Date
Attendees
Agenda
- Addition of Sophie Reed (Princeton) to the working group. Welcome!
- Plan to move forward with initial tests.
- single image and catalog tests
- Other tests besides those for Data Release Processing (e.g. nightly and prompt processing products)
- Planned work for each of us over the next few weeks.
Discussion items
Paul outlined the current state of the codes available for testing. Currently we expect some additional work to obtain catalog comparisons (derived from compressed vs. never-compressed images). Current software makes flux and centroid comparisons, additions for basic shapes/sizes are likely needed to demonstrate the impact on WL measurements.
The first tests will use HSC data. While this is likely sufficient for most tests, at some point observations from something other than a cosmological survey will be necessary (to determine how an algorithm behaves for other cases, e.g. crowded fields or where significant nebular emission is present).
Krzystof has made a first foray into what image products will be saved within AP pipelines (and measurements that might differ from those in the DRP case). The primary case of concern there is whether separate tests are needed to understand how forced photometry might perform on compressed products. This is currently not a big ticket item as the AP pipeline uses a 30 day store of PVI and difference images for pre-covery searches.
The question about where we should be accumulating knowlege/results. Current plan is to use Confluence either from the Main Lossy Compression WG page or through Child Pages as makes sense.
Action items
- Paul Price Sophie Reed will level-set with the intent to exercising the existing testing framework (and hopefully some preliminary example tests).
- Krzysztof Findeisen Robert Gruendl will continue trying to assemble knowledge-base with respect to requirements