AP Pipeline Meeting, 2020-10-26

2pm Pacific Time

BlueJeans: https://bluejeans.com/426716450

attending:

• Eric Bellm
• Ian Sullivan
• Unknown User (cmorrison)
• Kenneth Herner
• Meredith Rawls
• Krzysztof Findeisen
• Unknown User (gkovacs)

regrets:

Topics for discussion:

• Gen 3 conversion (all):
  • Krzysztof Findeisen :
    • working on make_apdb replacement (Gen2 deprecation blocker)
  • Yusra has given Ian a couple small deprecation blockers to distribute

• Precursor processing (Meredith Rawls ): 
  • working on DECam HiTS calibration products–it's a pain because the raws have erroneous filter info, so have to do some hacking.
    • can this be fixed with astro_metadata_translator?
• report from NCSA (Unknown User (emorganson)  + Unknown User (manninga) et al):
  • Eric Bellmto check in on what's happening at NCSA 02 Nov 2020 
• ap_association/SDM/fakes/alerts (Unknown User (cmorrison)):
  • ran the astrometry WCS inaccuracy test on HSC New Horizons (~6k reference objects, ~3k get matched typically)
  • shift and rotations okay, as for HiTS
  • a handful of failures for distortions–more sensitive due to larger number of reference objects.  Starts failing at ~1" distortion.
    • Eric Bellmdiscuss WCS inacurracy results with RHL 02 Nov 2020 
• Crowded field photometry (Ian Sullivan ):
  • stuck in management land
  • has been updating DCR model code to finish filter dependence
• Image differencing (Unknown User (gkovacs) ):
  • showing a notebook attempting direct PSF subtraction
  • trying to subtract PSF from calibPsfUsed sources
  • offsets in centroids vs. PSF yields a bunch of dipoles
    • any chance this is an indexing convention issue?  dipoles are all the same direction
    • which centroid is the right one?
  • used scipy to allow PSF to shift centroid–some improvement, but more details need exploration
  • going to switch to PsfEx, and more data
  • plan to present to RHL a week from tomorrow
• Preops project (Kenneth Herner  and all):
  • working on running ap_verify at FNAL, turning up various bugs in Gen 3-- Krzysztof Findeisen providing support
• Review CI (https://chronograf-demo.lsst.codes/):
  • nothing notable
• Review outstanding action items

• AOB
  • Doodle poll out for sprint planning

Action Items