AP Pipeline Meeting, 2021-07-26

2pm Pacific Time

BlueJeans: https://bluejeans.com/426716450

attending:

• Eric Bellm
• Krzysztof Findeisen
• Unknown User (cmorrison)
• John Parejko
• Kenneth Herner
• Meredith Rawls
• Unknown User (gkovacs)

regrets

• Ian Sullivan

Topics for discussion:

• Project updates (Eric Bellm , Ian Sullivan ):
  • not much of broad interest
• Pipelines Infrastructure (Krzysztof Findeisen ):
  • couple of fixes in to ap_verify to fix calib types, changes to logging from middleware
• Precursor processing (Meredith Rawls, Kenneth Herner ):
  • Meredith: has started a large BPS run on HSC with fakes, made it through ISR after a bunch of restarts–documenting on the ticket, opened a couple of tickets
    • Eric hopes to chain on some Zooniverse analysis to this run
  • Ken:
    • 6-year depth run appears to be unstuck, has processed more data than last time already. 
    • running the diffim sprint configs on a smaller dataset–no database entries stored as diaPipe gave no work found
      • Chris says it's not association, it's a butler issue.  something missing from the upstream inputs
      • Krzysztof thinks that this might be worth discussing with middleware
• Image differencing algorithms  (Unknown User (gkovacs), Ian Sullivan ):
  • implementing the decorrelation afterburner in the preconvolution case, thinking about how to test it
• association/SDM/fakes/astrometry (Unknown User (cmorrison)):
  • just merged refactoring of dia_calculation to aid solar system association
  • pinging Siegfried again
  • pulling in a backlog ticket to add variability to fakes so we can get variable star tests–starting with a fixed scatter.  will lead to a small new data product to store the input fakes, and will be used by a modified matchFakesTask.
• real-bogus (Eric Bellm, John Parejko ):
  • John is thinking about testing–Eric thinks maybe Meredith's HSC fakes set so we're getting purity numbers on the same dataset. 
  • John: also finishing up the chi-squared computation for PSF match to the DIASource
    • reduced chi-squared more like 1.2 on the calexp
    • just using the default PSF size–let's spawn a new ticket to investigate
• alert distribution (Spencer Nelson):
  
  • not on the call
• Preops DP0.2 preparation (all):
  •  
• Review CI (https://chronograf-demo.lsst.codes/):
  • no notable changes
• Review outstanding action items
• QA meeting next week
  • Meredith on COSMOS fakes
  • Ken TBD on DC2
• AOB

Action Items