RAFFLE

by Ned Thaddeus Taylor, Joe Pitfield, and Steven Paul Hepplestone

RAFFLE (pseudoRandom Approach For Finding Local Energetic minima) is a package for structural prediction applied to material interfaces. RAFFLE interfaces with the Atomic Simulation Environment (ASE).

RAFFLE is both a Fortran and a Python library, with the option of a Fortran executable. The code heavily relies on features of Fortran 2018 and above, so there is no backwards compatibility with Fortran95.

Documentation

Tutorials and documentation are provided on the docs website. The methodology is detailed in the Phys Rev B paper. The paper detailing the software package is currently avaialble on arXiv with plans for peer-reviewed publication.

Refer to the API Documentation section later in this document to see how to access the API-specific documentation.

Requirements

• Fortran compiler supporting Fortran 2018 standard or later
• fpm or CMake (fpm works only for Fortran installation)

Python-specific installation:

• Python 3.11 or later (might work on earlier, have not tested)
• NumPy.f2py
• f90wrap
• cython
• scikit-build-core
• meson
• make or ninja
• CMake
• ASE

The library bas been developed and tested using the following Fortran compilers: - gfortran -- gcc 11.4.0 - gfortran -- gcc 13.2.0 - gfortran -- gcc 14.1.0 - gfortran -- gcc 14.2.0

The library is known to not currently work with the intel Fortran compilers.

Installation

For the Python library, the easiest method of installation is to install it directly from pip:

pip install raffle

Once this is done, RAFFLE is ready to be used.

Alternatively, to download development versions or, if, for some reason, the pip method does not work, then RAFFLE can be installed from the source. To do so, the source must be obtained from the git repository. Use the following commands to get started: git clone https://github.com/ExeQuantCode/raffle.git cd raffle

Depending on what language will be used in, installation will vary from this point.

Python

For Python, the easiest installation is through pip: pip install .

Another option is installing it through cmake, which involves: mkdir build cd build cmake .. make install

Then, the path to the install directory (${HOME}/.local/raffle) needs to be added to the include path. NOTE: this method requires that the user manually installs the ase, numpy and f90wrap modules for Python.

Fortran

For Fortran, either fpm or cmake are required.

fpm

fpm installation is as follows:

fpm build --profile release

This will install both the Fortran library and the Fortran application for RAFFLE. The library can then be called from other fpm-built Fortran programs through normal means (usually referencing the location of RAFFLE in the program's own fpm.toml file). The application can be run using (NOTE: The application is not a priority for development, so is less likely to work): fpm run

The library can be tested to ensure compilation was successful (NOTE: Unit tests currently only provide minimal code coverage): fpm test --profile release

cmake

cmake installation is as follows: mkdir build cd build cmake [-DBUILD_PYTHON=Off] .. make install The optional filed (dentoted with [...]) can be used to turn off installation of the Python library. This will build the library in the build/ directory. All library files will then be found in: ${HOME}/.local/raffle Inside this directory, the following files will be generated: include/raffle.mod lib/libraffle.a

To check whether RAFFLE has installed correctly and that the compilation works as expected, the following command can be run: ctest This runs the unit tests (found in the test directory) to ensure procedures output as expected.

MacOS

Issues can arise with the built-in C and Fortran compilers when installing on MacOS, particularly for Python installation. It is recommended to manually install new versions (such as via Homebrew) and setting the CC and FC environment variables (i.e. defining the default compilers for the respective language). We have found the following to work:

brew install gcc brew install gfortran export CC=$(brew --prefix gfortran) export FC=$(brew --prefix gcc)

Now follow the instructions for the Python build methods.

Examples

After the library has been installed, a set of example programs can be found in the example directory (note, the test directory is for unit tests to ensure each procedure in the library produces expected outputs after compilation, they are not really needed to be looked at by potential users).

The example/fortran_exe example uses a shell script to run the Fortran installed application. It uses a user-editable input file param.in in the same directory to change values of the RAFFLE generator and provided database.

The example/python_pkg directory contains a set of subdirectories MATERIAL_learn that show how RAFFLE can be called using Python to implement it into existing structure search workflows. These examples are the recommended ones to run. To successfully run them, follow the above installation instructions for Python, then go to the example/python_pkg directory and run one of the scripts.

API documentation

API documentation can be generated using FORD (Fortran Documenter). To do so, follow the installation guide on the FORD website to ensure FORD is installed. Once FORD is installed, run the following command in the root directory of the git repository:

ford ford.md

Contributing

Please note that this project adheres to the Contributing Guide. If you want to contribute to this project, please first read through the guide. If you have any questions, bug reports, or feature requests, please either discuss then in issues.

License

This work is licensed under a GPL v3 license.

Code Coverage

Automated reporting on unit test code coverage in the README is achieved through utilising the cmake-modules and dynamic-badges-action projects.

References

If you use this code, please cite our papers: ```text @article{Pitfield2024PredictingPhaseStability, title = {Predicting Phase Stability at Interfaces}, author = {Pitfield, J. and Taylor, N. T. and Hepplestone, S. P.}, journal = {Phys. Rev. Lett.}, volume = {132}, issue = {6}, pages = {066201}, numpages = {8}, year = {2024}, month = {Feb}, publisher = {American Physical Society}, doi = {10.1103/PhysRevLett.132.066201}, url = {https://link.aps.org/doi/10.1103/PhysRevLett.132.066201} }

@article{Taylor2025RAFFLEActiveLearning, title = {RAFFLE: Active learning accelerated interface structure prediction}, author = {Taylor, Ned Thaddeus and Pitfield, Joe and Davies, Francis Huw and Hepplestone, Steven Paul}, year = {2025}, eprint={2504.02528}, archivePrefix={arXiv}, primaryClass={cond-mat.mtrl-sci}, doi = {10.48550/ARXIV.2504.02528}, url = {https://arxiv.org/abs/2504.02528}, publisher = {arXiv}, } ```