Status of jenkins jobs and slack channle for notifications: #dmj-s_verify_drp_metrics slack channel
Discussion items
Item
Who
Pre-meeting notes
Minutes
News, announcements and task review
All
CI status, bugs, issues of the week
All
verify_drp_metrics CI jobs
Check variation in metrics seen last week
Fluctuation still being seen in several metrics - strongly suspect this is due to fixes made to make CI work again a few weeks ago.
Fixes were made to the pipeline in in rc2_subset - https://github.com/lsst-dm/rc2_subset (pipelines directory). Was made to inherit as much as possible from obs_subaru.
JC & ED looked at the TE1 metirc diff at the tract level. Were calculating from deep_coadd_ForcedSource tableand the commparison we should do is from deep_coadd_meas table
KB: how shoudl we know this and find the doc?
JC: dug through the yaml policy defns for conveting to parquet.
Changed the default to use deep_coadd_meas in the base classes. This has mostly fixed the differences seen in comparison with the parquet tables. Still seeing small differences. Ready for internal team review.
Will ask someone in pipelines to review the implementation of the TEx algorithm, once the internal code change review is done
PF:
close to done with AA1 metric - needsthe filtering SNR/flag filtering to be updated to work with the parquet format.
DM-32611
-
Getting issue details...STATUS
mostly done with comparison with a complately external catalogue. PoC level.
RHL- take care about problems with the outside catalogues
AuxTel data reduction is proceeding, currently tweaking PSFEX parameters with Merlin to handle both dense and sparse imaging fields. Next AuxTel run is Dec 6-9, we should have a chance to get another set of imaging data then. Should discuss benefits of adding r' filter of same field before it sets
AuxTel is shallower so some exposures cannot be processed out of the box due to fewer stars
What is the process to ask deep technical questions and process questions? What channels should be used for what? rubinobs-sitcom-surveys v rubinobs-science-verification
Keith Bechtol to make a ticket to better understand mapping of these camera and calibration products characterization efforts to verification documents and the focus of these efforts. Discuss with the SCLT