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
Move to zoom from next week. We will use the existing zoom channel.
Co-work - look to merge with pipelines
Thought about feature requests → will assess influx 2 first and then look at what features we would like but that are not present. Alerts are in influx 2.0 which will be useful.
CI status, bugs, issues of the week
All
verify_drp_metrics CI jobs
Much variation in metrics in the last 7 days in AM/PA/TE metrics
Unlikely that there is a ticket commit every daya that led to this
Check input catalogs to faro first to check that the input columns are stable - CI intermediate data products are persisted
Checkout daily and see if we even get the same numbers.
Parquet conversion - tickets assigned to convert all the metrics that need matched catalogs as input. Base classes that used the forced photometry table will not be able to handle AM repeatability metrics due to fixed centroid, so cannot compare the astrometric solution. Need a new catalog matcher that takes the per-visit source catalogs. Keith planning to do this in the next week
The need to match the source catalogs for astrometric repeatability raises the N-way matcher topic again. One solution would be to use the object table as the reference and do matches of source catalogs against the object table, effectively associating sources with an object id.
TEx conversions - Erik & Jeff have it running for all analysis contexts.
Multiples found in the output table of a multimatch selection of individual visits. Erik to look into it.
TEx algorithm convert to accept parquet as input. Verified with FITS.
Ready for a PRPF: Comparison to truth table reference catalogs
Comparison with refcats that are not part of the regualr processing (e.g Gaia, Pan-STARRS), look to implement a simple metric to understand the process of working with any refcat in the butler context. presentation linkDM-31985
Erik: Merlin has finished a first run through of AuxTel data to handle SF processing. Handed over instructions. Keith ran to completion (see discussion here). Will start processing AuxTel data.
Data is at NCSA, latiss data repo
Can we move to Antu for processing. Erik will ping RHL on this.
Goal is to have near real-time processing at summit
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