mlr3proba
mlr3proba is shedding weight as its survival work moves into sibling packages
Comparison · Analytics
mlr3learners vs probably
A side-by-side editorial comparison of mlr3learners and probably — release velocity, themes, recent moves, and the top alternatives to consider.
mlr3learners vs probably: at a glance
What is mlr3learners?
mlr3learners spends its releases absorbing upstream churn
mlr3learners wraps the standard model packages — ranger, xgboost, glmnet, kknn — for mlr3. Its recent history is dominated by upstream events rather than its own plans: kknn was pulled from CRAN and its learners removed in 0.11.0, then restored in 0.12.0 when the package returned. The newest release absorbs glmnet 5.0 while adding a predict_raw flag across all learners and probit support to logistic regression.
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.
mlr3learners vs probably: editorial side-by-side
mlr3learners
0.0ANALYTICS
mlr3learners spends its releases absorbing upstream churn
◆ Current state
mlr3learners wraps the standard model packages — ranger, xgboost, glmnet, kknn — for mlr3. Its recent history is dominated by upstream events rather than its own plans: kknn was pulled from CRAN and its learners removed in 0.11.0, then restored in 0.12.0 when the package returned. The newest release absorbs glmnet 5.0 while adding a predict_raw flag across all learners and probit support to logistic regression.
◆ Where it's heading
The package's job is insulation, and the changelog shows what that costs — compatibility-only releases interleaved with small capability additions that expose more of each upstream model. The direction of travel is toward giving users access to the raw upstream objects rather than hiding them.
◆ Prediction
Expect the next release to track another upstream version bump, with incremental exposure of learner-specific fields continuing alongside.
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 mlr3learners 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 mlr3learners 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 mlr3learners alternatives → · See all probably alternatives →
Recent activity from mlr3learners and probably
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
- 2mo agomlr3learnerspredict_raw across all learners, plus probit and ranger internals
- 8mo agomlr3learnersxgboost 3.1.2.1 compatibility
- 9mo agomlr3learnersUncertainty estimation methods for ranger regression
- 10mo agoprobablyConformal objects gain required_pkgs() and butcher methods
- 10mo agomlr3learnersDevelopment snapshot: LDA test adjustment
- 1y agoprobablyggplot2 test updates and a clearer validation-set error
- 1y agomlr3learnerskknn learners restored after returning to CRAN
- 1y agoprobablyCalibration internals refactored; isotonic bootstrap bug fixed
- 1y agomlr3learnerskknn learners removed after CRAN archival
- 2y agoprobablyFix grouping sensitivity to variable type
- 3y agoprobablySplit conformal and conformal quantile regression added
- 3y agoprobablyCalibration and conformal inference arrive in tidymodels ⚡
Frequently asked questions
What is the difference between mlr3learners and probably?
They serve adjacent needs but don't currently overlap on shipped themes. mlr3learners 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 mlr3learners better than probably?
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. mlr3learners 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 mlr3learners?
Top mlr3learners alternatives in Analytics are ranked by recent ship velocity. Browse the "mlr3learners alternatives" section above for the current picks, or visit /alternatives/mlr3learners 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.