Seth Axen

If #67 is merged, we could add rules for `*(::Plan, ::StridedArray)`, so long as the plan is `Const` (if it's non-`Const`, then we would need the rule to support it...

I've paused work on this until EnzymeTestUtils (https://github.com/EnzymeAD/Enzyme.jl/pull/782) is registered, which will make testing these rules reliably much more straightforward.

Coming back to this, I think Enzyme rules should only be defined here abstractly for cases where we know they will not be breaking downstream code that otherwise Enzyme would...

> Any hope of having this PR revived? Enzyme has come a long way lately and FFT support would be another great step forward. I'm afraid I don't have the...

See also https://github.com/StanJulia/StanSample.jl/issues/60 for discussions on letting StanSample implement its own converter. We could then deprecate the `from_cmdstan` converter and point users to that.

@cscherrer it likewise would be nice to move the [conversion to `InferenceData` from `SampleChains`](https://github.com/arviz-devs/ArviZ.jl/blob/main/src/samplechains.jl) out of this package. This could either live in SampleChains/SampleChainsDynamicHMC via (conditional) dependence or in a...

Thanks @ErikQQY for pushing this forward! I'm pretty sure this implementation would work with Pathfinder (though we would probably need the expectations of the `factorization` field to be documented to...

On a semi-related note, this rank-update covariance is the same as the marginal covariance of latent factor models of the following structure: ```math \begin{aligned} z &\sim \mathcal{N}(0, D)\\ y &\sim...

If I comment out this line then all the warnings go away: https://github.com/rafaqz/DimensionalData.jl/blob/87a0b40fd2d4fa541161c729dd0a30ddc5cf9d88/src/DimensionalData.jl#L120

Yeah, so currently we already evaluate the target log-density for: - All draws that are recycled from the ELBO calculation - All draws returned by `multipathfinder` with `importance=True` But we...