relentless

Overview

relentless is a Python package for executing molecular simulations in larger computational workflows. The simulation protocol is specified as a reproducible, human-readable recipe that is run natively in popular engines such as LAMMPS and HOOMD-blue. relentless also has robust features for optimization with simulations, including independent and dependent variables with automatic differentiation, objective functions of simulations, and optimization methods. All features are readily extensible through a clean hierarchy of abstract objects, enabling you to quickly use your own simulation code or optimization objective. Our initial goal was to enable optimization of the relative entropy of structural coarse-graining and materials design: with less code.

Resources

• Documentation: Installation, examples, and commands.
• Source code: Download or contribute on GitHub.
• Issue tracker: Report issues or request features.

Installation

Install relentless from PyPI using pip:

pip install relentless

or from conda-forge using conda:

conda install -c conda-forge relentless

Contributing

Contributions are welcomed and appreciated! Fork and create a pull request on GitHub. We value the input and experiences all users and contributors bring to relentless.

Citing relentless

When using relentless in published work, please cite:

A. N. Sreenivasan, C. L. Petix, Z. M. Sherman, and M. P. Howard. relentless: Transparent, reproducible molecular dynamics simulations for optimization. The Journal of Chemical Physics 161, 212502 (2024). doi:10.1063/5.0233683