File: README.fix_python_move

package info (click to toggle)
lammps 20220106.git7586adbb6a%2Bds1-2
  • links: PTS, VCS
  • area: main
  • in suites: bookworm
  • size: 348,064 kB
  • sloc: cpp: 831,421; python: 24,896; xml: 14,949; f90: 10,845; ansic: 7,967; sh: 4,226; perl: 4,064; fortran: 2,424; makefile: 1,501; objc: 238; lisp: 163; csh: 16; awk: 14; tcl: 6
file content (16 lines) | stat: -rw-r--r-- 854 bytes parent folder | download | duplicates (4)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
This folder contains several LAMMPS input scripts and a python module
file py_nve.py to demonstrate the use of the fix style python/move
to reimplement NVE using Python.

in.fix_python_move_nve_melt:
This is a version of the melt example which replaces the default NVE integrator
with a Python implementation. Fix python/move is used to create an
instance of the py_nve.NVE class which implements the required interface.
It demonstrates how to access LAMMPS data as numpy arrays. This gives direct
access to memory owned by the C++ code, allows easy manipulation through numpy
operations and avoids unnecessary copies.

in.fix_python_move_nve_melt_opt:
This version of melt example uses NVE_Opt instead of NVE. While this Python
implementation is still much slower than the native version, it shows that
simple code transformations can lead to speedups.