\(\renewcommand{\AA}{\text{Å}}\)

2.3.6. System properties

Similar to what is described in System properties, the instances of lammps, PyLammps, or IPyLammps can be used to extract different kinds of information about the active LAMMPS instance and also to modify some of it. The main difference between the interfaces is how the information is exposed.

While the lammps is just a thin layer that wraps C API calls, PyLammps and IPyLammps expose information as objects and properties.

In some cases the data returned is a direct reference to the original data inside LAMMPS cast to ctypes pointers. Where possible, the wrappers will determine the ctypes data type and cast pointers accordingly. If numpy is installed arrays can also be extracted as numpy arrays, which will access the C arrays directly and have the correct dimensions to protect against invalid accesses.

Warning

When accessing per-atom data, please note that this data is the per-processor local data and indexed accordingly. These arrays can change sizes and order at every neighbor list rebuild and atom sort event as atoms are migrating between subdomains.

from lammps import lammps

lmp = lammps()
lmp.file("in.sysinit")

natoms = lmp.get_natoms()
print(f"running simulation with {natoms} atoms")

lmp.command("run 1000 post no");

for i in range(10):
   lmp.command("run 100 pre no post no")
   pe = lmp.get_thermo("pe")
   ke = lmp.get_thermo("ke")
   print(f"PE = {pe}\nKE = {ke}")

lmp.close()

Methods:

version(): return the numerical version id, e.g. LAMMPS 2 Sep 2015 -> 20150902
get_thermo(): return current value of a thermo keyword
last_thermo(): return a dictionary of the last thermodynamic output
get_natoms(): total # of atoms as int
reset_box(): reset the simulation box size
extract_setting(): return a global setting
extract_global(): extract a global quantity
extract_box(): extract box info
create_atoms(): create N atoms with IDs, types, x, v, and image flags

Properties:

last_thermo_step: the last timestep thermodynamic output was computed

In addition to the functions provided by lammps, PyLammps objects have several properties which allow you to query the system state:

L.system: Is a dictionary describing the system such as the bounding box or number of atoms
L.system.xlo, L.system.xhi: bounding box limits along x-axis
L.system.ylo, L.system.yhi: bounding box limits along y-axis
L.system.zlo, L.system.zhi: bounding box limits along z-axis
L.communication: configuration of communication subsystem, such as the number of threads or processors
L.communication.nthreads: number of threads used by each LAMMPS process
L.communication.nprocs: number of MPI processes used by LAMMPS
L.fixes: List of fixes in the current system
L.computes: List of active computes in the current system
L.dump: List of active dumps in the current system
L.groups: List of groups present in the current system

Retrieving the value of an arbitrary LAMMPS expressions

LAMMPS expressions can be immediately evaluated by using the eval method. The passed string parameter can be any expression containing global thermo command values, variables, compute or fix data (see Output from LAMMPS (thermo, dumps, computes, fixes, variables)):

result = L.eval("ke") # kinetic energy
result = L.eval("pe") # potential energy

result = L.eval("v_t/2.0")

Example

from lammps import PyLammps

L = PyLammps()
L.file("in.sysinit")

print(f"running simulation with {L.system.natoms} atoms")

L.run(1000, "post no");

for i in range(10):
   L.run(100, "pre no post no")
   pe = L.eval("pe")
   ke = L.eval("ke")
   print(f"PE = {pe}\nKE = {ke}")