-
@Vaibhavdixit02 can you take a look at it?
@mohamed82008 is there any ways to make this more stable? It seems like Turing breaks their interface every month and this is really hard to keep up with.
-
Platform: Windows 10, Julia 1.1.0
running the following produces an error:
`add Turing`
`test Turing`
.... lines about things being updated
> [bbc10e6e] + DynamicHMC v1.0.2+ #master (https…
-
[ Info: Precompiling Turing [fce5fe82-541a-59a6-adf8-730c64b5f9a0]
Building Libtask → `/opt/julia/packages/packages/Libtask/RjRkK/deps/build.log`
ERROR: LoadError: LoadError: UndefVarError: libta…
-
Currently, if `b isa Scale` then `inv(b)` results in `Scale(inv(b.a))`. After @willtebbutt educating me in performance and stability of matrix-inversion, it seems like it would be a better idea to do:…
-
Should we move ownership to the GPMLj organisation? I think that would make a lot of sense but I am not sure how the collaborator thingy would go?
-
I just updated to julia 1.3. Added Turing with no issue, but
using Turing
leads to the following error (this is on linux, but I received the same error earlier on a different machine running windo…
-
Hi,
Thanks for putting this work out in the open. I just recently finished McElreath's excellent book and was attempting to translate some of the models over to Julia. I am having a ton of trouble …
-
High dimensional Gaussian in Turing is slow (> 1 hour):
```julia
using Turing
@model high_dim_gauss(D) = begin
m ~ MvNormal(fill(0, D), 1)
end
chain = sample(high_dim_gauss(10_000), HMC(…
-
Turing's HMC sampler currently logs its progress in the terminal. While this is nice in a pure REPL, inside Juno it'd be even better to use the progress API. This has a number of nice features, includ…
-
Hi-
I noticed that `autocorplot` no longer works. Here is a MWE:
```
using MCMCChains, StatsPlots
val = rand(500,5, 2)
chn = Chains(val, ["a", "b", "c", "d", "e"])
autocorplot(chn)
```…