Just before ACoP, and just in time for the pre-conference workshop, we released Pumas version 2.3.0. This was a release with many new features and adjustments to Pumas.
As always, the full release notes are available, and can be found here: Release Notes for Pumas 2.3.0 · Pumas . Some notable highlights are:
- Vastly improved workflows and features for bayesian analysis - keep an eye out for the new tutorials. Also have a look at the documentation on bayesian inference Bayesian Workflow · Pumas.
- Steady-state solutions for dynamic models with analytical solutions (
Central1
,Central1Periph1Meta1
, linear ordinary differential equations, etc) are now calculated using a closed form expression. - Expose
optim_alg
andoptim_options
as separate keywords instead of using theoptimize_fn
interface infit
. - Export many functions and types that previously required to be qualified (needed
Pumas.
in front, complete list in release notes). - Add the option to choose the mode of parallelization in
predict
with aPopulation
as input. The new default isEnsembleThreads
.
As well as several fixes for various issues throughout Pumas, see the release notes for all of these improvements.
We hope you enjoy all the improvements we made to Pumas in this release!
Pumas-AI dev team