How to get matplotlib and latex work together?

Question

I have the following problem and I am totally new to matplotlib and python: To produce high qualitiy plots for my thesis I want to use matplotlib and I want to use the latex commands for the axes etc. When I try the following example (from http://matplotlib.org/users/usetex.html#usetex-tutorial):

import numpy as np 
import matplotlib.pyplot as plt


# Example data 
t = np.arange(0.0, 1.0 + 0.01, 0.01) 
s = np.cos(4 * np.pi * t) + 2

plt.rc('text', usetex=True) 
plt.rc('font', family='serif') 
plt.plot(t, s)

plt.xlabel(r'\textbf{time} (s)')
plt.ylabel(r'\textit{voltage} (mV)',fontsize=16) 
plt.title(r"\TeX\ is Number "
          r"$\displaystyle\sum_{n=1}^\infty\frac{-e^{i\pi}}{2^n}$!",
          fontsize=16, color='gray')
# Make room for the ridiculously large title. 
plt.subplots_adjust(top=0.8)

plt.savefig('tex_demo') 
plt.show()

I get the following message:

  File "test.py", line 21, in <module>
    plt.savefig('tex_demo')
  File "/usr/lib/pymodules/python2.7/matplotlib/pyplot.py", line 561, in savefig
    return fig.savefig(*args, **kwargs)
  File "/usr/lib/pymodules/python2.7/matplotlib/figure.py", line 1421, in savefig
    self.canvas.print_figure(*args, **kwargs)
  File "/usr/lib/pymodules/python2.7/matplotlib/backend_bases.py", line 2220, in print_figure
    **kwargs)
  File "/usr/lib/pymodules/python2.7/matplotlib/backends/backend_agg.py", line 505, in print_png
    FigureCanvasAgg.draw(self)
  File "/usr/lib/pymodules/python2.7/matplotlib/backends/backend_agg.py", line 451, in draw
    self.figure.draw(self.renderer)
  File "/usr/lib/pymodules/python2.7/matplotlib/artist.py", line 55, in draw_wrapper
    draw(artist, renderer, *args, **kwargs)
  File "/usr/lib/pymodules/python2.7/matplotlib/figure.py", line 1034, in draw
    func(*args)
  File "/usr/lib/pymodules/python2.7/matplotlib/artist.py", line 55, in draw_wrapper
    draw(artist, renderer, *args, **kwargs)
  File "/usr/lib/pymodules/python2.7/matplotlib/axes.py", line 2086, in draw
    a.draw(renderer)
  File "/usr/lib/pymodules/python2.7/matplotlib/artist.py", line 55, in draw_wrapper
    draw(artist, renderer, *args, **kwargs)
  File "/usr/lib/pymodules/python2.7/matplotlib/axis.py", line 1105, in draw
    self.label.draw(renderer)
  File "/usr/lib/pymodules/python2.7/matplotlib/artist.py", line 55, in draw_wrapper
    draw(artist, renderer, *args, **kwargs)
  File "/usr/lib/pymodules/python2.7/matplotlib/text.py", line 594, in draw
    self._fontproperties, angle, mtext=self)
  File "/usr/lib/pymodules/python2.7/matplotlib/backends/backend_agg.py", line 241, in draw_tex
    self._renderer.draw_text_image(Z, x, y, angle, gc)
OverflowError: cannot convert float infinity to integer

I'm new to Linux (Ubuntu 14.04) and maybe have done something wrong when installing matplotlib; here is what I have done to install matplotlib etc.:

sudo apt-get install python3
sudo apt-get install python-matplotlib
sudo apt-get build-dep python-matplotlib

I am using texmaker and thought this could be the problem; so I did the following:

sudo aptitude install texlive-fonts-recommended texlive-fonts-extra

If I try any other plots including latex commands I get similar error messages, or plots without any lables for the axes.

Answer 1

Your problem has been seen before as reported here on the matplotlib-user mailing list. The resolution is here and the confirmation that it works is here.

In summary, you need to clear the tex.cache directory.

To find where that directory is - go into python and execute matplotlib.get_cachedir()

For a standard Ubuntu 14.04 install - you should get something like the following:

~$ python3
Python 3.4.0 (default, Apr 11 2014, 13:05:11) 
[GCC 4.8.2] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import matplotlib as mpl
>>> mpl.get_cachedir()
'/home/<username>/.cache/matplotlib'

Then do rm <path/to/cache/from/above>/tex.cache/* and then re-run and it should all work.

---

Docs issue

Some tips like this are available in the usetex troubleshooting docs section, which lead to here to locate the directory, but they unfortunately don't mention get_cachedir()

Edit Pull Request generated based on this question and merged into the matplotlib docs.