#Build note for lammps with compute_pace (TO BE UPDATED - SEE README)
First download lammps from git
git clone -b unstable https://github.com/lammps/lammps.git mylammps_pace
or with SSH
git clone -b unstable [email protected]:lammps/lammps.git mylammps_pace
Next, set up initial lammps build by downloading the compute_pace enabled lammps-user-pace package move this package to your build directory
cd lammps-user-pace chmod +x setup.sh ./setup.sh
Now use cmake to install the default version of ML-PACE (for fitsnap compatibility, use the following flags) – Make sure you are in the build folder first, then execute:
export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/path/to/mylammps_pace/build
cmake -D LAMMPS_EXCEPTIONS=on -D PKG_PYTHON=on -D BUILD_SHARED_LIBS=on -D CMAKE_BUILD_TYPE=Release -D PKG_ML-IAP=on -D PKG_ML-PACE=on -D PKG_ML-SNAP=on -D BUILD_MPI=on -D BUILD_OMP=off ../cmake/
(Note: this may result in a hash mismatch error for the pace library. If this results in an MD5 hash mismatch error, then look in the error message for the actual hash, copy it, and replace the corresponding line in the ML-PACE cmake file (found in ../cmake/Modules/Packages/ML-PACE.cmake) If you are installing on a SRN cluster, there may be problems with downloading the libpace tarball from http. If you get an error “no rule to make pace”, then change the address in the ML-PACE.cmake file: https://github.com/ICAMS/... ->[email protected]:ICAMS/...
Modules and settings used for tested build: (blake)
module load intel/compilers/20.2.254 module load openmpi/4.0.5/intel/20.2.254 module load cmake/3.12.3
anaconda (python 3.8.8)