Kyurae Kim
Kyurae Kim
I'll close this since this seems to have been resolved.
Related issues in [Bijectors.jl](https://github.com/TuringLang/Bijectors.jl/issues/252) and [Turing.jl](https://github.com/TuringLang/Turing.jl/issues/1988)
@torfjelde Is there a reason we compute the first element first and then use that to initialize `mapreduce` in `Stacked`?
What do you mean by track? Was there any plan to include an RJMCMC sampler at any point?
As long as you can infer which variable is the model indicator and the dimensionality of the parameters can change, one could, in principle, implement such a backend for Turing....
Hi, [This issue](https://github.com/TuringLang/Bijectors.jl/issues/252) is possibly related. I've just tried the quick-and-dirty fix I posted there, but I now get a different error: ```julia julia> @eval Bijectors begin function with_logabsdet_jacobian(sb::Stacked, x::AbstractVector)...
@hdelecki Sorry for the late response. I'm not an AD expert either, but it does seen that `mapreduce` is still broken. Asking for help! @devmotion @torfjelde
Related [discussion](https://github.com/tpapp/LogDensityProblems.jl/issues/106) in `LogDensityProblems.jl`
Hi @torfjelde > Maybe we should make this into a discussion. I feel like there are several different approaches we can take here. Should we proceed here or create a...
@torfjelde Hi, I have significantly changed the sketch for the project structure. 1. As you previously suggested, the `ELBO` objective is now formed in a modular way. 2. I've also...