Currently, the specutils cross-correlation function just takes the peak pixel point instead of fitting a Gaussian to the peak (which it should do). It would be good to have the option to do both, so we can do fast CCFs against many thousands of spectra, but more accurate CCF fitting for a single spectrum.
The astropy modelling framework would be very useful here, and a PR could be created to feed the CCF function back into astropy.
Currently, the specutils cross-correlation function just takes the peak pixel point instead of fitting a Gaussian to the peak (which it should do). It would be good to have the option to do both, so we can do fast CCFs against many thousands of spectra, but more accurate CCF fitting for a single spectrum.
The astropy modelling framework would be very useful here, and a PR could be created to feed the CCF function back into astropy.