Sam

So if it is, should be fixed by https://github.com/TuringLang/Turing.jl/pull/1545.

Thanks, that workaround will be very helpful. I'm fitting a large number of models, and while the first few "step size found" notifications are useful, they get less interesting after...

Silencing warnings by default during during tuning is a good idea. The running tally and summary message is also a good idea; R prints out `There were 50 or more...

Ok, I wasn't totally sure about using `cross` for the matrix-multiplication case myself. The use case I'm picturing is something like the second model in the readme (with one latent...

Just ran into this issue with `::Type{IrrationalConstants.Log2π}`. Is there any reason not to add `AbstractIrrational` to `REAL_TYPES`? Happy to make a PR.

Delaunator.jl already only depends on base Julia, though DelaunayTriangulation.jl seems to be more fully-featured and actively developed at the moment. If you're hesitant to include it as a direct dependency...

Sure, that sounds good. Assuming both packages are loaded, this is all it takes: ```julia function SimpleMesh(tri::Triangulation) points = Point2.(eachcol(tri.points)) triangles = connect.(tri.triangles, Triangle) return SimpleMesh(points, triangles) end ```

I've started working on a PR and was wondering about what the return type for `cov(chns)` should be. A 2D `AxisArray` would seem to be a good choice. However, that's...