For practical reasons, I want to test a small piece of Pyton code on repl.it (webbased, so I do not need to install Python).

The code

import numpy as np
import matplotlib.pyplot as plttime = np.array([0, 1, 2, 3])
vec = np.array([1, 4, 3, -2])plt.plot(time, vec)

gives following error:

[GCC 4.8.2] on linuxTraceback (most recent call last):File "main.py", line 16, in <module>plt.figure()File "/goval_modules/python35/matplotlib/pyplot.py", line 535, in figure**kwargs)File "/goval_modules/python35/matplotlib/backends/backend_tkagg.py", line 84, in new_figure_managerreturn new_figure_manager_given_figure(num, figure)File "/goval_modules/python35/matplotlib/backends/backend_tkagg.py", line 92, in new_figure_manager_given_figurewindow = Tk.Tk()File "/usr/local/lib/python3.5/tkinter/__init__.py", line 1868, in __init__self.tk = _tkinter.create(screenName, baseName, className, interactive, wantobjects, useTk, sync, use)
_tkinter.TclError: no display name and no $DISPLAY environment variable
exited with non-zero status

What does this mean?

matplotlib by default picks an interactive backend, showing the plot in a GUI; the default is to use TkInter. See What is a backend in the matplotlib FAQ. That's not going to work when Python is run on a webserver, there is no GUI server to display this on.

You'll have to pick a non-interactive backend; Agg is supported and produces PNGs. Repl.it allows you to save files 'locally' ^(*):

import matplotlib as mpl
mpl.use('Agg')
import numpy as np
import matplotlib.pyplot as plttime = np.array([0, 1, 2, 3])
vec = np.array([1, 4, 3, -2])plt.plot(time, vec)
plt.savefig('graph.png')

When you run this, an extra graph.png tab appears that lets you download the result. See the live demo; the above produces:

^(*) For new scripts, enable project mode first by clicking on the new file icon. It looks like a document-with-plus-sign and sits underneath the 'settings' cog icon, above the line numbers of the editor.