mice
mice can finally predict, not just estimate, from multiply imputed data.
A side-by-side editorial comparison of midr and Skipper — release velocity, themes, recent moves, and the top alternatives to consider.
A black-box interpreter reaches CRAN, then learns multi-class and survival responses
midr explains black-box models by fitting an interpretable surrogate through Maximum Interpretation Decomposition — main effects plus second-order interactions, with exact Shapley values for the surrogate. Two months after its first CRAN release it can take a matrix response, which covers multi-class classification and survival models, and hold collections of fitted interpretations in midlist and midrib objects for comparison.
Skipper trims a 4x memory regression out of routesrv, days after shipping h2c
Zalando's HTTP router ships patch tags almost daily, and most carry a single dependency bump or a one-line auth fix. The substance in this window is v0.27.63, which stops routesrv holding an uncompressed route tree alongside the compressed buffer: retained memory in the release's own benchmark falls from 30.7 MB to 2.0 MB, undoing most of the 4x increase zone-aware routing had introduced. Around it sit narrow auth repairs, and an eskip parser change that finally accepts negative numeric arguments in predicates and filters.
midr explains black-box models by fitting an interpretable surrogate through Maximum Interpretation Decomposition — main effects plus second-order interactions, with exact Shapley values for the surrogate. Two months after its first CRAN release it can take a matrix response, which covers multi-class classification and survival models, and hold collections of fitted interpretations in midlist and midrib objects for comparison.
The releases move outward along two axes at once: what can be interpreted, and how much of it fits in memory. Version 0.5.3 rebuilt the fitting path to avoid materialising large design matrices and added a save.memory option; 0.6.0 widened the response from a vector to a matrix and added parametric link functions. Class and argument names were shortened in the same release, so the package is still willing to break itself this early.
With multiple models now held in one object and visualisation methods for them, comparison across models is the surface most likely to fill out next — the collection classes exist but the notes describe manipulation and plotting rather than any comparison metric.
Zalando's HTTP router ships patch tags almost daily, and most carry a single dependency bump or a one-line auth fix. The substance in this window is v0.27.63, which stops routesrv holding an uncompressed route tree alongside the compressed buffer: retained memory in the release's own benchmark falls from 30.7 MB to 2.0 MB, undoing most of the 4x increase zone-aware routing had introduced. Around it sit narrow auth repairs, and an eskip parser change that finally accepts negative numeric arguments in predicates and filters.
Two threads run through the recent tags. The data path keeps getting real work - leastRequests balancing, then h2c end to end in v0.27.57, now the memory cost of zone-aware routing being paid back. The auth filters, by contrast, are only being maintained: token introspection, grant auth, and now the OIDC Referer handling are fixes rather than new capability, and several are follow-ups to each other rather than independent bugs.
Zone-aware routing looks like the source of the recent memory attention, so expect further tuning around route storage and the hash computation done on every pull, which the release's second benchmark already isolates. The OIDC cookie change is the second link in an auth chain that started with the grant-auth fix, and is likely to draw another follow-up.
Other Infra & APIs 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 midr or Skipper.
mice can finally predict, not just estimate, from multiply imputed data.
A market-microstructure toolkit that keeps adding estimators as the papers land.
A vowel-analysis package trimming dependencies after an email address got it archived.
The R half of the EMU speech database system, fixing what was quietly broken.
A Bayesian model-averaging package spending its 2.0 on memory, not methods.
tidyplots keeps rebuilding its own foundations rather than layering around them.
See all midr alternatives → · See all Skipper alternatives →
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
They serve adjacent needs but don't currently overlap on shipped themes. Skipper is currently shipping more aggressively (velocity 5.0 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.
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. Skipper is currently shipping more aggressively (velocity 5.0 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 Infra & APIs products to evaluate alongside.
Top midr alternatives in Infra & APIs are ranked by recent ship velocity. Browse the "midr alternatives" section above for the current picks, or visit /alternatives/midr for the full list with editorial commentary on each.
Top Skipper alternatives in Infra & APIs are ranked by recent ship velocity. Browse the "Skipper alternatives" section above for the current picks, or visit /alternatives/skipper for the full list with editorial commentary on each.