AP Pipeline Meeting, 2023-03-27

2pm Pacific Time

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

attending:

• Kenneth Herner 
• Erin Howard 
• Brianna Smart 
• Harshit Rai 
• Nima Sedaghat 
• Ian Sullivan 
• John Parejko 
• Ari Heinze 
• Krzysztof Findeisen 

regrets

• Eric Bellm   (conference)
• Bruno Sanchez 

Topics for discussion:

• Project updates (Eric Bellm , Ian Sullivan ):
  • No recent news
• Pipelines Infrastructure (Krzysztof Findeisen , John Parejko ): 
  • Krzysztof: Nothing to report
  • John: Worked on new Characterize and Calibrate Task replacement with Eli
    • Called CalibrateImageTask 
    • Notes: https://gist.github.com/parejkoj/853f733ffdf8cff1cd8a48eae7375b89
    • Question: will this new task handle snaps?
• transinet (Nima Sedaghat Harshit Rai , Eric Bellm ):
  • Model packages PR is merged
  • RFC-909 converged, will implement as suggested.
    • Butler access is also nearly ready, but will postpone implementation until after testing existing neighbor mode (with git lfs) for ap_verify and monthly reprocessing
  • Harshit still waiting on SLAC access
    • Worked through DM Bootcamp tutorials. Also Transinet and other research papers
• Precursor processing ( Kenneth Herner , Erin Howard ):
  • Saha bulge processing
    • Split processing run into two runs around coaddition. Now i-band has generated most templates
      • Ap-pipe not finished, but there are ~2/3 failures with "Variance rescaling exceeds configured limits"
  • DECam and HSC have finished creating new templates
    • We will create a ticket to update the ap_verify templates using the new runs (DM-38449, DM-38450)
    • Also a ticket to create new DC2 templates (DM-38451, DM-38452)
    • Ken finished the HSM shape flag analysis, will present next week.
      • Recovers ~50% of negative sources
• Image differencing algorithms, DCR  ( Bruno Sanchez , Ian Sullivan ): 
  • No updates
• alert distribution (Brianna Smart , Eric Bellm ):
  • Updated technote to use secure password passing
• Solar-system processing (Ari Heinze ):
  • Ran a python-wrapped make_tracklets, and it worked (first impression)
    • Spot checks indicate it is working, but will do more in depth analysis
  • Python-wrapped Heliolinc is coming next.
  • Will attend Science Pipelines pair coding to check pybind-11 coding
• Review CI (https://chronograf-demo.lsst.codes/):
  • No changes last 30 days
  • February 19-20 timing increase has not recovered, still no culprit
• Review outstanding action items
• QA meeting in April:
  • comment on status of auxtel PP
  • Ken says he could follow up on the crossmatching results of the switch to HSM shape
• QA meeting in May:
  • edge flags? (& vignetting)
  • off-chip centroids – JP thinks it would be good to dig into why first
  • RB scoring?  still too early, want to show it one of the large runs
  • Eric on detection thresholds/fakes forced photometry?
  • Discussion of AP QA needs in the future that Commissioning/LINCC might assist with
  • preconvolution – this still needs some digging in to understand the edge behavior
  • Bruno share his diffim status report
• AOB
  • DM-38285 and DM-38286 to follow up on diaSource cutouts that are off the chip.
  • Bri also contacted about Commissioning support, we will discuss at sprint planning

Action Items