Installation¶
Quick version¶
The distribution is called astro-fitted because plain fitted on PyPI belongs to
an unrelated package. The import name is just fitted:
To work from a checkout instead — which you want if you plan to modify the code:
git clone https://github.com/fittingtransientswithdiscs/FitTeD.git
cd FitTeD
python3 -m pip install -e .
That is enough to run everything in tutorials 00–07. There are no compiled extensions, no compiler, and no build step: FitTeD is pure Python.
Why -e?
The editable install lets you edit the package in place and see the change without
reinstalling. It is not required — a plain pip install . works identically — but it
is the configuration everything is tested in.
Requirements¶
FitTeD is developed and tested on Python 3.10 and newer.
These are installed automatically by pip:
| Package | Used for |
|---|---|
numpy |
everything |
scipy |
special functions, optimisation |
astropy |
cosmology, units, constants |
matplotlib |
plotting |
emcee |
MCMC sampling |
corner |
corner plots |
h5py |
HDF5 chain backends |
numba |
the compiled Bessel-function kernels |
pandas, tqdm, importlib_resources |
data handling and progress bars |
Optional: manyTDE¶
To load published TDE photometry by IAU name — Data_Set(manyTDE_name='AT2019dsg', ...) —
install manyTDE:
Only tutorial 06 requires it. Everything else uses data that ships with FitTeD or that you supply yourself.
Checking it worked¶
FitTeD prints a banner on import that reports the same thing. manyTDE available: No
is a perfectly good installation — only tutorial 06 needs it.
Tutorial 00 runs this check and then computes a light curve, so if that notebook executes you are done.
Running the tutorials¶
The notebooks add the parent directory to sys.path, so they work from a bare checkout as
well as an installed copy. If FitTeD is installed, the installed copy takes precedence.
Every notebook on this site ships with its outputs already computed, so you can read them all without running anything.