These examples are meant to used with the ML-POD package from the following Git repo
https://github.com/cesmix-mit/lammps/tree/fastpod
Please download the above source code and install LAMMPS by using the commandlines
mkdir build
cd build
cmake -C ../cmake/presets/basic.cmake -D BUILD_SHARED_LIBS=on -D LAMMPS_EXCEPTIONS=on -D PKG_PYTHON=on -D PKG_ML-POD=on ../cmake
cmake --build .
Alternatively, you can copy the ML_POD source in the directory https://github.com/cesmix-mit/lammps/tree/fastpod/src/ML-POD to your existing LAMMPS source code and compile your LAMMPS source code.
TrainingData contains training datasets for many different materials
TestData contains test datasets for some materials
FastPOD include many examples on fitting and running MD with ML-POD in LAMMPS.
To fit FPOD potential, please go to a subfolder in FastPOD and do
lmp -in fit.pod
To run a MD simulation with the trained FPOD potential, please do
lmp -in mdrun.pod