bbotk
bbotk is generalizing from an optimizer toolkit into an evaluation framework.
Comparison · Analytics
pins vs themis
A side-by-side editorial comparison of pins and themis — release velocity, themes, recent moves, and the top alternatives to consider.
pins vs themis: at a glance
What is pins?
pins keeps adding a storage backend per release while retiring its original API
pins publishes and versions R objects to a board, where a board is whatever storage you have. The recent releases read as a steady list of new boards - Google Cloud Storage, Google Drive, Databricks Volumes, Connect vanity URLs - alongside serialization changes that track which formats R users actually want: parquet via nanoparquet, and qs replaced by qs2.
What is themis?
themis is back to adding real resampling algorithms after a documentation-heavy stretch.
themis supplies recipes steps for handling class imbalance in tidymodels. The 1.0.x line was consumed by documentation accuracy, message translation and internal consistency work. Version 1.1.0 returns to substance with two new under-sampling methods.
pins vs themis: editorial side-by-side
pins
0.0ANALYTICS
pins keeps adding a storage backend per release while retiring its original API
◆ Current state
pins publishes and versions R objects to a board, where a board is whatever storage you have. The recent releases read as a steady list of new boards - Google Cloud Storage, Google Drive, Databricks Volumes, Connect vanity URLs - alongside serialization changes that track which formats R users actually want: parquet via nanoparquet, and qs replaced by qs2.
◆ Where it's heading
Two long-running processes, neither dramatic. Backend coverage expands toward wherever teams already store artifacts, which increasingly means Databricks and cloud object storage rather than a shared drive. Meanwhile the legacy pin() API from before the board model has been in a staged deprecation across at least three releases, escalated each time rather than removed.
◆ Prediction
Expect another board or two as storage platforms are requested, and the legacy pin() functions to finally become errors; the format list will keep tracking whichever serializer the R community settles on.
themis
2.5ANALYTICS
themis is back to adding real resampling algorithms after a documentation-heavy stretch.
◆ Current state
themis supplies recipes steps for handling class imbalance in tidymodels. The 1.0.x line was consumed by documentation accuracy, message translation and internal consistency work. Version 1.1.0 returns to substance with two new under-sampling methods.
◆ Where it's heading
The package grows by adding algorithms rather than restructuring itself. tomek() was rewritten to handle multiple classes and drop the unbalanced dependency, case weights arrived at 1.0.0, and cluster-centroid and condensed-nearest-neighbour under-sampling arrive now — each shipped as both a recipes step and a direct-implementation function.
◆ Prediction
Expect further under- and over-sampling methods in the same paired form, as the package fills out coverage of the standard class-imbalance literature.
Alternatives to pins and themis
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 pins or themis.
bayesplot
bayesplot keeps widening its posterior-check catalogue while absorbing each ggplot2 break.
animation
Contribution-driven maintenance on a package whose last structural change was magick support.
tsibble
tsibble shipped one release in five and a half years - the data structure is finished
yardstick
yardstick made fairness metrics a first-class part of tidymodels evaluation
tune
tune extends tuning past the model itself to postprocessors, and adds a second parallel backend
Recent activity from pins and themis
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
- 10d agothemisthemis 1.1.0 adds cluster-centroid and CNN under-sampling
- 5mo agopinsqs2 replaces qs; pins can be written in multiple formats
- 1y agopinsPin previews on Connect; Databricks host normalization
- 1y agothemisthemis 1.0.3 corrects resampling direction in documentation
- 1y agopinsAdds board_databricks() and switches parquet to nanoparquet
- 2y agopinspin_write() arguments must be named; Connect caches removed
- 2y agopinsMessage clarity and Google Drive dribble handling
- 2y agothemisthemis 1.0.2 makes internal consistency and speed changes
- 2y agopinsboard_gdrive() added; cache location configurable
- 3y agothemisthemis 1.0.1 fixes upsampling errors when none is needed
- 4y agothemisthemis 1.0.0 adds case weights to up- and down-sampling
- 4y agothemisthemis 0.2.2 rewrites tomek() for multiclass, drops a dependency
Frequently asked questions
What is the difference between pins and themis?
They serve adjacent needs but don't currently overlap on shipped themes. themis is currently shipping more aggressively (velocity 2.5 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 pins better than themis?
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. themis is currently shipping more aggressively (velocity 2.5 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 pins?
Top pins alternatives in Analytics are ranked by recent ship velocity. Browse the "pins alternatives" section above for the current picks, or visit /alternatives/pins-r for the full list with editorial commentary on each.
What are the best alternatives to themis?
Top themis alternatives in Analytics are ranked by recent ship velocity. Browse the "themis alternatives" section above for the current picks, or visit /alternatives/themis for the full list with editorial commentary on each.