David Widmann

We are about to unify `PDMat` and `PDSparseMat` in #188. This will already introduce a `F

Models can be serialized, we even have a test for it: https://github.com/TuringLang/DynamicPPL.jl/blob/138bd40acdfc47d7b00e25a2adaf9fec986f9646/test/serialization.jl The serialization issues in DynamicPPL should have been fixed by https://github.com/TuringLang/DynamicPPL.jl/pull/134. I haven't checked the MWE above but...

Serialization should work (and is tested) - can you try if `Serialization.serialize` and `Serialization.deserialize` works for you?

`mymodel` is a regular Julia function, so it suffers from the same limitations regarding (de)serializations as any other Julia function, whereas `fit` is an object of type `DynamicPPL.Model` and behaves...

The example above works fine for me, I don't get this error. Did you try a more complicated example? My setup: ```julia (jl_L5QqA0) pkg> st Status `/private/var/folders/n6/98_7bm0j0hb57zv3l3tj8sxh0000gn/T/jl_L5QqA0/Project.toml` ⌃ [fce5fe82] Turing...

My gut feeling is that right now an `AbstractTransition` has only limited use since all these transitions have different fields and even different field names for the samples (e.g., different...

IIRC I moved a few of them (quite a while ago), and my impression has always been that they should be moved if they are of more general interest.

> F2F @SamuelBrand1 and I have discussed that the SciMl ecosystem has had to develop their own faster Poisson (presumably due to issues changing it in Distributions). That's basically a...

Issue in Distributions for the problem here: https://github.com/JuliaStats/Distributions.jl/issues/821 We should hopefully be able to fix it when overhauling `rand` etc. more generally. But my gut feeling is that it's somewhat...

The error was fixed (probably by https://github.com/JuliaSymbolics/Symbolics.jl/pull/1384) but the analysis seems to be incorrect: ```julia julia> Symbolics.isaffine(expr - Dx, [x]) false julia> Symbolics.islinear(expr - Dx, [x]) false ``` even though...