0

I have looked at this question and also this one. It looks like for me, matplotlib.pyplot.show() shows a figure from python, but not from jupyter console.

matplotlib.matplotlib_fname() returns the same matplotlibrc file location for both.

However, when I try to find the backend being used with matplotlib.rcParams['backend'] jupyter console tells me - 'module://ipykernel.pylab.backend_inline', regardless of which backend I have modified the matplotlibrc file to use.

Python on the other hand, correctly shows the backend I'm using; currently 'TkAgg'.

I installed matplotlib using python -mpip install -U matplotlib.

I'm using the following versions:

  • Windows 10
  • Jupyter console 5.2.0
  • Python 2.7.14
  • IPython 5.5.0

I can make do with using python, but it would be nice to figure this out for jupyter console as well.

Aniket Deshpande
  • 59
  • 2
  • 9

1 Answers1

2

First note that plt.show() works as expected, also in Juypter.

enter image description here

This uses the default 'module://ipykernel.pylab.backend_inline' backend. This backend is set by Jupyter, independently of the rcParams setting.

You may set the backend using matplotlib.use() 

import matplotlib
matplotlib.use("TkAgg")
import matplotlib.pyplot as plt
plt.plot([1,2,3])
plt.show()

or just using IPython magic %matplotlib backendname

%matplotlib tk
import matplotlib.pyplot as plt
plt.plot([1,2,3])
plt.show()

You may change the backend using pyplot.switch_backend()

plt.switch_backend("TkAgg")
plt.plot([1,2,3])
plt.show()

or using the same IPython magic

%matplotlib tk
plt.plot([1,2,3])
plt.show()

If you want to set the backend to be used by default, see this question: Change default backend for matplotlib in Jupyter Ipython

ImportanceOfBeingErnest
  • 251,038
  • 37
  • 461
  • 518