Getting started

Getting log data

from lammps_logfile import File

log = File("../../../examples/logfiles/crack_log.lammps")

t = log.get("Time")
temp = log.get("Temp")

Now the arrays t and temp contain the log data corresponding to the Time and Temp columns in the log file.

Plotting log data

import matplotlib.pyplot as plt 

plt.plot(t, temp)
plt.xlabel("Time (ps)")
plt.ylabel("Temperature (K)")
plt.ylim([215, 225])
plt.savefig("time_temp.png")

To make the plot pop up in a window rather than being saved to a file, run plt.show() rather than plt.savefig(…).

Running average

from lammps_logfile import running_mean

temp_avg = running_mean(temp, 100)

plt.plot(t, temp_avg)
plt.xlabel("Time (ps)")
plt.ylabel("Temperature (K)")
plt.ylim([215, 225])
plt.savefig("time_temp_avg.png")

What data are available in the log file?

To inspect what columns are available, you can run the get_keywords-method on the File object:

print(log.get_keywords())

This command yields an output like the one below, which shows what columns we may get from the File object:

['Density', 'E_angle', 'E_bond', 'E_coul', 'E_dihed', 'E_impro', 'E_long', 'E_vdwl', 'KinEng', 'Lx', 'Ly', 'Lz', 'PotEng', 'Press', 'Pxx', 'Pxy', 'Pxz', 'Pyy', 'Pyz', 'Pzz', 'Step', 'Temp', 'Time', 'TotEng', 'methaneM', 'waterMsd']

Command line tool

The following is the help message from the command line tool lammps_logplotter that comes with lammps-logplotter. This tool is meant to quicky inspect lammps log files without having to write a python script.

usage: lammps_logplotter [-h] [-x X] [-y Y [Y ...]] [-a RUNNING_AVERAGE] input_file

Plot contents from lammps log files

positional arguments:
input_file            Lammps log file containing thermo output from lammps simulation.

optional arguments:
-h, --help            show this help message and exit
-x X                  Data to plot on the first axis
-y Y [Y ...]          Data to plot on the second axis. You can supply several names to get several plot lines in the same figure.
-a RUNNING_AVERAGE, --running_average RUNNING_AVERAGE
                        Optionally average over this many log entries with a running average. Some thermo properties fluctuate wildly, and often we are interested in te
                        running average of properties like temperature and pressure.