Seth Axen

It gets better. Because DynamicHMC's API supports all initialization approaches we might use Pathfinder for, and because SampleChainsDynamicHMC exposes those configuration options untouched, we will be able to do this,...

> It will be really great to get it working with Soss and Tilde. Thanks for putting this together! You're welcome! Currently it returns the variational approximation and draws only...

@cscherrer it seems `SuperpositionMeasure` does not have a `rand` method: ```julia julia> μ = Dirichlet(10, 1.0) Dirichlet(α = Fill(1.0, 10),) julia> result = multipathfinder(μ, 1_000; nruns=20) ┌ Warning: Pareto shape...

Pausing development of this until Soss works with HMC again, see https://github.com/cscherrer/Soss.jl/issues/337

I believe that already happened with GalacticOptim v3. e.g. we now depend on GalacticOptim and GalacticOptimJL, the latter of which brings the Optim integration.

> And that's still too heavy thinking, interesting. Yes, here's what we get for GalacticOptim by itself: ```julia julia> @time_imports using GalacticOptim 11.0 ms ┌ MacroTools 18.6 ms ┌ ZygoteRules...

https://github.com/SciML/GalacticOptim.jl/issues/239

Relevant issue: https://github.com/SciML/Optimization.jl/issues/556

Coming back to this, Optimization/SciMLBase use `OptimizationFunction` purely to specify the function and derivatives. There's absolutely no information contained there to indicate whether this function will ultimately be maximized or...

Now that [PosteriorDB.jl](https://github.com/sethaxen/PosteriorDB.jl) has been registered, and BridgeStan is included in [StanSample.jl](https://github.com/StanJulia/StanSample.jl), it should be quite straightforward to set up a Julia script that generates samples from posteriordb models using...