Brian Ward

I accidentally hit 'Close with comment' instead of 'Comment' which triggered all of the above.

Vectorized versions of `abs`/`fabs` do not currently work for complex containers, @bob-carpenter thinks he might know why

A lot of these were resolved by #2753 and will be exposed in the language by https://github.com/stan-dev/stanc3/pull/1212. I believe the remaining all need their templates broadened, which I hope to...

Of the remaining ones in this issue, `inverse` and `mdivide_left` are just issues with using `std::is_arithmetic`. Adding an identical overload which uses `stan::is_complex` works, but autodiff is slow. For `determinant`,...

This has been implemented for real-valued matrices (see https://github.com/stan-dev/stanc3/pull/1192), but is still needed for when the input matrix is itself complex

+1 to `--disable-new-dtags` needing to change, we also are patching it on the Conda version of cmdstan

Thank you, it would be very helpful. I think I could rig something using some of the older code in the cmdstanpy io portion if you aren't able to find...

I think repository wikis or in make/local.example (or both) would be a good place for this kind of information

I think 6 is actually a better default value, but the issue is it is not documented. More discussion here: https://discourse.mc-stan.org/t/cmdstanpy-error-in-fit-summary/26939/13?u=wardbrian Furthermore, as is pointed out earlier in that thread,...

Thanks for looking into this @andrjohns. I guess the downside to this approach is losing the ability to 'batch' calls?