Utilities for converting data to and from LAMMPS dump formats.
Just clone and pip install as below:
git clone git@github.com:chrisk314/lammps-utils.git
cd lammps-utils
pip install -e .Installing in develop mode allows you to add your own custom dump loader functions. See the docstrings! ;-)
Data can be loaded from a specified LAMMPS dump file using the load_from_dump_file function .
from lammps_utils import load
fname = './dump.connectivity'
data = load.load_from_dump_file(fname)The above code will extract all data from the dump file into a list containing data for each time step in order. The data has the following format:
[
(
3000000, # timestep
383, # number of particles
array([ 0. , 0.0067592, 0. , 0.0067592, # domain dimensions
0. , 0.0067592]),
array([ # particle data
[ 0.00000000e+00, 5.59600000e-04, 8.75341000e-04,
4.43585000e-05, 1.00528000e-03],
[ 0.00000000e+00, 2.56164000e-04, 3.86693000e-04,
5.50167000e-04, 5.09240000e-04],
...
])
),
...
]
By default the format of the atom data in the dump file will be inferred by checking the line specifying the atom fields in the file such as this one:
ITEM: ATOMS id x y z diameter v_density c_conn
A structured array will be produced allowing access to the data by field name.
atom_data = data[0][3] # Get atom data from first time step
x = atom_data['x']
diam = atom_data['diameter']Often it is preferable to work with an ndarray object instead. There is a utility function to perform the
conversion.
atom_data = load.as_ndarray(atom_data) # Convert atom data to ndarray format