-
On DPPL 0.8.0:
```julia
julia> @model function test5(x)
μ ~ MvNormal(fill(0, 2), 2.0)
z = Vector{Int}(undef, length(x))
z .~ Categorical.(fill([0.5, 0.5], leng…
-
Please see https://github.com/TuringLang/Turing.jl/issues/1235 for details
-
Line https://github.com/TuringLang/DynamicPPL.jl/blob/f74e039cb7fff30a10be882b6e12e9b686dbc77c/src/varinfo.jl#L968 sums over the sizes of the varnames, which fails in case there are none. (You only n…
-
I think the current approach of dealing with states of samplers, e.g., in Turing is flawed. Currently, one defines an initial state when initializing the sampler. This is problematic since often the c…
-
This test (https://github.com/TuringLang/DynamicPPL.jl/blob/master/test/compiler.jl#L333) fails with Turing master. I am not sure why.
-
I spotted this typo in the docs: https://turing.ml/dev/tutorials/6-infinitemixturemodel/
Here's the snippet:
> , we found another issue with the model reported below. On Turing v0.8.0, the model fails when t…
-
Hi guys,
I am using the following package versions (updated to the current DistributionsAD module after last closed issue):
```
[76274a88] Bijectors v0.8.0
[31c24e10] Distributions v…