tern.rbmi
Reference-based multiple imputation tables, shipping only what CRAN checks demand.
Comparison · Analytics
geodist vs kernelshap
A side-by-side editorial comparison of geodist and kernelshap — release velocity, themes, recent moves, and the top alternatives to consider.
geodist vs kernelshap: at a glance
What is geodist?
geodist stays dependency-free and fast, and warns you when 'cheap' distances stop being honest.
geodist computes geodesic distances between coordinate pairs in C with no dependencies, offering several measures that trade accuracy for speed — including a 'cheap' approximation used by default. The API is small and largely finished; 0.1.0 added geodist_min() for nearest-match lookups and 0.1.1 is a compiler warning fix.
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.
geodist vs kernelshap: editorial side-by-side
geodist
0.0ANALYTICS
geodist stays dependency-free and fast, and warns you when 'cheap' distances stop being honest.
◆ Current state
geodist computes geodesic distances between coordinate pairs in C with no dependencies, offering several measures that trade accuracy for speed — including a 'cheap' approximation used by default. The API is small and largely finished; 0.1.0 added geodist_min() for nearest-match lookups and 0.1.1 is a compiler warning fix.
◆ Where it's heading
Development has been about making the speed-accuracy trade visible rather than hiding it. The 0.0.6 release added messages telling users to pick a different measure once the default cheap approximation is applied beyond 100km, where its error stops being negligible. Around that, the work is input handling — tibble support, better lon/lat column matching, vector inputs — and hardening the C code. It is a package that treats being small and correct as the feature.
◆ Prediction
Expect continued low-frequency maintenance: compiler warnings and geodesic source updates account for three of the last six releases, and the function surface has grown by only two entries in five years.
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 geodist 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 geodist or kernelshap.
tern.mmrm
An MMRM tabulation package that has published nothing since its 2024 CRAN releases.
ggmagnify
A single-purpose ggplot2 inset tool, refining the same three arguments.
symengine
An R symbolic-maths binding whose changelog is really the C++ core's release notes.
gtfstools
gtfstools stopped guarding its own object model and started accepting everyone else's.
quantities
The glue package that makes R carry units and uncertainty through the same calculation.
See all geodist alternatives → · See all kernelshap alternatives →
Recent activity from geodist and kernelshap
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
- 1y agokernelshapKernel weight bug fixed; parallelism moves to doFuture
- 1y agokernelshapSampling permutation SHAP with standard errors ⚡
- 1y agogeodistgeodist 0.1.1 clears a clang warning in geodesic.c
- 1y agokernelshapBackground data now optional; ranger survival support
- 2y agokernelshapFactor-valued predictions dropped
- 2y agokernelshapadditive_shap() explains additive models exactly
- 2y agogeodistgeodist 0.1.0 adds geodist_min() for nearest matches
- 2y agokernelshapFaster on plain data.frames
- 3y agogeodistgeodist 0.0.8 updates geodesic source, fixes clang warnings
- 5y agogeodistgeodist 0.0.7 improves lon/lat column matching and tibbles
- 5y agogeodistgeodist 0.0.6 warns when cheap distances exceed 100km
- 6y agogeodistgeodist 0.0.4 adds geodist_vec() for vector inputs
Frequently asked questions
What is the difference between geodist and kernelshap?
They serve adjacent needs but don't currently overlap on shipped themes. geodist 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 geodist better than kernelshap?
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. geodist 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 geodist?
Top geodist alternatives in Analytics are ranked by recent ship velocity. Browse the "geodist alternatives" section above for the current picks, or visit /alternatives/geodist-r 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.