Jupyter making 3D matplotlib graphs extremely small

Question

Having read many of the posts on this site about resizing graphs and setting limits on graph sizes in Jupyter, I am virtually convinced there is something different when it

Answer 1

Try replacing

%pylab inline
pylab.rcParams['figure.figsize'] = (20, 16)
pylab.rcParams['figure.dpi'] = 200

import matplotlib.pyplot as plt
import matplotlib

with

%matplotlib inline

import matplotlib.pyplot as plt
import matplotlib

matplotlib.rcParams['figure.figsize'] = (20, 16)
matplotlib.rcParams['figure.dpi'] = 200

Answer 2

Sometimes that happens to me as well on my Mac.

First use this:

import matplotlib.pyplot as plt
import matplotlib
%matplotlib inline

matplotlib.rcParams['figure.figsize'] = (20, 16)
matplotlib.rcParams['figure.dpi'] = 200

The trick for my case: First import and then use the %matplotlib inline command. However, seems like a bug.

Answer 3

This is due to a bug in matplotlib 3.0.0. It should not occur in matplotlib 3.0.1.

Options you have:

• Update to matplotlib 3.0.1

• Set the following option in your jupyter notebook before plotting

  %config InlineBackend.print_figure_kwargs = {'bbox_inches':None}
  

• Use the %matplotlib notebook backend instead of the %matplotlib inline one.