Skip to content

IndiraOcampo/CMB_ML_based_model_selection

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

60 Commits
 
 
 
 
 
 

Repository files navigation

CMB Machine Learning based model selection and interpretability

License Badge arXiv DOI DOI DOI

Table of Contents


Overview

Repository showing how to use Neural Networks (NN) to test beyond ΛCDM models against Cosmic Microwave Background (CMB) data, at the level of the angular power spectra. All the analysis is found within the corresponding jupyter notebooks.

Description

The repository is structured as follows. Within the neural_networks folder, you can find the folders:

  • modified_gravity
  • feature

Each folder contain the corresponding jupyter notebooks

Installation

Prerequisites

  • Required software: python
  • Dependencies: numpy, matplotlib, tensorflow, shap
  • It is recommended to run the notebooks to train the Neural Networks in a cluster with GPUs
  • The data to train the Neural Networks are found at these Zenodo repositories: modified gravity and primordial feature.

How to get started

# Example to get it running
pip install numpy matplotlib tensorflow
git clone https://github.com/IndiraOcampo/CMB_ML_based_model_selection.git
cd neural_networks

Usage

If you are using the content provided in this repository to do your own analysis, please cite this repository and the manuscript:

@misc{CMB_ML_based_model_selection,
  author = {Ocampo, I},
  title = {CMB ML based model selection},
  year = {2024},
  publisher = {GitHub},
  journal = {GitHub repository},
  howpublished = {\url{https://github.com/IndiraOcampo/CMB_ML_based_model_selection.git}},
}
@article{Ocampo_2025,
doi = {10.1088/1475-7516/2025/02/004},
url = {https://dx.doi.org/10.1088/1475-7516/2025/02/004},
year = {2025},
month = {feb},
publisher = {IOP Publishing},
volume = {2025},
number = {02},
pages = {004},
author = {Ocampo, I. and Cañas-Herrera, G. and Nesseris, S.},
title = {Neural Networks for cosmological model selection and feature importance using Cosmic Microwave Background data},
journal = {Journal of Cosmology and Astroparticle Physics}.
}

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgements

This reaseach acknowledges support from the ESA Archival Research Visitor Programme: a programme to increase the scientific return from ESA space science missions by supporting scientists interested in pursuing research based on publicly available data in the ESA Space Science Archives. This work is also supported by the fellowship LCF/BQ/DI22/11940033 from “la Caixa” Foundation (ID 100010434) and acknowledges the use of the Finis Terrae III Supercomputer which was financed by the Ministry of Science and Innovation, Xunta de Galicia and ERDF (European Regional Development Fund). This research uses ESA Planck Legacy archives.

About

Neural networks for model selection and feature importance at the level of the CMB angular power spectra for JCAP02(2025)004 (arXiv:2410.05209)

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors