mlr3proba
mlr3proba is shedding weight as its survival work moves into sibling packages
Comparison · Analytics
parameters vs probably
A side-by-side editorial comparison of parameters and probably — release velocity, themes, recent moves, and the top alternatives to consider.
parameters vs probably: at a glance
What is parameters?
easystats' parameters package absorbs one more model class every few weeks
parameters extracts and formats coefficients from an enormous range of R model objects, and its releases read as a running ledger of that range expanding — lavaan and lavaan.mi, survey, lcmm, glmmTMB, fixest, marginaleffects, ordinal. Recent versions ship roughly monthly with a mix of new support, new arguments, and fixes for label handling and standard errors. The most consequential recent change is behavioral: post-hoc standardization no longer standardizes the intercept, setting it and its inferential statistics to NA.
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.
parameters vs probably: editorial side-by-side
parameters
0.0ANALYTICS
easystats' parameters package absorbs one more model class every few weeks
◆ Current state
parameters extracts and formats coefficients from an enormous range of R model objects, and its releases read as a running ledger of that range expanding — lavaan and lavaan.mi, survey, lcmm, glmmTMB, fixest, marginaleffects, ordinal. Recent versions ship roughly monthly with a mix of new support, new arguments, and fixes for label handling and standard errors. The most consequential recent change is behavioral: post-hoc standardization no longer standardizes the intercept, setting it and its inferential statistics to NA.
◆ Where it's heading
The package's job is to be the universal adapter for model output, so its roadmap is effectively set by what the R modelling ecosystem produces. Two threads are visible beyond coverage: getting standard errors right for awkward cases such as frailty terms and robust vcov matrices, and getting labels right when factors are converted on the fly or character variables appear in a formula. Interoperability inside easystats keeps tightening, with equivalence_test() gaining methods for modelbased objects.
◆ Prediction
Given the cadence, the next release will most likely add another model class alongside label and standard-error fixes rather than change how the package works.
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 parameters 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 parameters 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
tidyposterior
A finished Bayesian model-comparison package in pure maintenance mode
See all parameters alternatives → · See all probably alternatives →
Recent activity from parameters and probably
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
- 1mo agoparametersparameters 0.29.2 extends lavaan support and fixes label dropping
- 2mo agoparametersparameters 0.29.1 adds a cluster argument and fixes vcov handling
- 3mo agoparametersparameters 0.29.0 stops standardizing the intercept in post-hoc methods
- 8mo agoparametersparameters 0.28.3 adds Kenward-Roger and Satterthwaite for glmmTMB
- 10mo agoprobablyConformal objects gain required_pkgs() and butcher methods
- 11mo agoparametersparameters 0.28.2 updates tests for the latest fixest release
- 11mo agoparametersparameters 0.28.1 adds robust standard errors for glmmTMB
- 1y agoprobablyggplot2 test updates and a clearer validation-set error
- 1y agoprobablyCalibration internals refactored; isotonic bootstrap bug fixed
- 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 parameters and probably?
They serve adjacent needs but don't currently overlap on shipped themes. parameters 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 parameters better than probably?
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. parameters 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 parameters?
Top parameters alternatives in Analytics are ranked by recent ship velocity. Browse the "parameters alternatives" section above for the current picks, or visit /alternatives/parameters-r 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.