I think we want to treat disk thickness as a nuisance parameter to fit for, marginalizing over some prior on the distribution.
I've tried to be generally agnostic to the number of fit parameters in the code, but changing the "pars" array will probably require changes in a few different places, as well as the "guess" and "fixed" arrays. Might want to add a general method to define the fit parameters so that models can be swapped in and out in a single spot.
Thick disks will also need some thought when drawing intensity and velocity maps. Let's start by assuming they are optically thin and rotating as cylinders. Thickness can be treated as an exponential scale height (or use Sersic?), and the flux-weighted velocities will require an integral along the line of sight.
I think we want to treat disk thickness as a nuisance parameter to fit for, marginalizing over some prior on the distribution.
I've tried to be generally agnostic to the number of fit parameters in the code, but changing the "pars" array will probably require changes in a few different places, as well as the "guess" and "fixed" arrays. Might want to add a general method to define the fit parameters so that models can be swapped in and out in a single spot.
Thick disks will also need some thought when drawing intensity and velocity maps. Let's start by assuming they are optically thin and rotating as cylinders. Thickness can be treated as an exponential scale height (or use Sersic?), and the flux-weighted velocities will require an integral along the line of sight.