CederGroupHub / chgnet

Pretrained universal neural network potential for charge-informed atomistic modeling https://chgnet.lbl.gov
https://doi.org/10.1038/s42256-023-00716-3
Other
220 stars 60 forks source link

Fine tuning CHGNet with DFT energies and forces #61

Closed Atefeh-Yadegarifard closed 1 year ago

Atefeh-Yadegarifard commented 1 year ago

Dear CHGNet developers, Hello. I would like to request an example for how CHGNet can be trained with additional DFT data (calculated by the users). Specifically the format of the data that has to be used for training. Thank you.

BowenD-UCB commented 1 year ago

For fine-tuning code, please refer to: https://github.com/CederGroupHub/chgnet/blob/main/examples/fine_tuning.ipynb

For parsing VASP output into fine-tuning dataset, please refer to: https://github.com/CederGroupHub/chgnet/blob/main/chgnet/utils/vasp_utils.py