BAND-NN

BAND-NN is a deep learning architecture for prediction of atomization energy and for geometry optimization.

Requirements:

• Python3
• Keras
• Tensorflow
• Numpy
• Scipy

Usage:

• Examples for prediction and geometry optimization are provided in examples/ folder.
• Required scripts in scripts/ folder.
• Trained models and weights are provided in models/ and weights/ folders.