RMVMR
RMVMR is being tidied in lockstep with MVMR, the package it wraps
Comparison · Analytics
Fulcrum vs kernelshap
A side-by-side editorial comparison of Fulcrum and kernelshap — release velocity, themes, recent moves, and the top alternatives to consider.
Fulcrum vs kernelshap: at a glance
| Feature | Fulcrum | kernelshap |
|---|---|---|
| Sector | Analytics | Analytics |
| Velocity score | 6.3 | 0.0 |
| Sparks · 30d | 0 | 0 |
| Top themes | gis, esri-migration, offline-maps, field-data-capture | shap, model explainability, sampling algorithms, numerical correctness |
| Last editorial update | 7h ago | 1h ago |
| Website | — | Visit → |
What is Fulcrum?
Fulcrum is betting its whole map stack on Esri, with a hard Google Maps cutoff on September 1.
Fulcrum is a field data collection platform, and nearly every entry in the last month touches mapping. The web app ships weekly fix batches for layer rendering (KML/KMZ, MBTiles, ArcGIS Feature Services), while iOS and Android push near-weekly builds against the ArcGIS SDK. Feature work is thin right now; the visible surface is stabilization around a mapping engine migration already in flight.
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.
Fulcrum vs kernelshap: editorial side-by-side
Fulcrum
6.3ANALYTICS
Fulcrum is betting its whole map stack on Esri, with a hard Google Maps cutoff on September 1.
◆ Current state
Fulcrum is a field data collection platform, and nearly every entry in the last month touches mapping. The web app ships weekly fix batches for layer rendering (KML/KMZ, MBTiles, ArcGIS Feature Services), while iOS and Android push near-weekly builds against the ArcGIS SDK. Feature work is thin right now; the visible surface is stabilization around a mapping engine migration already in flight.
◆ Where it's heading
The direction is a full consolidation onto Esri. The legacy Google Maps engine is being retired on September 1 with automatic migration for anyone who hasn't switched, and Esri now carries Google's satellite and street basemaps so the imagery argument is neutralized. Underneath, the mobile SDK moved to ArcGIS 300.0.0 and ONNX on-device inference was dropped for a new INFERENCE format with labels.txt support. Offline is the second thread: downloadable layers can now be updated in place rather than deleted and recreated.
◆ Prediction
Expect the weeks before September 1 to stay dominated by migration-shaped fixes and Esri feature parity work, with the MMPK layer and Photo FastFill early-access programs the most likely to graduate to general availability next.
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 Fulcrum 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 Fulcrum or kernelshap.
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 Fulcrum alternatives → · See all kernelshap alternatives →
Recent activity from Fulcrum and kernelshap
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
- 16h agoFulcrumWeb fixes: shared-view exports, MBTiles popups, KML and ArcGIS layers
- 2d agoFulcrumAndroid: update offline map layers in place, non-blocking Photo FastFill
- 3d agoFulcrumiOS fix: slow location resolution blocked record saves
- 8d agoFulcrumWeb: SSO email wording and ArcGIS/deck.gl version bumps
- 9d agoFulcrumAndroid fixes: ArcGIS stability, signature button, photo markup
- 9d agoFulcrumiOS fixes: basemap loading and black map screen
- 1y agokernelshapKernel weight bug fixed; parallelism moves to doFuture
- 1y agokernelshapSampling permutation SHAP with standard errors ⚡
- 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 Fulcrum and kernelshap?
They serve adjacent needs but don't currently overlap on shipped themes. Fulcrum is currently shipping more aggressively (velocity 6.3 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 Fulcrum better than kernelshap?
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. Fulcrum is currently shipping more aggressively (velocity 6.3 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 Fulcrum?
Top Fulcrum alternatives in Analytics are ranked by recent ship velocity. Browse the "Fulcrum alternatives" section above for the current picks, or visit /alternatives/fulcrum 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.