Seth Axen

> BTW all three ESS implementations use the same ESS algorithm from the split-rhat paper, really the only difference is how the autocorrelation terms are computed. Which split-rhat paper? If...

On the Python side, ArviZ does indeed have reloo for some PPLs. It's implementation is PPL-specific, one approach per supported PPL. ArviZ.jl hasn't added anything like this for Julia PPLs...

I haven't looked at the CV implementations in MLJ, but LOO doesn't actually do CV. It does approximate CV using importance sampling (hence the PSIS implementation). I'd guess MLJ is...

Would it make sense to put this script in a lightweight package? Then I could easily install the package and call the downgrade function providing a local path to a...

Agreed, this would be useful to have.

I think something like this would be sufficient: ```julia _quad(d::AbstractMvNormal, t::AbstractVector) = dot(t, cov(d), t) _quad(d::MvNormal, t::AbstractVector) = PDMats.quad(cov(d), t) _quad(d::MvNormalCanon, t::AbstractVector) = PDMats.invquad(d.J, t) function cf(d::AbstractMvNormal, t::AbstractVector) return exp(im...

> StanSample.jl v7.5.0 contains a fix for this issue. Thanks! Indeed, it works for me! > You've probably seen [Brian's suggestion](https://github.com/StanJulia/StanSample.jl/issues/74#issue-1979350826) to move Stan related I/O to a separate package....

The error when calling `DataFrame` on a `Dataset` is unrelated to InferenceObjects and can be reproduced in DimensionalData itself. I've opened an issue: https://github.com/rafaqz/DimensionalData.jl/issues/459

There's no plan for it. TBH I haven't used FileIO much and am not certain the best way to support it, but I would be happy to review a PR.

I probably will not be able to look into this again for several weeks, but I think this is because in an MRC file data is written row-major, while Julia...