BlueJeans: https://bluejeans.com/426716450

attending: Eric Bellm John Swinbank Unknown User (cmorrison) Krzysztof Findeisen Meredith Rawls Unknown User (emorganson) Unknown User (gkovacs)

Topics for discussion:

  • HiTS processing (Meredith Rawls ):
    • put  DM-20535 in review.  Using higher spatial resolution templates, plus better background handling makes the spatial plots much nicer
    • still a fair bit of separation between totFlux forced photometry and the calexp photometry; many constituent DIASources are flagged, but even the unflagged ones are often off. 
    • one truly variable source seems to be better, which makes us wonder if this is just still bad differencing–looking at the cutouts it seems like there are still artifacts, but maybe somewhat improved?
    • we are discussing how much we'd expect the centroids to wander based on just SNR and statistical fluctations
  • DES processing (Unknown User (emorganson)):
    • continuing to work on Kubernetes/Docker
  • ap_association (Unknown User (cmorrison)):
    • Yusra finished the review about stripping cal objects; then is blocked on SDM so will move forward on timeseries feature plugins
  • DCR (Ian Sullivan ):
    • Ian away
  • Image differencing (Unknown User (gkovacs) ):
    • working on debugging; most debug plots in ip_diffim are broken due to a change in afw
    • John asks: is the work useful as we get ready to rework ip_diffim?  Gabor thinks so unless we plan to throw away large chunks of the C++, which he doesn't think is the plan.  Eric thinks also could help check that the restructure/Gen3 port worked as expected.
  • Review CI (https://chronograf-demo.lsst.codes/):
    • nothing this week
    • Chris says when SDM standardization merges there will be a new step to store calibrations, so we should see if has an effect
  • AOB
    • Eric: RFC-630 is approved, Eric put the implementation ticket in review, so hopefully next week we'll see some (good) effects in chronograph