I would like to use WDPhotTools for fitting models to some directly imaged WDs as part of a survey program I am conducting, since this is such a great package for model use and fitting. I appreciated the discussion in the docs about how symmetric uncertainties in flux translate to asymmetric uncertainties in magnitudes since they are the log of flux. Since I am measuring photometry in images, I measure symmetric uncertainties in flux values, but the fitter (according to the docs) only supports entering photometry in magnitudes. So I am propagating a bias forward by trying to estimate my magnitude uncertainties as symmetrical when they're not. You mention that the fitting is done in flux space for the fitter because of this reason. It would be a great feature to enable users to feed flux values into the fitter rather than magnitudes to avoid this problem. Or alternatively, enable users to enter asymmetrical uncertainties for magnitude (which in my case would become symmetrical about the mean when converted back into flux during the fitting). Apologies if this feature already exists, but I don't see it in the API or tutorials.
I would like to use WDPhotTools for fitting models to some directly imaged WDs as part of a survey program I am conducting, since this is such a great package for model use and fitting. I appreciated the discussion in the docs about how symmetric uncertainties in flux translate to asymmetric uncertainties in magnitudes since they are the log of flux. Since I am measuring photometry in images, I measure symmetric uncertainties in flux values, but the fitter (according to the docs) only supports entering photometry in magnitudes. So I am propagating a bias forward by trying to estimate my magnitude uncertainties as symmetrical when they're not. You mention that the fitting is done in flux space for the fitter because of this reason. It would be a great feature to enable users to feed flux values into the fitter rather than magnitudes to avoid this problem. Or alternatively, enable users to enter asymmetrical uncertainties for magnitude (which in my case would become symmetrical about the mean when converted back into flux during the fitting). Apologies if this feature already exists, but I don't see it in the API or tutorials.