Mitzi Morris

has this been fixed already? I saw these too, but now everything just works.

regarding the question, does Stan return variables on the unconstrained scale, in CmdStan, sampler argument 'diagnostic_file' is used to report the model variables on the unconstrained scale. because this is...

the underlying mechanism here is calling the Stan model class member function `log_prob` to get the unconstrained parameter values.

right, files are a big problem - happy to discuss possible DASK integrations

agree with Daniel - it needs to work for all legal filepaths on *nix, Mac, and Windows.

does this work with the git commands - `make stan-update`?

I agree with the suggestion that we write a script to compile models - I don't see how you can change the makefile, because there are a bunch of tasks...

note that proposal for CmdStan3 would remove need for this feature: https://github.com/stan-dev/design-docs/blob/cmdstan3/designs/0005-cmdstan3.md

@ahartikainen and @evgenyneu - thoughts? is this feasible?

super cool! but I think what Andrew was asking for was to superimpose *all* chains so that you get the classic "fuzzy catapiller" when things converge - here's an example...