README

Hey.

Thanks for using FitTeD.

The paper describing this package is here: https://ui.adsabs.harvard.edu/abs/2025MNRAS.544.2225M/abstract Please cite if you make use of this code in your research.

Setup

To get this all up and running, you will need to run the following:

• Pull into its own folder
• Within the folder, run "python3 -m pip install -e ."

That is the whole installation. FitTeD is pure Python: there is no compiler, no build step, and the same install works on Linux, macOS and Windows.

• The -e flag is optional. It makes the package editable in place; a plain "python3 -m pip install ." works identically.

To run the below examples you will need to also have manyTDE installed (Andy and Sjoert's database of optical/UV TDE light curves).

• This is only necessary if you want to get TDE data sets by IAU name, rather than loading your own data.
• manyTDE can be found here: https://github.com/sjoertvv/manyTDE
• Again just run “ python3 -m pip install -e . ” in the manyTDE directory to set up.

Running the code

There are 3 example scripts in the fitted/examples directory which run through how this all works.

Running

data_loading.py

Will set up a FitTeD Data_Set class for the tidal disruption event AT2019dsg and do some processing (see code/paper for details).

Running

fitting_models.py

will show you how to generate FitTeD models, and fit them to data in various ways.

Switch the variables yes_i_want_to_run_a_chain and yes_i_want_to_find_a_best_fit to True if you want to do a proper analysis, although running the chain will take ~ 5 hours with current settings (on my laptop).

If you do run the chain, then

analysis.py

shows you some science results you can get from the fit.

Happy TDE-ing.

Cheers, Andy*, Ed, and Adam

P.S., any comments/questions, drop me a line.

• amummery@ias.edu