Page History

Mondays 12pm - 12:50pm ET

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

Postponed from Nov 29. 

Attendees:

• Yusra AlSayyad 
• Lauren MacArthur 
• Eli Rykoff 
• Unknown User (brendal4) 
• Clare Saunders 
• Arun Kannawadi 
• Jeffrey Carlin 
• Jim Bosch 
• Leanne Guy 
• Hsin-Fang Chiang 
• Christopher Waters 
• Merlin Fisher-Levine 
• Matthias Wittgen 
• Colin Slater 
• Sophie Reed 
• Lee Kelvin 
• Dan Taranu 

Agenda:

• Announcements:
• Review Action items from last week
• Processing Status
• Review HSC RC2 rerun w_2021_46
  • What was noteworthy in this rerun?
    • Gen2: Still in progress. 

• • • Gen3: DM-32563
      • Went smoothly!
      • wPerp still needs to be run on lsst-devl by hand. 
      • JFB: We expect this to be fixed by parquet tables. 
      • LG: We expect this to be wrapped up imminently.
  • What changed? Annotate
    • stellar_locus_width_wPerp has gone up – not sure why. Leanne Guy suggests that this may be related to the behavior seen in CI Gen3 where the stellar locus fluctuated nightly, but it's not possible that this could be because of something in Science Pipelines (changed are not made on a nightly basis, so this would be impossible). It may not be possible to trust the numbers coming out of faro right now. We also need to investigate why the dates for the stellar locus only go back to October.
  • What do we expect from w_2021_50 IN 2 DAYS (last RC2 of the year)?
    • Processing changes:
    • Changes toward Gen2-Gen3 parity:
      • LAM would like a quick discussion on the last few comments on 

        Jira
        server JIRA columnIds issuekey,summary,issuetype,created,updated,duedate,assignee,reporter,priority,status,resolution columns key,summary,type,created,updated,due,assignee,reporter,priority,status,resolution serverId 9da94fb6-5771-303d-a785-1b6c5ab0f2d2 key DM-30284

      • In skyCorr there are very small expected differences.
      • In directWarp the differences are more significant. In Gen3, the sky correction is not getting applied correctly and is not updating the calexp. After correcting this the major differences are gone. There are still small differences which are probably explained by differences in FGCM. It looks like the root issue is that a reference to the maskedImage is getting changed, but the maskedImage is not changed. JFB suggests that the problem is that the the maskedImage was not deep-copied.

• • • What new metric can we expect next time 
      • DC2:  DM-32624 (the secret bank of DC2 configs)
      • RC2: 
        • check that skyCorr is being applied during makeWarp. w50
• Review DC2
  • w_2021_44
  • • gen3: DM-32391 
      • expected error in step1. also expected error from DM-31777
    • gen2: 
    • Gen3 Pipeline for DP0.2 (backport-v23 label)

• • Why is the astrometry metric so bad for DC2? The photometry metric is better than what we see from HSC. Eli says that jointcal will not help astrometry because the simulation has steps between detectors that do not follow the model of smooth variations over the field of view expected by jointcal or gbdes.

• Processing schedule for the next 6 weeks:
• AOB:
  

Action Items