Chronostar, the next generation of discovery and characterisation of stellar associations
The full docs have a more detailed installation and quickstart.
Create a conda environment and install packages available through conda.
git clone https://github.com/tcrundall/chronostar-tng.git
cd chronostar-tng
conda env create -n chron -f environment.yml # set up py39 environment
conda activate chron
pip install . # this will put chronostar-trial in your site-packages
You will now have multiple command-line tools at your disposal.
Prepare your data into a numpy array of shape (n_stars, n_features),
where the features are in RHS cartesian coordinates centred on the local
standard of rest (XYZUVW).
In any directory you can call:
>>> fit-component path/to/data.npy [path/to/comp_memberships.npy]
>>> fit-mixture NCOMPONENTS path/to/data.npy [path/to/mixture_memberships.npy]
or
>>> fit-chronostar path/to/data.npy
Where comp_memberships.npy is a stored numpy array of shape (n_stars) with
entries between 0 and 1, and mixture_memberships.npy has shape (n_stars, n_comps).