AP Pipeline Meeting, 2022-04-11

2pm Pacific Time

Zoom: https://washington.zoom.us/j/98846655020

attending:

• Eric Bellm 
• John Parejko 
• Kenneth Herner 
• Meredith Rawls 
• Karlo Mrakovcic 
• Krzysztof Findeisen 

regrets

• Ian Sullivan (vacation)
• Nima Sedaghat (travel)

Topics for discussion:

• Project updates (Eric Bellm , Ian Sullivan ):
  • all-hands meeting this Thursday
  • SITCOM reorg coming
  • RHL visiting April 19-20
• Pipelines Infrastructure (Krzysztof Findeisen , John Parejko ):
  • Prompt Processing Demo at Wednesday Pipelines meeting
• Precursor processing (Meredith Rawls, Kenneth Herner ):
  • Ken: progress on the "shape flag for negative sources" ticket.  seems to have just required calling SDSSShapeAlgorithm with negative=True?
    • KF: it looks like what you actually did was change the default in the pybind11 binding, instead of changing a function call in the task code–is that a good idea?
    • JP: need to keep C++ and pybind11 defaults the same.  Better to change this where this is called
• Image differencing algorithms, DCR  ( Ian Sullivan ):
  • out today
• real-bogus (Nima Sedaghat, Eric Bellm, John Parejko ):
  • out today
• alert distribution (Eric Bellm ):
  • continuing work on estimating packet size increase with three stamps–exploring a few different ways to approach this
  • Talked about multi-resolution stamps with Francisco Forster.  Didn't make a huge classifier difference except in the case of glints...
    • (glints examples from ZTF + Fink: https://ui.adsabs.harvard.edu/abs/2022arXiv220205719K/abstract)
    • added a backlog project to figure out how to post-process and remove these from LSST
  • SST discussed sub-threshold alert requirements; Leanne, Eric, Melissa will work on a technote.
• Cosmic rays (Karlo Mrakovcic ):
  • working on report (already at 50 pages...)
  • global conclusion: current algorithm appears to "good enough".  still opportunities to fine tune, but it's not terrible.  Alternatives not obviously better.
  • Eric: are the numbers going to be enough to help Zeljko resolve the snaps question? Karlo: think so
• Review CI (https://chronograf-demo.lsst.codes/)
  • nothing post-Piff.  Josh Meyers may be looking at relevant ticket ( DM-34254 - Getting issue details... STATUS )?  Eric to confirm.
• Review outstanding action items
  • none
• QA meeting in May:
  • (Ian: DCR:  defer to May+)
• AOB
  • none

Action Items