Metadata-Version: 2.1
Name: fiestaEM
Version: 0.0.2
Summary: Fast inference of electromagnetic signals with JAX
Home-page: https://github.com/ThibeauWouters/fiestaEM
Author: Thibeau Wouters
Author-email: thibeauwouters@gmail.com
License: MIT
Keywords: sampling,inference,astrophysics,kilonovae,gamma-ray bursts
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: jax<=0.4.31
Requires-Dist: jaxlib<=0.4.31
Requires-Dist: numpy<2.0.0
Requires-Dist: pandas<2.0.0
Requires-Dist: jaxtyping
Requires-Dist: beartype
Requires-Dist: tqdm
Requires-Dist: scipy<=1.14.0
Requires-Dist: ml_collections
Requires-Dist: astropy
Requires-Dist: sncosmo
Requires-Dist: flowMC
Requires-Dist: joblib
Requires-Dist: h5py

# fiesta 🎉

`fiesta`: **F**ast **I**nference of **E**lectromagnetic **S**ignals and **T**ransients with j**A**x

![fiesta logo](docs/fiesta_logo.jpeg)

**NOTE:** `fiesta` is currently under development -- stay tuned!

## Installation

pip installation is currently work in progress. Install from source by cloning this Github repository and running
```
pip install -e .
```

NOTE: This is using an older and custom version of `flowMC`. Install by cloning the `flowMC` version at [this fork](https://github.com/ThibeauWouters/flowMC/tree/fiesta) (branch `fiesta`).

## Training surrogate models

To train your own surrogate models, have a look at some of the example scripts in the repository for inspiration, under `trained_models`

- `train_Bu2019lm.py`: Example script showing how to train a surrogate model for the POSSIS `Bu2019lm` kilonova model. 
- `train_afterglowpy_tophat.py`: Example script showing how to train a surrogate model for `afterglowpy`, using a tophat jet structure.  

## Examples

- `run_AT2017gfo_Bu2019lm.py`: Example where we infer the parameters of the AT2017gfo kilonova with the `Bu2019lm` model.
- `run_GRB170817_tophat.py`: Example where we infer the parameters of the GRB170817 GRB with a surrogate model for `afterglowpy`'s tophat jet. **NOTE** This currently only uses one specific filter. The complete inference will be updated soon.

## Acknowledgements

The logo was created by [ideogram AI](https://ideogram.ai/). 
