Run metadetect on HSC

[arunkannawadi]

Run metadetect on HSC

The title says it all

Contacts: @arunkannawadi @chihway @empEvil Day/Time: Main communication channel: GitHub repo: Zoom room (if applicable):

Goals and deliverable

Get a cookie at the SLAC sprint day. Stretch goal: Run metadetect on HSC.

Resources and skills needed

NERSC/USDF access.

Detailed description

[add additional details about the project]

[arunkannawadi]

We are unlikely to get anything done here during the sprint week since 1) the environment to run this is planned for, but not available at USDF (HSC is not available via the new middleware at NERSC AFAIK) 2) there are other pressing sprints to work on. But if you are interested in doing anything this week, let me know.

[empEvil]

When do you think USDF will have the environment ready? If this is soonish it might be worth it to go through and figure out what configuration choices should be picked for the next steps, but otherwise we can also wait till later.

[sidneymau]

I head TQ brought some HSC data over to NERSC, though I don't know whether it's the data that would be needed for this

[empEvil]

I would assume what TQ brought over is already processed HSC data, to use for photo-z estimation. But I think the configurations for running RAIL etc. that TQ and co. settles on will be valuable to keep in mind when this project get's going

[arunkannawadi]

I don't know the timeline and it's beyond my payscale. The environment will be ready whenever DM-40695 gets done. And yes, what TQ has on NERSC will be processed data and won't be in cell-based coadd format. I don't think it adds value to run metadetect on those images, and doing validation because I am sure it'll look bad.