← Back to home
Comparison · Analytics

audubon vs kernelshap

A side-by-side editorial comparison of audubon and kernelshap — release velocity, themes, recent moves, and the top alternatives to consider.

audubon vs kernelshap: at a glance

Featureaudubonkernelshap
SectorAnalyticsAnalytics
Velocity score0.00.0
Sparks · 30d00
Top themesjapanese-nlp, text-processing, r-package, budouxshap, model explainability, sampling algorithms, numerical correctness
Last editorial update1h ago5h ago
WebsiteVisit →Visit →

What is audubon?

audubon's release feed is almost entirely Renovate bumping the JavaScript toolchain behind its Japanese text splitter.

An R package for Japanese text processing — normalisation, tokenisation via MeCab and SudachiPy, and phrase splitting through budoux. Ten releases since 2022, but the changelogs are dominated by automated dependency updates to a webpack, babel and prettier toolchain, because the budoux component is JavaScript that has to be bundled. Actual R-facing changes appear in perhaps one release in three.

Read the full audubon trajectory →

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.

Read the full kernelshap trajectory →

audubon vs kernelshap: editorial side-by-side

A
audubon
ANALYTICS
0.0

audubon's release feed is almost entirely Renovate bumping the JavaScript toolchain behind its Japanese text splitter.

◆ Current state

An R package for Japanese text processing — normalisation, tokenisation via MeCab and SudachiPy, and phrase splitting through budoux. Ten releases since 2022, but the changelogs are dominated by automated dependency updates to a webpack, babel and prettier toolchain, because the budoux component is JavaScript that has to be bundled. Actual R-facing changes appear in perhaps one release in three.

◆ Where it's heading

The package appears feature-stable and in maintenance. The last substantive R-level addition visible here is bind_lr() for bigram LR values back in 0.5.0; everything since has been dependency hygiene, a tokeniser refactor, and platform-specific test fixes. That is a reasonable end state for a wrapper whose value is the binding rather than ongoing invention, but it does mean the release feed carries almost no signal about the package itself — a reader watching this feed would learn more about webpack's version history than about Japanese text processing.

◆ Prediction

Expect the Renovate cadence to continue setting the release rhythm, with R-facing changes arriving only when budoux itself gains capability or a platform breaks.

K
kernelshap
ANALYTICS
0.0

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 audubon 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 audubon or kernelshap.

See all audubon alternatives → · See all kernelshap alternatives →

Recent activity from audubon and kernelshap

Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.

  1. 3mo agoaudubonM1 Mac locale crash worked around in examples
  2. 7mo agoaudubonaudubon 0.6.2
  3. 7mo agoaudubonAutomated dependency bumps, including a webpack security update
  4. 1y agokernelshapKernel weight bug fixed; parallelism moves to doFuture
  5. 1y agokernelshapSampling permutation SHAP with standard errors
  6. 1y agokernelshapBackground data now optional; ranger survival support
  7. 2y agokernelshapFactor-valued predictions dropped
  8. 2y agokernelshapadditive_shap() explains additive models exactly
  9. 2y agoaudubonbudoux bumped to 0.6.2; Renovate configured
  10. 2y agokernelshapFaster on plain data.frames
  11. 3y agoaudubonMeCab and SudachiPy tokenisers refactored
  12. 3y agoaudubonbind_lr() computes LR values for bigrams

Frequently asked questions

What is the difference between audubon and kernelshap?

They serve adjacent needs but don't currently overlap on shipped themes. audubon 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 audubon better than kernelshap?

Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. audubon 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 audubon?

Top audubon alternatives in Analytics are ranked by recent ship velocity. Browse the "audubon alternatives" section above for the current picks, or visit /alternatives/audubon-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.