v0.3.2

MCMCTempering v0.3.2

Diff since v0.3.1

Closed issues:

• Leverage existing ways in TuringLang (if they exist) / Julia ecosystem to approximate K in the case of Simulated Tempering (#3)
• Is there any reason to return all of the PT chains, or should only the one where beta = 1 be returned in all cases? (#20)
• Solve the problem of sampler specific types obscuring access to the VarInfo (#63)
• Figure out a consistent way to return swap history and other chains if desired (#123)
• Extend the suite of tests to match level of coverage in Turing (#124)
• Support multiple samplers (#148)
• Support passing a chain type and returning in line with it from tempered_sample (#149)

Merged pull requests:

• Renaming of some fields (#142) (@torfjelde)
• Implementing and correcting the base set of swap strategies, plus wrapping swap moves as part of steps (#143) (@HarrisonWilde)
• Introduction of compositions and products of samplers and models (#151) (@torfjelde)
• Introduction of SwapSampler + make TemperedSampler a fancy version of CompositionSampler (#152) (@torfjelde)
• Documentation (#153) (@torfjelde)
• Use process_to_chain in roundtrips instead (#154) (@torfjelde)
• Improvements to state_from and more extensive docs (#155) (@torfjelde)

v0.3.1

MCMCTempering v0.3.1

Diff since v0.3.0

Merged pull requests:

• Fixed significant bugs (#141) (@torfjelde)

v0.3.0

MCMCTempering v0.3.0

Diff since v0.2.0

Merged pull requests:

• Support for LogDensityProblems.jl (#139) (@torfjelde)

v0.2.0

MCMCTempering v0.2.0

Diff since v0.1.1

Closed issues:

• Implement online adaptation of the temperature schedule (#69)
• Add Disclaimer in README that this repo is non-functional (#135)

Merged pull requests:

• Quality of life improvements to docs and code (#129) (@HarrisonWilde)
• Implemented adaptation per Miasojedow 2013 (#131) (@HarrisonWilde)
• Adding missed change (#132) (@HarrisonWilde)
• General improvements and fixes (#133) (@torfjelde)
• add warning about package status (#136) (@itsdfish)
• Added a tempered_sample helper function and no-swap burn-in functionality (#137) (@HarrisonWilde)
• CompatHelper: add new compat entry for ProgressLogging at version 0.1, (keep existing compat) (#138) (@github-actions[bot])

v0.1.1

MCMCTempering v0.1.1

Diff since v0.1.0

Closed issues:

• Register the package (#97)
• Remove unicode characters from the API function names (#127)

Merged pull requests:

• Change api function names (#128) (@HarrisonWilde)

v0.1.0

MCMCTempering v0.1.0

Closed issues:

• Package name discussions (#1)
• How best to generalise the swap_acceptance_x functions for each algorithm (#2)
• Should I be worried about Base.Docs warnings when pre-compiling a package for the first time (#4)
• How best to build the tempered model for a step of the tempered algorithms (#5)
• Create temperature schedule plot templates to aid debugging (#6)
• Best way to parallelise parallel tempering (#7)
• Reconstruct chains correctly from parallel tempering (#19)
• Implement Non-Rev Rev sampling strategy for PT (#21)
• index temperature by integer (#25)
• Avoid evaluating the same logdensity twice (#42)
• Propagate the Selector through the DynamicPPL.Samplers in order to use the Sampler to index the VI and return Theta per step (#58)
• Store the swap history (#59)
• Plot swap history helper function (#61)
• Enable parallelism (#64)
• Create a minimal function-based API for greater general applicability of this package (#68)

Merged pull requests:

• Selector and store swap history (#60) (@HarrisonWilde)
• Closes #61 (#62) (@HarrisonWilde)
• Closes #64 enabling MCMCThreads style parallelism across chains (#65) (@HarrisonWilde)
• Working general API for usage with Turing, AHMC etc. (#122) (@HarrisonWilde)