Hi everyone,
We’ve recently released RAFFLE, an open source Python/Fortran library for predicting interface structures.
RAFFLE takes an iterative approach to structure search by generating candidate interfaces, learning their energetics, and refining its atom placement rules for subsequent iterations. This active learning strategy improves the efficiency of sampling interface energy landscapes, biasing the search towards low-energy configurations.
I wanted to share it here in case it’s useful to anyone. We’re actively developing the library and exploring new features, so feedback and contributions are very welcome!
Documentation & Tutorials: RAFFLE — RAFFLE documentation
RAFFLE is developed by Ned Thaddeus Taylor, Joe Pitfield, and Steve Hepplestone.
Feel free to ask any questions or share any suggestions!