This project uses Graph Neural Networks (GNNs) to predict the frequency-dependent dielectric function from crystal structures. The data and model provided in this repository support the findings presented in the paper:
"Prediction of Frequency-Dependent Optical Spectrum for Solid Materials: A Multioutput and Multifidelity Machine Learning Approach" (https://pubs.acs.org/doi/10.1021/acsami.4c07328).