Skip to main content

Python Gnuplot wrapper

Project description

PyGnuplot: Python wrapper for Gnuplot

Author: Ben Schneider


Gnuplot (


Using pip

pip install PyGnuplot

Using conda

conda install -c benschneider pygnuplot


pip install --upgrade  PyGnuplot

Basic Usage:

from PyGnuplot import gp
figure1 = gp()  # Create a new figure handle
figure2 = gp(r"C:\Program Files\gnuplot\bin\gnuplot.exe")  # Can also specify which gnuplot to use
figure1.a("plot sin(x)")
figure2.a("plot cos(x)")
pi = figure.a("print pi")

Functions available with each figure:


pipe a command to gnuplot as if in gnuplot command promt

c('plot sin(x)')

save(data, filename=’tmp.dat’)

save arrays into file (filename = ‘tmp.dat’) easily read by Gnuplot

s([X,Y,Z])  # creates tmp.dat
c('plot "tmp.dat" u 1:2')

a(command=’’, vtype=str, timeout=0.05)

asks gnuplot: it sends a command to gnuplot and returns its response This is paricularly handy when using gnuplots fitting features vtype can be used to change the return format timeout is the time to wait for a response

a('print pi')  # returns the value of pi
a('print pi; print pi')  # returns 2 times the value of pi

r(vtype=str, timeout=0.05)

reads the gnuplot return buffer until its empty

plot(data, filename=’tmp.dat’)

Plot some data. Sends plot instructions and the data to Gnuplot


plot_b(data, v1=’d’, v2=’%double’)

Similar to plot: Sends plot instructions and the data to Gnuplot However it sends them in binary format, which can be beneficial when the dealing with larger quanities of numbers

p(filename=’’, width=14, height=9, fontsize=12, term=’x11’)

Create postscript file (overwrites existing)


pdf(filename=’tmp.pdf’, width=14, height=9, fontsize=12, term=’x11’)

Create a pdf file (overwrites existing)



Closes windows,then gnuplot, then the pipe

Setup terminal

This script will use the same default terminal that gnuplot used (it reads the GPVAL_TERM value when gnuplot starts up) it can still be modified by the ‘default_term’ variable:

from PyGnuplot import gp
fig1 = gp()
fig1.default_term = 'wxt'

New features:

fit2d(data, func=’y(x)=a + b*x’, via=’a,b’, limit=1e-9)

Quickly Fit a simple 2-D data set and return the fitting results. This uses the new ask function “a()” Here we gather the fitting info from gnuplot


fit(self, data, func=’y(x)=a + b*x’, via=’a,b’, limit=1e-9, filename=’tmp.dat’, wait=1)

Allows for sligtly more complex fitting, filename: stores data first into a temporary file default: tmp.dat wait: define a waiting time in sec for gnuplot to finish its fitting default: 1sec

import numpy as np
f1 = gp()
x = np.linspace(0, 20, 1001)
yn = np.random.randn(1001)/10
y = np.sin(x)
data = [x, y+yn]
func = 'y(x) = a + b*cos(x + c)'  # define a fitting function here.
(a, b, c), report = f1.fit2d(data, func, via='a,b,c', limit=1e-9) # sending in the data the function used to fit and the variables that are to be found., "tmp.dat")
f1.a('plot "tmp.dat" w lp')
f1.a('replot y(x)')


  • 1 Example code

from PyGnuplot import gp
import numpy as np
X = np.arange(10)
Y = np.sin(X/(2*np.pi))
Z = Y**2.0
fig1 = gp()[X,Y,Z])
fig1.c('plot "tmp.dat" u 1:2 w lp)
fig1.c('replot "tmp.dat" u 1:3' w lp)
  • 2 Example file


Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

PyGnuplot-0.12.3.tar.gz (7.0 kB view hashes)

Uploaded source

Built Distribution

PyGnuplot-0.12.3-py3-none-any.whl (7.0 kB view hashes)

Uploaded py3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page