This is a tool that allows you to fetch DESI spectra without manually downloading them. This tool can be run on your laptop and does not have to run on NERSC. (it still can run on NERSC, in this case it will be typically faster)
You may also consider using SPARCL https://astrosparcl.datalab.noirlab.edu/ for another interface.
Author: Sergey Koposov, Email: skoposov AT ed DOT ac _DOT uk, University of Edinburgh
Please acknowledge the software if you use it.
To install just do
pip install desi_retriever
or the version from github
pip install -U desi_retriever@git+https://github.com/segasai/desi_retriever
desi_retriever works with public DR1 data and private DESI data. If you want to use non-public DESI data, you must set up the DESI user / password
You need to put in the standard DESI web login/password in the home folder in the file: $HOME/.desi_http_user in the following format login:password
The login password are the same ones as used to access data.desi.lbl.gov
To fetch a given coadd spectrum you can do
import desi_retriever.dr1
SP = desi_retriever.dr1.get_specs(survey='sv1',
program='dark',
hpx=17683,
targetid=39627652591521181)[0]
You could also get the RVSpecFit models by using
Dmod = desi_retriever.dr1.get_rvspec_models(survey='sv1', program='dark', hpx=17683, targetid=39627652591521181)[0]
More examples are in jupyter notebooks in example folder.