LAMMPS¶

UPET models can drive LAMMPS molecular dynamics via pair_style metatomic, on both CPU and GPU (with KOKKOS).

1. Install LAMMPS with metatomic support¶

Follow the metatomic LAMMPS install instructions. We recommend the pre-built conda binaries.

2. Export the UPET model¶

Fetch and convert a UPET checkpoint from the HuggingFace repository:

mtt export https://huggingface.co/lab-cosmo/upet/resolve/main/models/pet-mad-s-v1.5.0.ckpt -o model.pt

This downloads the model and converts it to a TorchScript file compatible with LAMMPS, using metatomic and metatrain under the hood. Other UPET models can be prepared in the same way:

mtt export https://huggingface.co/lab-cosmo/upet/resolve/main/models/pet-omat-xs-v1.0.0.ckpt -o model.pt
mtt export https://huggingface.co/lab-cosmo/upet/resolve/main/models/pet-omatpes-l-v0.1.0.ckpt -o model.pt

3. CPU version¶

Prepare a LAMMPS input file using pair_style metatomic and defining the mapping from LAMMPS atom types in the data file to the element UPET expects, using pair_coeff. Below, LAMMPS atom type 1 is silicon (atomic number 14). The silicon.in and silicon.data files are also checked into the repository and can be downloaded directly:

units metal
atom_style atomic

read_data silicon.data

pair_style metatomic model.pt device cuda  # change device to 'cpu' to use the CPU instead
pair_coeff * * 14

neighbor 2.0 bin
neigh_modify one 100000 page 1000000 binsize 5.5

timestep 0.001

dump myDump all xyz 10 trajectory.xyz
dump_modify myDump element Si

thermo_style multi
thermo 1

velocity all create 300 87287 mom yes rot yes

fix 1 all nvt temp 300 300 0.10
run_style verlet


run 100

Create the silicon.data data file:

# LAMMPS data file for Silicon unit cell
atoms
atom types

0  5.43  xlo xhi
0  5.43  ylo yhi
0  5.43  zlo zhi

Masses

28.084999992775295 # Si

Atoms # atomic

 1   0   0   0
 1   1.3575   1.3575   1.3575
 1   0   2.715   2.715
 1   1.3575   4.0725   4.0725
 1   2.715   0   2.715
 1   4.0725   1.3575   4.0725
 1   2.715   2.715   0
 1   4.0725   4.0725   1.3575

Run LAMMPS:

lmp -in lammps.in                # serial version
mpirun -np 1 lmp -in lammps.in   # MPI version

Warning

The neigh_modify settings above are particularly important for running PET-MAD v1.5 models, especially for dense systems with a large number of neighbors. This is due to the adaptive cutoff strategy, which requires a large initial buffer of neighbors before truncation. If your system is extremely dense, you may need to increase the one and page parameters even further to avoid neighbor-list overflow. Finally, binsize is important for the KOKKOS version of the code to avoid a bug with an empty neighbor list.

4. KOKKOS-enabled GPU version¶

Running LAMMPS with KOKKOS and GPU support is similar to the CPU version, but with a few additional flags. The lammps.in and silicon.data files are unchanged:

lmp -in in.lammps -k on g 1 -pk kokkos newton on neigh half -sf kk                 # serial
mpirun -np 1 lmp -in in.lammps -k on g 1 -pk kokkos newton on neigh half -sf kk    # MPI

The -k on g 1 -sf kk flags activate the KOKKOS subroutines. g 1 specifies how many GPUs the simulation is parallelized over — adjust it if you run a large system on multiple GPUs.

Important notes¶

For CPU calculations, use a single MPI task unless simulating large systems (30+ Å box size). Multi-threading can be enabled with:
```
export OMP_NUM_THREADS=4
```
For GPU calculations, use one MPI task per GPU.