Two naming problems live in the Zenodo data package (record 18298544, fetched by bin/download_data.py), not in the repo, so they can't be fixed
in-tree alone — the files and the survey config that references them must change together.
DES file basenames use des_y6_ instead of des_yr6_.
See:
des_y6_5_sig_maglim_band_{g,r,i,z,y}_nside_512.hsp
des_y6_stellar_efficiency_beta.csv
des_y6_photoerror_r.csv
Misspelled efficiency column classifiction_eff.
The stellar-efficiency CSVs carry a misspelled header column classifiction_eff (should be classification_eff):
data/others/lsst_stellar_efficiency_cutr.csv
data/surveys/des_yr6/des_y6_stellar_efficiency_beta.csv
header: mag_r,delta_mag,detection_eff,classifiction_eff,classification_detection_eff.
As a stopgap, streamobs.surveys.Survey.set_completeness already accepts both spellings (commit 6674845), and new_survey.md documents the correct
classification_eff. But the distributed data still ships the typo.
Fix (coordinated): correct the column header to classification_eff in the generation step (notebooks/create_streamobs_files_hlwas.ipynb produces
the misspelling), regenerate/re-upload the data package (same Zenodo bump as above), then remove the legacy classifiction_eff fallback in
set_completeness and drop the "legacy" note from its docstring.
Note: both fixes touch the same Zenodo archive + bin/download_data.py version bump, so it's natural to do them in one data-package release.
Two naming problems live in the Zenodo data package (record 18298544, fetched by bin/download_data.py), not in the repo, so they can't be fixed
in-tree alone — the files and the survey config that references them must change together.
DES file basenames use des_y6_ instead of des_yr6_.
See:
des_y6_5_sig_maglim_band_{g,r,i,z,y}_nside_512.hsp
des_y6_stellar_efficiency_beta.csv
des_y6_photoerror_r.csv
Misspelled efficiency column classifiction_eff.
The stellar-efficiency CSVs carry a misspelled header column classifiction_eff (should be classification_eff):
data/others/lsst_stellar_efficiency_cutr.csv
data/surveys/des_yr6/des_y6_stellar_efficiency_beta.csv
header: mag_r,delta_mag,detection_eff,classifiction_eff,classification_detection_eff.
As a stopgap, streamobs.surveys.Survey.set_completeness already accepts both spellings (commit 6674845), and new_survey.md documents the correct
classification_eff. But the distributed data still ships the typo.
Fix (coordinated): correct the column header to classification_eff in the generation step (notebooks/create_streamobs_files_hlwas.ipynb produces
the misspelling), regenerate/re-upload the data package (same Zenodo bump as above), then remove the legacy classifiction_eff fallback in
set_completeness and drop the "legacy" note from its docstring.
Note: both fixes touch the same Zenodo archive + bin/download_data.py version bump, so it's natural to do them in one data-package release.