cheukhinhojerry / parareal_md

Description:

This page contains the code from the paper

"Combining machine-learned and classical force fields with the parareal algorithm: application to the diffusion of atomistic defects"

by Olga Gorynina, Frédéric Legoll, Tony Lelièvre and Danny Perez

Dependencies:

-Python 3.8+

-Compile LAMMPS as a shared library with python support: https://docs.lammps.org/Python_install.html

Running:

python run.py [options]

--N: Number of time steps in parareal algorithm (default value 2000)

--fine: choice of potential for fine propagator (default value 14, corresponds to SNAP-205 potential)

availible potentials:

14 - SNAP-205

12 - SNAP-141

10 - SNAP-92

8 -  SNAP-56

6 -  SNAP-31

4 -  SNAP-15

2 -  SNAP-6

0 -  EAM potential

--coarse: choice of potential for coarse propagator (default value 0, corresponds to EAM potential)

--dc: convergence parameter (default value 10^(-5))

--de: explosion threshold parameter, de > dc (default value 0.35)