DRP Metrics Monitoring 2022-12-19

Mondays 12pm - 12:50pm ET

Yusra's Zoom: https://princeton.zoom.us/my/yusra

Attendees: 

Agenda:

• Meeting recorder - Jim Bosch (last 6 meetings were:  Keith, Fred, Nate, Jeff, Clare, Arun) 
• Announcements
• Review Action items from last month
  • astrometry plots (visit and coadd sky plots) now in review
  • sky coverage plots ( DM-34625 - Getting issue details... STATUS ) now in review
  • Piff robustification ( DM-36930 - Getting issue details... STATUS ):
    • Lauren MacArthur will make a new ticket to take over adding a check on max distance to nearest PSF star, and do it in the visit summary stat code, starting from code snippet provided by Joshua Meyers :  DM-37411 - Getting issue details... STATUS
    • some of these CCDs are salvageable
    • some of these are not salvageable, and could silently cause problems downstream (maybe not FGCM, jointcal at worst weakly, definitely coadds, maybe CRs)
    • this is complicated by the fact that the coadd input selection PSF checks still incorrectly use the original PSF
    • Jim Bosch will add a new round of visit-summary calculation after the final PSFs, and switch back to PSFEx for the initial PSFs ( DM-35207 - Getting issue details... STATUS )
  • stellar locus in 9813 is wider than other tracts (long-standing, no updates)
• Processing Status
• Review w_2022_48 rerun:
  • "payload" errors (from tasks) now being included better in pilot logging
  • same characterizeImage and calibrate failures as before
  • ~40 fewer subtractImages failures compared to last run (expected)
  • 9697 patch 7 detection:
    
    • failed in _44 ("unable to evaluate coadd PSF")
    • succeeded in _48
    • thought the fix was  DM-35722 - Getting issue details... STATUS but that landed in _38
    • also thought the fix might be  DM-36762 - Getting issue details... STATUS , but while timing is right, there's no connection from the code change to the behavior
    • need to track this down to make sure it gets backported to v24
    • want to check that the right version of the stack was actually used for step1 in _48
  • templates (and other things holding lots of PSFs) do indeed take half the space
  • DM-36763 - Getting issue details... STATUS
  • Recall history:
    
    • w12: psfex
    • w16: piff  (bad size residuals)
    • w20: finalizeCharacterize (bad apcorr configs → bad stellar locus)
    • w22: lanczos11 + apcorr configs (better stellar locus and size residuals approx equivalent to psfEx!
    • w24: PIFF kernelSize to 25. new scarlet lite storage.
    • w28: attempt at fixing measure failures:  DM-35722 - Getting issue details... STATUS
    • w32: First RC2 at SLAC. subtractImages compatibility mode on
    • w36: subtractImages Compatibility mode off
    • w40: 
    • w44:
    • w48: 9697/7 succeeds, extra subtractImage failures gone, 
  • Chronograf, plot-navigator review of _48
    
    • "Everything looks okay"
    • But the first thing Chronograph showed us for wPerp was very strange and concerning (before picking times), and we can't get back to it now.
    • analysis_tools plots apparently failed to run
      • maybe this was expected, because we had 'u' band configured in HSC; this would have been fixed on  DM-36305 - Getting issue details... STATUS (landed in _50)
      • unclear how much (if any of this) was working prior to that ticket
      • no one can explain why it ran in _44
  • There will not be a _52 run.  First HSC RC2 run will w_2023_02, but be prepared for surprises in tagging.
    
    • Expect analysis_tools object core plots should be working again.
    • Other tickets still pending, motivated by HSC PDR4 deadlines.
    • SkyObject histogram plots should have been added (landed in _49).
    • Piff performance improvements in external review, don't know when that will land.
    • fgcmcal will soon be using isolated star catalog.
    • new rho statistics plots landing soon in analysis_tools, including multi-tract plots
    • some dataset type names will be changing to be more predictable/descriptive
• Expect a doodle for a  new meeting time!
• AOB:
  • DM-34625 - Getting issue details... STATUS needs input on dataset type naming conventions and some display choices (will discuss in slack)
  • PSA: skyPlots (at least for skyObjects) seem to have inconsistent labels and numbers outside the plots (included in DM-37172 - Getting issue details... STATUS )

Action Items