to keep rising as the package tracks CRAN's compiled-code policy.
kernelshap
0.0ANALYTICS
kernelshap makes permutation SHAP practical past eight features, then fixes the kernel weights it had wrong.
◆ Current state
kernelshap computes model-agnostic SHAP values in R through Kernel SHAP, permutation SHAP and an exact additive explainer. Version 0.8.0 added a sampling permutation-SHAP algorithm with standard errors and early stopping, lifting the practical feature ceiling past what the exact method allows. Version 0.9.0 then corrected a bug in how kernel weights were computed — exact Kernel SHAP now agrees with exact permutation SHAP — and moved parallelism from foreach to doFuture.
◆ Where it's heading
Two concerns drive this package: making exact methods reach further, and being demonstrably right. The first shows in the additive explainer, the optional background dataset and the sampling permutation algorithm; the second in unit tests written against Python's shap, credited fixes from outside contributors, and a willingness to ship a correctness fix that changes numbers people have already published. Speed work runs continuously underneath — direct solves replacing the Moore-Penrose pseudo-inverse, roughly 10% less memory.
◆ Prediction
The 0.6.0 and 0.7.0 notes each promised a stable 1.0.0 that has not arrived; with the weighting bug fixed and parallelism reworked, a 1.0 release is the most plausible next step.
Alternatives to fastplyr and kernelshap
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 fastplyr or kernelshap.
lineup2
lineup2 ships once every few years, and 2026's release is a logo and a core-count tweak.
stringi
stringi has spent two years on build hardening since its Unicode 15.1 reset.
OHPL
A 2017 chemometrics method frozen in place, visited only when CRAN changes its documentation rules.
trias
Belgium's invasive-species indicator toolkit is in steady refinement, one plotting edge case at a time.
icesSAG
The ICES stock assessment client took upload away in 2024 and spent two years giving it back.
See all fastplyr alternatives → · See all kernelshap alternatives →
Recent activity from fastplyr and kernelshap
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
- 4mo agofastplyrNon-API C functions dropped, R 4.5.0 now required
- 8mo agofastplyrIn-place sorting arrives with a C++17 requirement
- 10mo agofastplyrGroup-unaware expressions evaluated on the whole frame
- 1y agokernelshapKernel weight bug fixed; parallelism moves to doFuture
- 1y agokernelshapSampling permutation SHAP with standard errors ⚡
- 1y agofastplyrf_mutate and f_reframe complete the verb set
- 1y agofastplyrDynamic argument evaluation and f_pull
- 1y agofastplyrf_fill added and grouped joins repaired
- 1y agokernelshapBackground data now optional; ranger survival support
- 2y agokernelshapFactor-valued predictions dropped
- 2y agokernelshapadditive_shap() explains additive models exactly
- 2y agokernelshapFaster on plain data.frames
Frequently asked questions
What is the difference between fastplyr and kernelshap?
They serve adjacent needs but don't currently overlap on shipped themes. fastplyr and kernelshap 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 fastplyr better than kernelshap?
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. fastplyr and kernelshap 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 fastplyr?
Top fastplyr alternatives in Analytics are ranked by recent ship velocity. Browse the "fastplyr alternatives" section above for the current picks, or visit /alternatives/fastplyr for the full list with editorial commentary on each.
What are the best alternatives to kernelshap?
Top kernelshap alternatives in Analytics are ranked by recent ship velocity. Browse the "kernelshap alternatives" section above for the current picks, or visit /alternatives/kernelshap for the full list with editorial commentary on each.