Cameron Pfiffer

I'm beginning to not be very fond of `AxisArrays`. It's not that good a data format for our purposes.

They do, but AxisArrays are kind of a non-starter for long-term development. They basically require all variables share a type, so including integer and continuous variables tends to make arrays...

I think we used to have this in the super old `MCMCChains` but it was removed by assuming that it would be added as a parameter if it was important....

Yeah, that would actually be quite nice -- tracking them separately would be a lot better than this weirdo group map thing we're doing internally.

Probably 4.16.0 since #310 is a bigger thing and probably won't have too much effect here.

> Hello! I am attempting to use MCMCChains to help with chains of HMMs optimized by EM. Performing the heideldiag tests on the autotransition probabilities is a reasonable way to...

Dunno, I'm kind of the wrong person to ask -- I don't know much about licensing issues.

I kinda love it and would be happy to see this in the docs. I think also a glue-on package that handles some common interactions between MCMCChains and Makie/AlgebraOfGraphics would...

David's got good points here. There are _a lot_ of far better ways to do what we currently do, but switching over to them is the hard bit. We could...

You should be able to generate a separate chain with the generated quantities and then just call `hcat(chain1, chain2)`, which (I hope) will smush them together.