Error when using StackFitter for redshift fitting.

[Zhiyuan-G]

Hello, I am reporting an error when trying to do redshift estimation with stack fitter when running st.xfit_redshift() with st = grizli.stack.StackFitter()

File /path/to/envs/grizli39/lib/python3.9/site-packages/grizli/fitting.py:2792, in GroupFitter.xfit_redshift(self, prior, templates, fwhm, line_complexes, fsps_templates, zr, dz, zoom, verbose, fit_background, fitter, bounded_kwargs, delta_chi2_threshold, poly_order, make_figure, figsize, use_cached_templates, get_uncertainties, Rspline, huber_delta, get_student_logpdf)
   2789 fit.meta['chimax'] = (chi2.max(), 'Maximum chi2 of template fit')
   2790 fit.meta['fitter'] = (fitter, 'Minimization algorithm')
-> 2792 fit.meta['as_epsf'] = ((self.psf_param_dict is not None)*1,
   2793                        'Object fit with effective PSF morphology')
   2796 # Bayesian information criteria, normalized to template min_chi2
   2797 # BIC = log(number of data points)*(number of params) + min(chi2) + C
   2798 # https://en.wikipedia.org/wiki/Bayesian_information_criterion
   2800 scale_chinu = self.DoF/chi2.min()

AttributeError: 'StackFitter' object has no attribute 'psf_param_dict'

I have one more thing I'd like to double-check to ensure my understanding is correct in terms of MultiBeam() or StackFitter(). I understand that StackFitter relies on stacked spectra derived from a single source across multiple exposures, and the subsequent redshift estimation is based on the stacked spetra. However, I'm curious if MultiBeam (mb = grizli.multifit.MultiBeam(..)) also takes into consideration all individual beams during redshift estimation performed with mb.xfit_redshift()? Are those individual beams collectively analyzed or integrated in any manner during the process of redshift fitting?

I would really appreciate your help.