To prepare, make sure your environment has the latest version of SEDkit and AstrodbKit2:
pip install sedkit
pip install AstrodbKit2
You will also need to have a copy of the SIMPLE database. You can download a SQLite binary version of SIMPLE from https://github.com/SIMPLE-AstroDB/SIMPLE-binary or can pull the raw data and build it yourself from https://github.com/SIMPLE-AstroDB/SIMPLE-db (see more instructions in that repo)
Once you have the database, you can place it in your current working directory and can connect to it with:
from astrodbkit2.astrodb import Database
db = Database('sqlite:///SIMPLE.sqlite')
This assumes the database file is in the same working directory, otherwise connect to it with something like:
db = Database('sqlite:///path/to/SIMPLE.sqlite')
This Database
object will need to be passed to SEDSIMPLE
for it to work.
The key functionality in this package is a new class (SEDSIMPLE
) that inherits from SED
in order
to provide wrappers around SIMPLE Database calls.
For convenience, you may want to have sedsimple.py in your current working directory.
Initiating with SEDSIMPLE
is as simple as:
from sedsimple import SEDSIMPLE
from astrodbkit2.astrodb import Database
db = Database('sqlite:///SIMPLE.sqlite')
sed = SEDSIMPLE(db, 'Trappist-1', auto_db=True)
The parameter auto_db=True
tells it to load as much information from the provided database as it can.
This includes coordinates, parallaxes, spectral types, photometry, and spectra.
If you choose auto_db=False
(the default), you can load whichever data you need manually with:
- sed.load_coords_db()
- sed.load_parallax_db()
- sed.load_photometry_db()
- sed.load_spectral_type_db()
- sed.load_spectra_db()
You can use sed.fetch_single_spectrum_db()
to fetch the wavelength, flux, flux uncertainty,
and reference for a single spectrum if you need to do further transformations.
The input is a Spectra row from db.inventory
;
see https://astrodbkit2.readthedocs.io/en/latest/ for more information on using AstrodbKit2 queries
and the SIMPLE Documentation for more info on the database structure.
Note: upon initialization SEDSIMPLE
will search for the specified target in the database.
The internal database name may be different from what provided name,
but db.search_object
is used to find the correct match.
You may want to manually check database entries with queries like:
db.search_object('Trappist-1')
db.inventory('2MASS J23062928-0502285') # primary source name in the database for Trappist-1
All methods available in SED
are available in SEDSIMPLE
which means you can proceed as normal
(eg, sed.add_spectrum()
, sed.plot()
, sed.results
).
A tutorial notebook is available that walks you through the Trappist-1 example.