sdss / mangadap

The MaNGA Data Analysis Pipeline
BSD 3-Clause "New" or "Revised" License
16 stars 9 forks source link

Issue with use of extinction law instead of polynomials in fitting emission lines #39

Closed francbelf closed 6 years ago

francbelf commented 6 years ago

It does not work. Error message below. The key problems seem to be in the last line of the Traceback, File "/home/francesco/CODE/mangadap/mangadap/python/mangadap/proc/sasuke.py", line 1120, in _save_results model_fit_par['EBV'][spec_to_fit,:] = model_reddening IndexError: too many indices for array

Full traceback

File "/home/francesco/.conda/envs/pybelf3/lib/python3.5/multiprocessing/process.py", line 249, in _bootstrap self.run() File "/home/francesco/.conda/envs/pybelf3/lib/python3.5/multiprocessing/process.py", line 93, in run self._target(*self._args, **self._kwargs) File "/data/manga/belfiore/DAP-paper/extinction/test_tpl_switch.py", line 43, in fit_one_cube return manga_dap(obs, plan, verbose=2, directory_path=directory_path, analysis_path=analysis_path) File "/home/francesco/CODE/mangadap/mangadap/python/mangadap/survey/manga_dap.py", line 307, in manga_dap clobber=plan['elfit_clobber'][i], loggers=loggers) File "/home/francesco/CODE/mangadap/mangadap/python/mangadap/proc/emissionlinemodel.py", line 375, in __init__ quiet=quiet) File "/home/francesco/CODE/mangadap/mangadap/python/mangadap/proc/emissionlinemodel.py", line 1165, in fit loggers=self.loggers, quiet=self.quiet) File "/home/francesco/CODE/mangadap/mangadap/python/mangadap/proc/sasuke.py", line 1576, in fit_SpatiallyBinnedSpectra loggers=loggers, quiet=quiet) File "/home/francesco/CODE/mangadap/mangadap/python/mangadap/proc/sasuke.py", line 2470, in fit model_kin_err, model_mask, model_fit_par, model_eml_par) File "/home/francesco/CODE/mangadap/mangadap/python/mangadap/proc/sasuke.py", line 1120, in _save_results model_fit_par['EBV'][spec_to_fit,:] = model_reddening IndexError: too many indices for array

kbwestfall commented 6 years ago

Hopefully this is addressed in #40 , that I just merged into master. Update your master and then try again.

francbelf commented 6 years ago

New problem

Process Process-1:nuum for spectrum: 1/2816 Traceback (most recent call last): File "/home/francesco/.conda/envs/pybelf3/lib/python3.5/multiprocessing/process.py", line 249, in _bootstrap self.run() File "/home/francesco/.conda/envs/pybelf3/lib/python3.5/multiprocessing/process.py", line 93, in run self._target(*self._args, **self._kwargs) File "/data/manga/belfiore/DAP-paper/extinction/test_tpl_switch.py", line 43, in fit_one_cube return manga_dap(obs, plan, verbose=2, directory_path=directory_path, analysis_path=analysis_path) File "/home/francesco/CODE/mangadap/mangadap/python/mangadap/survey/manga_dap.py", line 375, in manga_dap clobber=plan['elmom_clobber'][i], loggers=loggers) File "/home/francesco/CODE/mangadap/mangadap/python/mangadap/proc/emissionlinemoments.py", line 1261, in measure continuum = self.emission_line_model.fill_continuum_to_match(binid) File "/home/francesco/CODE/mangadap/mangadap/python/mangadap/proc/emissionlinemodel.py", line 1703, in fill_continuum_to_match dispersion_corrections=dispersion_corrections) File "/home/francesco/CODE/mangadap/mangadap/python/mangadap/proc/emissionlinemodel.py", line 1767, in construct_continuum_models dispersion_corrections=dispersion_corrections, dvtol=1e-9) File "/home/francesco/CODE/mangadap/mangadap/python/mangadap/proc/sasuke.py", line 2591, in construct_continuum_models moments=smoments, degree=degree, mdegree=mdegree, reddening=ebv) File "/home/francesco/CODE/mangadap/mangadap/python/mangadap/proc/ppxffit.py", line 461, in __init__ raise ValueError('LAM must be given with REDDENING or GAS_REDDENING keyword') ValueError: LAM must be given with REDDENING or GAS_REDDENING keyword

francbelf commented 6 years ago

by this I mean, fit one cube still fails at the emission line moments stage

kbwestfall commented 6 years ago

I just put in a fix for this, and I'm testing it now.

kbwestfall commented 6 years ago

fit_one_cube works for me now using elfit_key='EFITMR', re-pull master and give it another go.

francbelf commented 6 years ago

Works now!

kbwestfall commented 6 years ago

Great! Closing.