Details
-
Type:
Story
-
Status: Done
-
Resolution: Done
-
Fix Version/s: None
-
Component/s: None
-
Labels:None
-
Story Points:12
-
Epic Link:
-
Team:Data Release Production
Description
For the QA dashboard project, we need to deliver a test dataset to work on. This will be the output of DM-18300, processed with the matchVisits.py task. This ticket involves:
- Running matchVisits.py on all tracts/filters for the /datasets/hsc/repo/rerun/RC/w_2019_14/
DM-18300-jointcal/ dataset. - Delivering this dataset to Quansight in a way that they can use.
For 2., there are different ways we could do it, but I would lean towards having them use the stack, so they can butler-load the tables they need (easier reproducibility for us, and guarantees fewer surprises later). One thing that I don't know is how to copy a data repo to an arbitrary location where the full root repos wouldn't necessarily be available.
This has been done; the test dataset, based on running matchVisits.py on /datasets/hsc/repo/rerun/RC/w_2019_18/
DM-19151, is in /project/tmorton/tickets/DM-19684. It should have all the data necessary to make all the pipe_analysis style plots for the dashboard, as well as the visit->coadd match data.Future work along this direction will require orienting Quansight developers to this dataset, as well as working to prepare a standalone version for them to be able to use outside the LSP. This future work is ticketed in
DM-20015.