Code, Datasets, Model Weights and Source Data for: Efficient equivariant framework for universal charge density prediction

This record contains the code, datasets, pretrained model weights, and source data used in the study “Efficient equivariant framework for universal charge density prediction.” The code.zip file includes the source code of EdenGNN. The code is also available at https://github.com/rubenlee11/EdenGNN. A runnable example is provided via Google Colab at https://colab.research.google.com/drive/1tSGPZk4XI71GEylYeNDD1218smOEaFKc?usp=sharing. The data.zip file includes the POTCAR versions and hashes, the dataset splits, and example CHGCAR files of superposition of atomic charge density and self-consistent density. The model_weight.zip file includes the model weights and config.yaml files. The source_data.zip file includes source files for figure generation in the main text, and 1,000 band-comparison plots for test structures from the Materials Project database.