v0.24.1

Turing v0.24.1

Diff since v0.24.0

Closed issues:

• Function to check HMC diagnostics (#1524)
• PG doesn't work multivariate distributions (#1592)
• Support MCMCTempering functionality (#1624)
• Reverse-mode AD extremely slow for large number of observations (#1642)
• Unused variable in benchmark (#1652)
• Linear SDEs (#1814)
• Precompile fails with "UndefVarError: NodeType not defined" (#1941)
• turing.ml isn't responding (#1952)
• turing.ml can not be reached currently! (#1953)
• UndefVarError: GLOBAL_RNG from Bijectors.jl (#1955)
• Cannot load / read chains from disk UndefVarError: ForwardDiffLogDensity (#1956)

Merged pull requests:

• CompatHelper: bump compat for ForwardDiff to 0.10 for package test, (keep existing compat) (#1908) (@github-actions[bot])
• Fix test errors with LogDensityProblemsAD >= 1.2 (#1942) (@devmotion)
• Update TuringLang homepage url (#1954) (@jamesmaino)
• CompatHelper: bump compat for MCMCChains to 6, (keep existing compat) (#1959) (@github-actions[bot])
• CompatHelper: bump compat for MCMCChains to 6 for package test, (keep existing compat) (#1960) (@github-actions[bot])

v0.24.0

Turing v0.24.0

Diff since v0.23.3

Closed issues:

• Changing Sampled Parameter Type? (#1922)
• The package dependency on Setfield is old (v0.8.2), which makes conflict with other packages. (#1935)

Merged pull requests:

• Reduce usage of sampler (#1936) (@torfjelde)
• Cleanup benchmarks (#1937) (@yebai)

v0.23.3

Turing v0.23.3

Diff since v0.23.2

Merged pull requests:

• CompatHelper: bump compat for AdvancedHMC to 0.4, (keep existing compat) (#1926) (@github-actions[bot])
• Fix expectation of s² and m in gdemo (#1927) (@rdiaz02)
• Fix error in example and add simpler example (#1928) (@rdiaz02)
• CompatHelper: bump compat for AdvancedHMC to 0.4 and for AdvancedMH to 0.7, (keep existing compat) (#1929) (@github-actions[bot])
• CompatHelper: bump compat for AdvancedMH to 0.7 for package test, (keep existing compat) (#1930) (@github-actions[bot])
• Fix test failure (#1932) (@devmotion)

v0.23.2

Turing v0.23.2

Diff since v0.23.1

Closed issues:

• Posterior Predictive Checks (#1809)
• StableRNG in Turing is not yielding reproducible output (#1923)

Merged pull requests:

• Fix non-reproducible step sizes (#1924) (@devmotion)

v0.23.1

Turing v0.23.1

Diff since v0.23.0

Closed issues:

• @model scope and influence from global variables (#1915)

Merged pull requests:

• CompatHelper: bump compat for LogDensityProblems to 2, (keep existing compat) (#1917) (@github-actions[bot])
• Fix tests for Gibbs (#1920) (@yebai)

v0.23.0

Turing v0.23.0

Diff since v0.22.0

Merged pull requests:

• Match AdvancedPS 0.4 API (#1858) (@FredericWantiez)
• Attempt to tighten assertion accurary bounds by increasing MCMC samples (#1905) (@yebai)
• Variational Inference docs fix (#1912) (@v-i-s-h)

v0.22.0

Turing v0.22.0

Diff since v0.21.13

Closed issues:

• [BNP] Bayesian Nonparametrics project (#370)
• [BNP] Q-class (#374)
• [MH] change of variable (#376)
• [Traces]: add a member field to Trace to mark reference particle (#396)
• More robust mechanism for mapping model variables to samplers (#397)
• Particle Gibbs does not return log probability (#643)
• Improvement of sampler descriptions. (wip) (#699)
• Test for RandomMeasures fail (#973)
• Update benchmarks on wiki (#1006)
• Sanity check is missing for space of Gibbs sampler and model parameter space (#1012)
• Handle gradient = nothing case when doing reverse diff with Zygote (#1081)
• Failure when dimension changes (#1115)
• Simple benchmarking for Tracker, ReverseDiff, and Zygote (#1140)
• initial parameters / NUTS step size (#1314)
• Integrating out variables via Laplace approximation (or other methods) (#1382)
• Add instructions to check the gradient function (#1383)
• Precompilation fails while using ReverseDiff backend (#1400)
• RFC: Sampler-specific AD settings instead of global AD settings (#1402)
• rdcache not cleared while using MLE,MAP estimation (#1418)
• Coeftable throws error on mle and map estimates (#1447)
• GibbsConditionals: proper variable subsumption (#1452)
• Wrong interpolation symbol? (#1551)
• Sampler initialization (init_params) (#1563)
• Some tests for MH fail for certain random seeds (#1587)
• The chains do not converge to the same posterior (#1593)
• Numerical error messages: real problem or just terminal noise? (#1621)
• Feature request: Add LKJCholesky (#1629)
• Sometimes this error is thrown: Cannot convert an object of type Geometric{Float64} to an object of type Dirac{Float64} (#1669)
• Improper (?) chains resuming (#1697)
• AbstractMCMC.step with different model at each step (#1699)
• Stein Thinning (#1704)
• Question: looks like the order of variables are inconsistent (#1712)
• Bijector error with VI and arraydist/filldist (#1717)
• Easy way to get gradient evaluation timing (#1721)
• Automatic lazy broadcasting/optimization for arrays of distributions (#1723)
• Invalid test case for Gibbs? (#1725)
• Refactoring Glue Code (#1735)
• Way to Declare Data (#1736)
• Improve usage and support for immutable AbstractVarInfo (#1752)
• Batch size reduced to zero when doing inference on ODEs (#1760)
• Massive test time regressions (#1774)
• Broken doctests (#1796)
• there is an erro in Variational inference (VI) of the Tutorials (#1820)
• Concrete terse structs (#1830)
• Question: Which package is the MCMCserial() part of? been working through the notebook and have had problems with it not being defined hence that part of the notebook does not run (#1842)
• Custom distribution sampling does not perform well (#1859)
• Proposal: Interest in faster logpdf for BernoulliLogit (logistic regression) (#1890)

Merged pull requests:

• Compatibility with new DPPL version (#1900) (@torfjelde)
• CompatHelper: bump compat for Setfield to 1, (keep existing compat) (#1906) (@github-actions[bot])
• Bump minor version (#1909) (@torfjelde)

v0.21.13

Turing v0.21.13

Diff since v0.21.12

Closed issues:

• Hessian AD for logjoint and second-order optim interface broken (#1878)
• Error with bunchkaufman factorization (#1889)

Merged pull requests:

• Fix LogPoisson typo in docstring (#1888) (@storopoli)
• Conditionally remove BernoulliLogit (#1892) (@devmotion)

v0.21.12

Turing v0.21.12

Diff since v0.21.11

Merged pull requests:

• vi should not work with Matrixvariate distributions (#1545) (@torfjelde)
• Support LogDensityProblems 1 (#1883) (@devmotion)

v0.21.11

Turing v0.21.11

Diff since v0.21.10

Closed issues:

• Getting the preconditioner/checking that it is functional (#1860)
• Libtask error when using PG (#1873)
• VarInfo internal error with MH (#1874)
• Console logging of sampling broken (#1875)

Merged pull requests:

• Use LogDensityProblems instead of gradient_logp (#1877) (@devmotion)
• Update gibbs.jl (#1879) (@yebai)
• Remove outdated and unused Stan interface and tests (#1880) (@devmotion)