What AI says about AugLiChem

AugLiChem is a Python-based data augmentation library for chemical structures (molecules and crystalline materials) designed to improve machine learning model performance. It provides predefined augmentations for molecular systems (random atom masking, random bond deletion, substructure removal) and crystalline systems (random perturbation, random rotation, swap axes, random translation, supercell transformations), along with automatic dataset downloading/cleaning and pre-implemented models for fingerprint-based ML and graph neural networks. The project is developed by Barati Lab at Carnegie Mellon University, with an MIT license and a published paper.