Changing only the line properties inside the circle when using pie in matplotlib - matplotlib

Question

When I am segmenting a circle with pie from matplotlib I would like to change the properties of the lines only inside the circle:
plt.rcParams['patch.edgecolor'] = 'lightgrey'
plt.rcParams['patch.linewidth'] = 1
Affect all the lines including the line of the circle itsef.

Answer 1

Step 1 - changing 'inner' lines
As usual it is a good idea to look at the matplotlib API documentation, where we find that pie plot provides a lot of arguments, one of which is wedgeprops
wedgeprops: [ None | dict of key value pairs ]
Dict of arguments passed to the wedge objects making the pie. For example, you can pass in wedgeprops = { ‘linewidth’ : 3 } to set the width of the wedge border lines equal to 3. For more details, look at the doc/arguments of the wedge object.
One of the arguments to Wedge is edgecolor, another is linewidth.
So in total you have to call
plt.pie([215, 130], colors=['b', 'r'],
wedgeprops = { 'linewidth' : 1 , 'edgecolor' : 'lightgrey'} )
However, since this also changes the outline of the pie diagram we need...
Step 2 - setting circonference circle
Now, in order to get a circle around the pie, or restore the initial linestyle for the circonference of the pie, we can set a new Circle patch with the desired properties on top of the pie.
The complete solution then looks something like this
import matplotlib.pyplot as plt
import matplotlib.patches
fig, ax = plt.subplots(figsize=(3,3))
ax.axis('equal')
slices, labels = ax.pie([186, 130, 85], colors=['b', 'r','y'],
wedgeprops = { 'linewidth' : 1 , 'edgecolor' : 'lightgrey'} )
# get the center and radius of the pie wedges
center = slices[0].center
r = slices[0].r
# create a new circle with the desired properties
circle = matplotlib.patches.Circle(center, r, fill=False, edgecolor="k", linewidth=2)
# add the circle to the axes
ax.add_patch(circle)
plt.show()

Answer 2

For a solution that works also with any pie chart, including exploded pie charts, e.g.
import numpy as np
import matplotlib as plt
data = [1, 2, 3, 1, 4, 2]
explode = [0.05] * len(data)
labels = list('ABCDEFGHIJKLMNOPQRSTUVWXYZ'[:len(data)])
fig, ax = plt.subplots()
pie = ax.pie(data, labels=labels, explode=explode)
use one of the following options:
Option A, add lines for each wedge of the pie
pie = ax.pie(data, labels=labels, explode=explode)
for wedge in pie[0]:
ax.plot([wedge.center[0], wedge.r*np.cos(wedge.theta1*np.pi/180)+wedge.center[0]], [wedge.center[1], wedge.r*np.sin(wedge.theta1*np.pi/180)+wedge.center[1]], color='k')
ax.plot([wedge.center[0], wedge.r*np.cos(wedge.theta2*np.pi/180)+wedge.center[0]], [wedge.center[1], wedge.r*np.sin(wedge.theta2*np.pi/180)+wedge.center[1]], color='k')
fig.show()
Option B, add edges to the pie wedges then overwrite the radial edge with another color (e.g. white)
from matplotlib import patches
pie = ax.pie(data, labels=labels, explode=explode, wedgeprops=dict(ec='k')
for wedge in pie[0]:
arc = patches.Arc(wedge.center, 2*wedge.r, 2*wedge.r, 0, theta1=wedge.theta1, theta2=wedge.theta2, ec='w', lw=1.5)
ax.add_patch(arc)
fig.show()