Releases: liesel-devs/liesel
Releases Β· liesel-devs/liesel
Liesel v0.2.9
- π§ Pinned scipy version to <=1.12 (@jobrachem)
- π§ Handeled error in
liesel.model.model.transform
for duplicated nodes (#167, @GianmarcoCallegher) - β¨
lsl.Calc
nodes will now try to evaluate their function upon initialization. This can make it much easier to spot errors early. The behavior can be turned off by passingupdate_on_init=False
. (#92, @jobrachem)
liesel v0.2.8
liesel v0.2.7
- β¨ Large documentation updates! (#149, #146, #145, #143, #135, #134, @jobrachem; #144, #140, @GianmarcoCallegher)
- β¨ The method
gs.Summary.error_df()
is now publicly available. (#137, @jobrachem) - β¨ The class
liesel.goose.engine.SamplingResults
is now exported vialiesel.goose
, which means it can be used asgs.SamplingResults
. (@jobrachem) - β¨ Improved the efficiency of the
liesel.distributions.mvn_degen.MultivariateNormalDegenerate.from_penalty
constructor (#101, @GianmarcoCallegher) - β¨ Added alternative constructor
.from_penalty_smooth
toMultivariateNormalDegenerate
(#133, @GianmarcoCallegher) - π§ Added
observed=True
to apd.DataFrame.groupby()
call ingoose/summary_m.py
to silence a warning due to a deprecation in pandas v2.1.0 - π§ Renamed
lsl.Param
tolsl.param
andlsl.Obs
tolsl.obs
to reflect the fact that those are functions, not classes. The old names are deprecated and scheduled for removal in v0.4.0. (#130, @jobrachem) - π§ Renamed/moved the following symbols. The old names are deprecated and scheduled for removal in v0.4.0. (#136, @jobrachem)
liesel.model.goose.GooseModel
-> (renamed/moved)liesel.goose.interface.LieselInterface
liesel.goose.models.DictModel
-> (renamed/moved)liesel.goose.interface.DictInterface
liesel.goose.models.DataClassModel
-> (renamed/moved)liesel.goose.interface.DataclassInterface
- π§ Some maintenance updates to stay up to date with dependencies (#147, #141, #102, @GianmarcoCallegher; #139, @wiep; #109, @jobrachem)
Edit: The version number was too high in __version__.py
. I updated the release title to fit the version number. Now there is just no liesel v0.2.6 on pypi.
liesel v0.2.5
- π§ Updated for compatibility with Blackjax 1.0.0 (#100, @wiep & @hriebl)
- π§ Updated for compatibility with the latest mypy update (#97, @wiep & @hriebl)
- β¨ Added functionality for easy setup and customization of initial value jittering (#72, @GianmarcoCallegher & @hriebl)
- β¨ Improved error messages in
lsl.Calc.update()
(#84, @jobrachem) - π§ Fixed a bug in
gs.plot_param()
(#81, @viktoriussuwandi) - π§ Fixed an error in the tutorial on linear regression (#85, @jobrachem)
- π§ Fixed the display of the plot title in
gs.plot_scatter()
(#98, @hriebl)
liesel v0.2.4
What's new?
- π§ Removed all references to
jax.numpy.DeviceArray
to make Liesel compatible with Jax 0.4.14 (#73, @jobrachem) - β¨ Added a visual distinction for edges that represent a connection to a variable's distribution or value (#76, @GianmarcoCallegher)
- β¨ Added
ls.Model.simulate()
, which provides a convenient way to draw random samples from a Liesel model using the specified priors. (#70, @hriebl) - β¨ Added
liesel.model.goose.finite_discrete_gibbs_kernel
, which helps you to automatically set up ags.GibbsKernel
for a discrete variable (#64 & #65, @jobrachem and @hriebl) - β¨ Added an intialization message to
gs.Engine
(#66, @GianmarcoCallegher)
liesel v0.2.3
What's new?
- π Improved documentation (#47, #48, @jobrachem, @hriebl)
- β¨ Added convert_dtype to graph builder (#50, @hriebl)
- β¨ New overview page for tutorials (#62, @jobrachem)
- β¨ New tutorial on advanced group usage (#63, @jobrachem)
- β¨ Added a method to convert sampling results to arviz's inference data (#49, @wiep)
- π§ Changes
__repr__
for multiple classes inliesel.model
(#57, @wiep)
liesel 0.2.2
What's new?
- π The tutorials have been updated to v.0.2.2 and are now part of the documentation (@jobrachem, @wiep, @hriebl, @GianmarcoCallegher)
- β¨ Added new tutorial showcasing the interface to PyMC (@GianmarcoCallegher)
- β¨ Added node / variable groups (#28, @jobrachem)
- β¨ Sampling from the
MultivariateNormalDegenerate
is now possible (#34, @jobrachem) - π§ Fixed undefined behaviour in the distreg module (#20, @hriebl)
- π§ The distreg module will now use variables names as position keys (#22, @hriebl)
liesel 0.2.1
What's new?
- π¦’ Updated the experimental PyMC interface for Goose to PyMC 5.0 (@wiep)
- π¦’ Fixed a bug with the column dtype being
object
instead offloat
in the MCMC summary (@wiep)
liesel 0.2.0
For this release, the Liesel modeling library has been rewritten from scratch. We are currently working on updated tutorials explaining the new concepts introduced in v0.2.0 in full detail.
What's new?
- π§ Rewrote the Liesel modeling library from scratch, introducing the
Var
and theGraphBuilder
- π§ Replaced the
SmoothPrior
TFP distribution with the more generalMultivariateNormalDegenerate
- π¦’ Removed deprecated functionality from the Goose summary modules
β οΈ The import paths have changed:import liesel.liesel as lsl
β‘οΈimport liesel.model as lsl
import liesel.tfp.jax.distributions as lsld
β‘οΈimport liesel.distributions as lsld
import liesel.tfp.jax.bijectors as lslb
β‘οΈimport liesel.bijectors as lslb
Contributors
liesel 0.1.4
What's new?
- π We have a new project homepage: https://liesel-project.org
- π Migrated the docs from pdoc to Sphinx: https://docs.liesel-project.org (@jobrachem)
- π¦’ An MCMC summary can now be created with
gs.Summary(results)
instead ofgs.Summary.from_results(results)
(#13, @jobrachem) - π¦’ Fixed a bug with the reported quantiles in the per-chain MCMC summary (#14, @jobrachem)