search

LSSTScienceCollaborations / StackClub

Learning the LSST software Stack, by writing jupyter notebook tutorials.
https://stackclub.readthedocs.io/
MIT License
45 stars 17 forks source link

Project/density maps/jeffcarlin #250

Closed jeffcarlin closed 3 years ago

jeffcarlin commented 4 years ago

Hi Nushkia - since you expressed interest in reviewing my Stack Club project a few weeks ago, I have invited you to review. Let me know if you are unable to review this notebook, or if you have any questions about it.

Alex - your comments are welcome as well, but if you want to weigh in after Nushkia, I can ping you then.

kadrlica commented 4 years ago

Thanks Jeff. I'll weigh in after Nushkia (if she is available).

kadrlica commented 4 years ago

This was super cool Jeff, thanks for writing it! I'll admit, it does make me a bit scared about semi-resolved dwarf searches, though I assume that the algorithms will be tuned up during commissioning.

jeffcarlin commented 3 years ago

I see this cell

! cd ~/stackclub/StackClub/ && python setup.py -q develop --user && cd -

I don't think you can use an absolute path like that, since not everyone has the Stack Club notebooks installed in the same place.

In the FindingDocs notebook (under GettingStarted), Phil does something like this:

! cd .. && python setup.py -q develop --user && cd -

Good point - I updated it to the relative path.

jeffcarlin commented 3 years ago

I see this call to see if a data set exists. Does this actually "guarantee" that the data set is on disk?

butler.datasetExists('deepCoadd_forced_src', dataId=dataref14)

I thought there may have been some subtleties here that involved getting the URI

I added the following to the notebook to clarify:

"Note, however, that in the Gen2 butler, this doesn't guarantee that the catalog actually exists as a file on disk. It only tells us that some processing steps were executed that intended to create the catalog, and thus added its entry to the registry. This issue will be remedied in the Gen3 butler, so it is not important to dwell on it further."

jeffcarlin commented 3 years ago

What does immediate=True mean here? I hadn't seen that before. Is it worth noting?

image14 = butler.get('deepCoadd_calexp', immediate=True, dataId=dataref14)

Indeed, that was a holdover to ancient times, but unnecessary now. See this community post

kadrlica commented 3 years ago

@jeffcarlin want to merge? This notebook is a nice holiday present!