BETE-NET

This repo contains the models, training data and test predictions for the work described in the paper: Accelerating superconductor discovery through tempered deep learning of the electron-phonon spectral function

Files and directories

The CSO/ CPD/ and FPD/ directories contain the trained models described in the paper.

The database.json file contains the database used to train the model.

The structures directory contains the corresponding cif files.

The indices/ directory contain the training indices/idx_train_full.txt, and testing indices/idx_test_full.txt indices as well as the indices used for bootstrapping.

The notebook/ directory contains notebooks to visualize the predictions, train the models and make predictions on a dataset.

The test_preds/ directory contain the predictions published in the paper.

Prerequisites

This package requires:

• PyTorch
• scikit-learn
• pymatgen

If you are new to Python, the easiest way of installing the prerequisites is via conda. After installing conda, run the following command to create a new environment named bete_net and install all prerequisites:

conda create --name bete_net python=3.9
conda activate bete_net
conda install pytorch==1.10.0 torchvision==0.11.0 torchaudio==0.10.0 cudatoolkit=11.3 -c pytorch -c conda-forge
pip install -r requirements.txt -f https://pytorch-geometric.com/whl/torch-1.10.0+cu113.html

Paper

Our paper can be found here

Citation

If you use the code in your work, please cite:

@misc{gibson2024acceleratingsuperconductordiscoverytempered,
      title={Accelerating superconductor discovery through tempered deep learning of the electron-phonon spectral function}, 
      author={Jason B. Gibson and Ajinkya C. Hire and Philip M. Dee and Oscar Barrera and Benjamin Geisler and Peter J. Hirschfeld and Richard G. Hennig},
      year={2024},
      eprint={2401.16611},
      archivePrefix={arXiv},
      primaryClass={cond-mat.supr-con},
      url={https://arxiv.org/abs/2401.16611}, 
}