r2rtf
The clinical-report table engine learned Chinese, then learned to leave RTF entirely
A side-by-side editorial comparison of brulee and USAboundaries — release velocity, themes, recent moves, and the top alternatives to consider.
tidymodels' torch backend grew from MLPs into a tabular deep learning suite with foundation models.
brulee fits neural networks for tidymodels on torch, and 1.0.0 redefined what that means: alongside the original MLP it now ships Regularization Learning Networks, ResNet with skip connections and batch normalization, AutoInt with columnwise attention, SAINT with row and column attention, and Chronos2, a foundational forecasting model. GPU acceleration arrived in the same release with automatic CUDA selection and opt-in MPS. Version 1.1.0 added TabICL, an open-source tabular foundation model, and 1.1.1 spent its time cleaning up the consequences of shipping models that need weight downloads.
Seven years dormant, then two releases dragging every census boundary from 2020 to 2024
USAboundaries supplies contemporary and historical US boundary data — states, counties, congressional districts, cities, ZIP code tabulation areas — as sf objects, with the bulk data held in a companion USAboundariesData package. After a gap running from 2018 to late 2025, two releases a month apart refreshed the contemporary census vintage from 2020 to 2024: 0.5.0 covered everything except states, and 0.5.1 finished the job.
brulee fits neural networks for tidymodels on torch, and 1.0.0 redefined what that means: alongside the original MLP it now ships Regularization Learning Networks, ResNet with skip connections and batch normalization, AutoInt with columnwise attention, SAINT with row and column attention, and Chronos2, a foundational forecasting model. GPU acceleration arrived in the same release with automatic CUDA selection and opt-in MPS. Version 1.1.0 added TabICL, an open-source tabular foundation model, and 1.1.1 spent its time cleaning up the consequences of shipping models that need weight downloads.
The package has crossed from a torch convenience wrapper into a catalog of current tabular architectures, and the recent releases show it absorbing what that costs. Pretrained weights meant a 400MB download, so 1.1.1 stopped fetching them on attach and moved the cache to the platform-appropriate R_user_dir location. Numerical robustness is the other constant thread — 64-bit tensors, Gaussian initialization, gradient clipping extended architecture by architecture, and a ResNet batch-normalization bug where a single-row trailing batch produced all-NA predictions.
Gradient clipping has been rolled out one architecture at a time and TabICL is the newest arrival, so the likely next step is bringing the foundation models to parity with the trained ones on device selection, prediction types, and the tuning surface.
USAboundaries supplies contemporary and historical US boundary data — states, counties, congressional districts, cities, ZIP code tabulation areas — as sf objects, with the bulk data held in a companion USAboundariesData package. After a gap running from 2018 to late 2025, two releases a month apart refreshed the contemporary census vintage from 2020 to 2024: 0.5.0 covered everything except states, and 0.5.1 finished the job.
For a data package the release cycle is the data vintage, and the 0.3.0 split into a separate data package was designed precisely so those refreshes would not require a code release. That the 2024 update still arrived as two package versions seven years later says the mechanism is being used sparingly. Nothing in the feed shows work on the API itself since us_cities() gained an sf return type and a states argument in 2018.
The next release is most likely another vintage refresh when the census data moves again, rather than new geographies or functions. The split of 0.5.0 and 0.5.1 suggests state boundaries are handled on a separate path from the rest and may lag again.
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 brulee or USAboundaries.
The clinical-report table engine learned Chinese, then learned to leave RTF entirely
New stewardship at openpharma, then two releases adding the methods MCP-Mod was missing
The stubbing library added httr2 support, then spent a year cutting itself free of everything else
crul took mocking back from webmockr and made it a property of the client itself
Six releases, six identical bodies — the feed carries the package abstract instead of release notes
chattr deleted every LLM integration it had written and outsourced the lot to ellmer
See all brulee alternatives → · See all USAboundaries alternatives →
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
They serve adjacent needs but don't currently overlap on shipped themes. brulee and USAboundaries 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.
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. brulee and USAboundaries 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.
Top brulee alternatives in Analytics are ranked by recent ship velocity. Browse the "brulee alternatives" section above for the current picks, or visit /alternatives/brulee for the full list with editorial commentary on each.
Top USAboundaries alternatives in Analytics are ranked by recent ship velocity. Browse the "USAboundaries alternatives" section above for the current picks, or visit /alternatives/usaboundaries for the full list with editorial commentary on each.