Skip to content

Equivariant machine learning model for predicting self-consistent Hubbard parameters

License

Notifications You must be signed in to change notification settings

camml-lab/hubbardml

Repository files navigation

HubbardML

Code style: black Commitizen friendly security: bandit DeepSource

About

This repository contains source code for our machine learning model for predicting self-consistent Hubbard parameters, as presented in this work:

Uhrin, M., Zadoks, A., Binci, L., Marzari, N., & Timrov, I. (2025). Machine learning Hubbard parameters with equivariant neural networks. Npj Computational Materials, 11(1), 19. https://doi.org/10.1038/s41524-024-01501-5i

The experiments carried out in this work can be found in the experiments/ folder along with all the notebooks to generate the plots.

As an example, from experiments you can use:

python run.py experiment=predict_hp model=u

to run an experiment that trains a model to predict Hubbard U values from a linear-response dataset.

Additional experiments can be found in the experiments/experiment/ folder.

About

Equivariant machine learning model for predicting self-consistent Hubbard parameters

Resources

License

Stars

Watchers

Forks

Packages

No packages published