mlr3proba
mlr3proba is shedding weight as its survival work moves into sibling packages
Comparison · Analytics
hardhat vs probably
A side-by-side editorial comparison of hardhat and probably — release velocity, themes, recent moves, and the top alternatives to consider.
Shared themes:tidymodels
hardhat vs probably: at a glance
What is hardhat?
hardhat keeps adding the contracts tidymodels needs next
hardhat is the infrastructure layer under tidymodels, defining the preprocessing and extraction contracts other packages implement. Its releases read as a list of new generics and vector classes: extract_postprocessor(), extract_fit_time(), extract_tailor(), and a quantile_pred() class for quantile-regression output. The newest release is narrow warning and missing-value handling in mold().
What is probably?
The package that made calibration a step instead of an afterthought.
probably started as a small utility for class predictions and equivocal zones, and version 1.0.0 turned it into tidymodels' calibration and uncertainty package: cal_plot_*, cal_estimate_*, cal_validate_* and cal_apply across binary, multiclass and regression problems, plus conformal prediction intervals. Since then the work has been consolidation — a large internal refactor with no API change, split conformal and conformal quantile regression, bound_prediction(), and required_pkgs() and butcher methods so conformal objects can be deployed and stripped.
hardhat vs probably: editorial side-by-side
hardhat
0.0ANALYTICS
hardhat keeps adding the contracts tidymodels needs next
◆ Current state
hardhat is the infrastructure layer under tidymodels, defining the preprocessing and extraction contracts other packages implement. Its releases read as a list of new generics and vector classes: extract_postprocessor(), extract_fit_time(), extract_tailor(), and a quantile_pred() class for quantile-regression output. The newest release is narrow warning and missing-value handling in mold().
◆ Where it's heading
Each addition here lands ahead of a user-facing feature elsewhere in tidymodels — the postprocessor and tailor generics precede the postprocessing workflow, quantile_pred() precedes quantile prediction in parsnip. The package's own surface stays deliberately small and its cadence follows what the rest of the stack is about to need.
◆ Prediction
Expect further extraction generics and prediction-type classes as tidymodels builds out postprocessing, with hardhat's own API remaining thin.
probably
0.0ANALYTICS
The package that made calibration a step instead of an afterthought.
◆ Current state
probably started as a small utility for class predictions and equivocal zones, and version 1.0.0 turned it into tidymodels' calibration and uncertainty package: cal_plot_*, cal_estimate_*, cal_validate_* and cal_apply across binary, multiclass and regression problems, plus conformal prediction intervals. Since then the work has been consolidation — a large internal refactor with no API change, split conformal and conformal quantile regression, bound_prediction(), and required_pkgs() and butcher methods so conformal objects can be deployed and stripped.
◆ Where it's heading
The recent releases are about making these objects survive leaving the session. butcher and required_pkgs() methods are what a model needs to be pinned, containerised and served, and their arrival alongside workflows adding a tailor postprocessing stage and vetiver adding probably support points the same way: calibration is being moved out of analysis scripts and into the deployed pipeline. The cal_*_none() reference implementations are the tell that calibration is now something people tune rather than apply once.
◆ Prediction
Expect the calibration functions to be reachable directly from a tuned workflow's postprocessing stage rather than applied to predictions afterwards, following the tailor integration that workflows just shipped.
Alternatives to hardhat and probably
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 hardhat or probably.
mlr3viz
mlr3viz keeps the ecosystem's plots working while the plots themselves move out
mlr3tuning
mlr3tuning is rebuilding its async machinery under a stable public surface
timetk
timetk swallowed anomalize whole, then went quiet for two years
modelbased
modelbased is turning marginal effects into a full contrast grammar
parameters
easystats' parameters package absorbs one more model class every few weeks
See all hardhat alternatives → · See all probably alternatives →
Recent activity from hardhat and probably
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
- 4mo agohardhatmold() warning and quantile missing-value fixes
- 10mo agoprobablyConformal objects gain required_pkgs() and butcher methods
- 11mo agohardhatextract_tailor() generic added
- 1y agoprobablyggplot2 test updates and a clearer validation-set error
- 1y agoprobablyCalibration internals refactored; isotonic bootstrap bug fixed
- 1y agohardhatquantile_pred() class for quantile regression output
- 2y agohardhatextract_postprocessor() and extract_fit_time() generics
- 2y agoprobablyFix grouping sensitivity to variable type
- 2y agohardhatDocumentation topic renamed at CRAN's request
- 3y agoprobablySplit conformal and conformal quantile regression added
- 3y agoprobablyCalibration and conformal inference arrive in tidymodels ⚡
- 3y agohardhatMulti-outcome prediction helpers and one-hot factor encoding
Frequently asked questions
What is the difference between hardhat and probably?
Both compete on the same themes — tidymodels — within Analytics. hardhat and probably 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 hardhat better than probably?
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. hardhat and probably 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 hardhat?
Top hardhat alternatives in Analytics are ranked by recent ship velocity. Browse the "hardhat alternatives" section above for the current picks, or visit /alternatives/hardhat for the full list with editorial commentary on each.
What are the best alternatives to probably?
Top probably alternatives in Analytics are ranked by recent ship velocity. Browse the "probably alternatives" section above for the current picks, or visit /alternatives/probably for the full list with editorial commentary on each.