Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

guadalupe LSS catalogs #36

Open
ashleyjross opened this issue Jan 21, 2025 · 11 comments
Open

guadalupe LSS catalogs #36

ashleyjross opened this issue Jan 21, 2025 · 11 comments
Assignees

Comments

@ashleyjross
Copy link

The LSS catalogs here desi/survey/catalogs/edav1/da02/ (organized like sv3) should get copied to the public directory, mirroring sv3.

While I thought we planned for this during the EDR, perhaps not, because the only way I see to make it work with the sv3 data model is to make the version 'guadalupe'. This would put everything in

/global/cfs/cdirs/desi/public/dr1/vac/dr1/lss/guadalupe/

I don't think it works, but an alternative would be

/global/cfs/cdirs/desi/public/dr1/vac/m2/lss/v1.0/

(In the early BAO paper, the sample is referred to as DESI-M2, for 2 months of main survey.)

A couple of minor notes would be needed in the EDR data model, e.g., the alt MTL directory existing only for SV3. I think we should also add a simple text readme within each directory making it clear that these are not the full DR1 LSS catalogs.

@weaverba137
Copy link
Member

When you say "I don't think it works", could you please specifically say why not? There are other VACs that do follow this pattern.

A couple of minor notes would be needed in the EDR data model, e.g., the alt MTL directory existing only for SV3.

OK. Is there a PR for this already?

I think we should also add a simple text readme within each directory

Please do this before any data are moved into place. The final data directory will be unwriteable.

@ashleyjross
Copy link
Author

Oh, if there are others that follow that pattern and we can use /global/cfs/cdirs/desi/public/dr1/vac/m2/lss/v1.0/, I think it is better. (I find them on my quick look.)

@weaverba137
Copy link
Member

/global/cfs/cdirs/desi/public/dr1/vac/m2/lss/v1.0/

No, it would be /global/cfs/cdirs/desi/public/dr1/vac/dr1/lss/guadalupe/v1.0/.

@weaverba137
Copy link
Member

When you said "I don't think it works" I assumed you were referring to

/global/cfs/cdirs/desi/public/dr1/vac/dr1/lss/guadalupe/.

@ashleyjross
Copy link
Author

Ah, yeah, the problem is that is mismatched with the LSS EDR data model: DESI_ROOT/vac/RELEASE/lss/VERSION/ (https://desidatamodel.readthedocs.io/en/latest/DESI_ROOT/vac/RELEASE/lss/VERSION/index.html) and we want to re-use that (since the catalogs were produced at the same time)

@ashleyjross
Copy link
Author

So, I'm assuming /global/cfs/cdirs/desi/public/dr1/vac/dr1/lss/guadalupe (with guadalupe as the version and so no v1.0) is the best we can do. If so, I'm ok with it...

@weaverba137
Copy link
Member

I think we're writing past each other just now, so I'm going to wait for a bit.

@ashleyjross
Copy link
Author

The issue is, I think, we cannot use

/global/cfs/cdirs/desi/public/dr1/vac/dr1/lss/guadalupe/v1.0/

with the vac data model, because it is DESI_ROOT/vac/RELEASE/lss/VERSION/

(so VERSION would have to be guadalupe/v1.0 , which I don't think is allowed. If it is, then, great, we can actually use it.)

If that is correct, the best option I see is to use

/global/cfs/cdirs/desi/public/dr1/vac/dr1/lss/guadalupe/

and then add a note on the data model page that v1.0 is the only version for EDR, and guadalupe is the only version for DR1 for this data model. Then, we would add a sentence linking to the full DR1 LSS catalog data model.

But, someone might have a better idea!

@weaverba137
Copy link
Member

You can absolutely describe VERSION as guadalupe/v1.0.

@ashleyjross
Copy link
Author

Ok, that is obviously the best solution then!

@aureliocarnero
Copy link

Checking the guadalupe/v1.0 data. There is one directory in clustering named recon. Since this is not documented in the current EDR datamodel, I'm going to remove it from the public repo

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants