DynamicPPL.jl

DynamicPPL.jl copied to clipboard

This pull request changes the compat entry for the `Setfield` package from `0.7.1, 0.8` to `0.7.1, 0.8, 1`. This keeps the compat entries for earlier versions. Note: I have not...

github-actions[bot]

Improvements to DynamicPPLBenchmarks

10

Produces results such as can be seen here: https://github.com/TuringLang/DynamicPPL.jl/pull/309#issuecomment-985080224

torfjelde

Linearization/flattening of SimpleVarInfo

10

This PR introduces a couple of things, though these are related: 1. `unflatten(original[, spl], x)`: converts from a certain input `x`, usually a `Vector`, into an instance similar to `original`....

torfjelde

"[DynamicPPL] attempt to link a linked vi" warning when aborting sampling and returning minus infinity

5

I have some code that may have failures and need to reject the sample, which I do with something like ```julia if !(sol.retcode == :Success) @addlogprob! -Inf return end ```...

jlperla

Enhance wrapped distributions

17

Add basic `WrappedDistribution` type for `NoDist` and `NamedDist` and teach them a few tricks like `length()` and `bijector()`. I've discovered that these methods are missing when trying to do ```...

alyst

Probability interface tutorial

7

First addition to the DynamicPPL tutorials; breaking this up as Hong suggested. Goes over how to use the basic interfaces (e.g. logjoint, loglikelihood, logdensityof).

ParadaCarleton

Roadmap for depreciating `VarInfo` in favour of `SimpleVarInfo`

5

This is meant as an umbrella issue for documenting the steps required to depreciate `VarInfo` in favour of `SimpleVarInfo`-equivalwent. Some known items: - [x] Lineralisation for `SimpleVarInfo`: **solved by `VarNamedVector`**...

yebai

Adopting DensityInterface

5

Implements #340 TODO: - [x] Upgrade to AbstractPPL 0.4 as soon as released - [x] Require Distributions 0.25.25 - [ ] Rename `logdensity` to `logdensityof` (deprecating the former), check interface...

phipsgabler

[WIP] More flexibility in RHS of `~`, e.g. MeasureTheory.jl

14

I've recently been thinking a bit about how it would be nice to support more than just `Distribution` from Distributions.jl on RHS of a `~` statement (and other values as...

torfjelde

Improve docs, tutorials and README

13

ParadaCarleton