Jonah Gabry

Thanks for pointing that out. I think for now the easiest way to get the MAD for _all_ parameters is just by computing it from the posterior draws. You could...

Good question, I'm not sure, it's been so long! I'll have to take a look at the source code and refresh my memory when I have a chance. I don't...

Thanks for the heads up. A separate package for the formula-processing stuff makes sense. A pull request would be great, thank you!

Thank you for this! I'm super busy the rest of this week unfortunately, so I may not have time to comment for a little while. But I'll come back to...

Actually, it just occurred to me that there's already `posterior::split_chains()`. So something like this should already work with bayesplot: ``` x % as_draws_array() %>% # not necessary if already a...

So maybe we should add an example of doing this using `posterior::split_chains` to the doc instead of changing the bayesplot function? Or alternatively we could just call `posterior::split_chains` internally to...

I agree with an argument that defaults to `FALSE`. I think given the way the bayesplot internals are written (they were written mostly before posterior was developed), in order to...

Thanks @kruschke, I agree we should provide an option to for using other functions for averaging. Thanks @tjmahr for the PR. Will review soon.

Thanks! In the rstantools package we typically implement generic and default methods that don't take fitted model objects as input. So I think if this uses fitted model objects it...

I didn't realize there would be this sort of difficulty with the dot plots. I think handling those edge cases and having appropriate warnings is probably enough of a benefit...