SIMPLE-NN uses ASE to handle output from ab initio programs like VASP or Quantum espresso.
All output types supported by ASE can be used in SIMPLE-NN, but they need to contain essential information such as atom coordinates, lattice parameters, energy, and forces. (Release_notes)
- Tensorflow >= r1.6.
- PYTHON - 2.7.X or 3.4 - 3.6 (http://www.python.org/)
- mpi4py (optional)
The BSD 3-Clause Licence
How to cite:
If you used SIMPLE-NN for your work, please cite the following article.
- "Toward Reliable and Transferable Machine Learning Potentials: Uniform Training by Overcoming Sampling Bias"
Wonseok Jeong, Kyuhyun Lee, Dongsun Yoo, Dongheon Lee, and Seungwu Han, J. Phys. Chem. 7, pp 22790-22795 (2018)