JuliaBUGS-v0.13.0

JuliaBUGS JuliaBUGS-v0.13.0

Diff since JuliaBUGS-v0.12.3

Breaking changes

• Bump AbstractPPL to 0.14

Merged pull requests:

• Bump AbstractPPL to 0.14, release v0.13.0 (#422) (@github-actions[bot])
• DoodleStan: Generate Stan Code for DoodleBUGS graph (#431) (@shravanngoswamii)
• fix Dependabot alerts (#444) (@shravanngoswamii)

Closed issues:

• Imcompatibility of the latest DynamicPPL and JuliaBUGS (#442)

JuliaBUGS-v0.12.3

JuliaBUGS JuliaBUGS-v0.12.3

Diff since JuliaBUGS-v0.12.2

Merged pull requests:

• CompatHelper: bump compat for MetaGraphsNext to 0.8 for package JuliaBUGS, (keep existing compat) (#425) (@github-actions[bot])
• fix serializing BUGSmodel and add tests (#435) (@hardik-xi11)
• CI: consolidate test matrix, fix benchmark fork permissions, misc cleanup (#438) (@sunxd3)
• Add codecov.yml for multi-upload coverage (#440) (@sunxd3)
• Release v0.12.3 (#441) (@sunxd3)

Closed issues:

• Distributed AbstractMCMC.sample fails (#333)
• CI improvements: consolidate test matrix, fix benchmark fork permissions, misc cleanup (#437)

JuliaBUGS-v0.12.2

JuliaBUGS JuliaBUGS-v0.12.2

Diff since JuliaBUGS-v0.12.1

Merged pull requests:

• fix dependabot security alert (#424) (@shravanngoswamii)
• improve DoodleWidget (#426) (@shravanngoswamii)
• Tons of Improvements and reduce code duplication across DoodleBUGS and DoodleWidget (#428) (@shravanngoswamii)
• fix maths and inline latex (#430) (@shravanngoswamii)
• fix security issues -- Dependabot Alerts (#432) (@shravanngoswamii)
• Catch DomainError in logdensity to prevent HMC crashes (#434) (@sunxd3)

Closed issues:

• math on docs page not rendering correctly (#429)
• DomainError during leapfrog integration (#433)

JuliaBUGS-v0.12.1

JuliaBUGS JuliaBUGS-v0.12.1

Diff since JuliaBUGS-v0.12.0

Add support for TensorBoardLogger.jl via AbstractMCMC.mcmc_callbaks. See here.

Merged pull requests:

• Add AbstractMCMC Callbacks support (#423) (@shravanngoswamii)

JuliaBUGS-v0.12.0

JuliaBUGS JuliaBUGS-v0.12.0

Diff since JuliaBUGS-v0.11.1

Breaking changes

This release brings major improvements to gradient-based inference with DifferentiationInterface.jl replacing LogDensityProblemsAD, and automatic marginalization of discrete parameters.

Highlights

• DifferentiationInterface.jl for gradient computation (#397): Gradient computation now uses DifferentiationInterface.jl instead of LogDensityProblemsAD. Pass an adtype parameter to compile() to get a gradient-enabled model.

  • Example: model = compile(model_def, data; adtype=AutoReverseDiff())
  • Supports AutoReverseDiff, AutoForwardDiff, AutoMooncake

• Auto-marginalization for discrete parameters (#385): Automatically marginalize discrete latent variables to enable gradient-based inference on models with discrete parameters.

  • Example: model = set_evaluation_mode(settrans(compile(model_def, data), true), UseAutoMarginalization())
  • Supports models where discrete parameters have finite support (e.g., Categorical, Bernoulli)

• On-demand log density function generation (#416): Log density functions are now generated on-demand when set_evaluation_mode(model, UseGeneratedLogDensityFunction()) is called, rather than at compile time. All models start with UseGraph() mode.

Breaking Changes

• LogDensityProblemsAD.ADgradient is no longer supported. Use compile(...; adtype=...) or BUGSModelWithGradient(model, adtype) instead.

Migration Notes

With on-demand generation (#416), some internal options from 0.10.x are no longer needed:

• skip_source_generation parameter removed from compile() - no longer necessary since functions are generated lazily.
• regenerate_log_density kwarg removed from condition() - conditioned models start in UseGraph mode; call set_evaluation_mode(model, UseGeneratedLogDensityFunction()) when needed.
• regenerate_log_density_function() removed - use set_evaluation_mode(model, UseGeneratedLogDensityFunction()) instead.

Improvements

• Expanded support for generated log density functions via dependence vectors (#390)
• Julia 1.12 compatibility improvements (#404)

Merged pull requests:

• Auto-marginalization discrete model parameters (#385) (@sunxd3)
• Integrate DifferentiationInterface.jl for gradient computation (#397) (@shravanngoswamii)
• Remove macOS test workflow (#414) (@sunxd3)
• Simplify benchmark logic and output (#415) (@sunxd3)
• Make log density function generation on-demand (#416) (@sunxd3)
• Prepare for release 0.12 (#417) (@sunxd3)
• DoodleBUGS: Build as Library for Integration in Any Web App (#418) (@shravanngoswamii)
• Reorganize documentation structure (#420) (@sunxd3)
• Fix UseGeneratedLogDensityFunction to support @bugs_primitive functions (#421) (@sunxd3)

Closed issues:

• Add Aqua tests (#285)
• Extend Code Generation to Support Models with Cyclic Dependencies (e.g., Time Series) (#290)
• Remove type piracy methods (#312)
• Simplify the workflow for computing model gradients (#380)
• chain_type argument is not respected in sample calls (#381)

JuliaBUGS-v0.11.1

JuliaBUGS JuliaBUGS-v0.11.1

Diff since JuliaBUGS-v0.11.0

Merged pull requests:

• DoodleBUGS: Make zooming smooth (#360) (#384) (@Govindggupta)
• Add DoodleBUGS PR Preview cleanup workflow (#387) (@Copilot)
• Use dependence vectors to support generating Julia code for more models (#390) (@sunxd3)
• DoodleBUGS: Implement Undo/Redo functionality (#398) (@kundangosavii)
• CompatHelper: bump compat for JSON to 1 for package JuliaBUGS, (keep existing compat) (#399) (@github-actions[bot])
• Use invokelatest in tests (because of Julia 1.12) (#404) (@sunxd3)
• DoodleBUGS: fix Dependabot security alerts (#407) (@shravanngoswamii)
• DoodleBUGS: Use PrimeVue for UI & Implement “Figma-like” Layout & Model Sharing Feature (#408) (@shravanngoswamii)
• DoodleBUGS: fix sharing model with validation errors (#410) (@shravanngoswamii)
• DoodleBUGS: Remove node only when Alt/Option (⌥) is pressed (#413) (@shravanngoswamii)

Closed issues:

• Compile to JAX to enable GPU/TPU acceleration and vmap. (#209)
• DoodleBUGS: Add GHA for PR Preview cleanup (#342)
• DoodleBUGS: Backend (#355)
• DoodleBUGS: Allow plate resizing (#358)
• DoodleBUGS: Undo/Redo Functionality (#359)
• DoodleBUGS: Make zooming smooth (#360)
• JuliaBUGS: Plates not supported in plotting (#412)

JuliaBUGS-v0.11.0

JuliaBUGS JuliaBUGS-v0.11.0

Diff since JuliaBUGS-v0.10.5

v0.11.0 Breaking Changes

Simplified sampling API in AbstractPPL.evaluate!!: Replaced sample_all and respect_observed (added in 0.10.5) kwargs with sample_observed.

Migration:

• From 0.10.5: Default behavior unchanged (samples latents, keeps observed fixed)
  • sample_all=true, respect_observed=false → sample_observed=true
  • Other uses → remove (now default)
• From 0.10.4 or earlier: Default behavior changed
  • Old default sampled all nodes including observed
  • New default keeps observed fixed; use sample_observed=true for old behavior

Merged pull requests:

• Simplify sampling API: replace sample_all and respect_observed with sample_observed (#406) (@sunxd3)

JuliaBUGS-v0.10.5

JuliaBUGS JuliaBUGS-v0.10.5

Diff since JuliaBUGS-v0.10.4

Merged pull requests:

• Add a kwarg to allow respecting observed values when sampling with evaluate_with_rng!! (#405) (@sunxd3)

JuliaBUGS-v0.10.4

JuliaBUGS JuliaBUGS-v0.10.4

Diff since JuliaBUGS-v0.10.3

Merged pull requests:

• DoodleBUGS: Now works in WebKit broswers (Safari, all iPad/iOS browsers) (#402) (@shravanngoswamii)
• Allow disabling source generation during compilation (#403) (@sunxd3)

Closed issues:

• DoodleBUGS: Make it work for iPad (#350)
• example 6 does not compile (#400)

JuliaBUGS-v0.10.3

JuliaBUGS JuliaBUGS-v0.10.3

Diff since JuliaBUGS-v0.10.2

Merged pull requests:

• Formatting some code and adjust formatting configs (#396) (@sunxd3)