Hong Ge
Hong Ge
Let's add experimental support for `BUGS` syntax via a package extension depending on `JuliaBUGS`.
This PR makes it easy to run each test file individually locally: ```julia shell> pwd /Users/hg344/projects/Turing.jl (test) pkg> activate test Activating project at `~/projects/Turing.jl/test` julia> using Turing; include(pkgdir(Turing)*"/test/test_utils/test_utils.jl") julia> include("test/modes/OptimInterface.jl")...
The current implementation of the Gibbs sampler is a bit fragile and based on the out-of-date design for `AbstractVarInfo`. As part of the transition into a more intuitive, modular and...
After #375, we have most Turing docs in this repo. We might want to consider renaming this repo to something more generic.
Hidden Markov Models are quite common in time series analysis. Since they involve discrete variables, HMC is not always appropriate, although we can still apply HMC for continuous parameters of...
For efficiency reasons, there are cases where we want to handle model parameters differently in logdensity and sampling functions. This might be of general interest to Turing users, so would...
The current figure for MCMC on the landing page looks like: where MCMC sampling is depicted by loss surface with scattering points. This is not very representative of MCMC IMO....
We have debated for length about how to handle returned values from a DynamicPPL model, and use that as a way to store extra non-parameter model variables (e.g. transformed variables...
The following model will ignore the fact `v` is a derived variable from data and treat `v` as a random variable. This is a known limitation of `DynamicPPL`. I am...