Drawing multiple donut charts

Question

I am writing a code to output 3 donut charts. I draw 3 pie charts at first, and then I want to fill each one with a white circle.

I am successfully outputting 3 pie charts, but the white circle appears only on the last one.

How do I make it appear on all three?

fig, axes = plt.subplots(1,3, figsize=(10,5))
for ax, col in zip(axes, df.columns):
    ax.pie(df[col], labels=None,  colors=colors)
    ax.set(ylabel='', title=col, aspect='equal')
    centre_circle = plt.Circle((0,0),0.70,fc='white')
    fig = plt.gcf()
    fig.gca().add_artist(centre_circle)

axes[0].legend(labels, bbox_to_anchor=(0, 0.5))



fig.savefig('your_file.png')
plt.show()

Answer 1

Because fig.gca() holds the last created axis in the figure, so the white circle is placed there three times independently of the loop variables.
You can simply add the circle artist to the axis in each iteration by using your ax variable

ax.add_artist(centre_circle)

However, your approch seems to be an unnecessary workaround for me, because you can accomplish the donut shape by using the wedgeprops kwarg with its width parameter (see here for further ones):

try e.g.

ax.pie(df[col], labels=None, colors=colors, wedgeprops=dict(width=.3))

instead of painting the white circle manually after each pie creation.

---

full example:

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

np.random.seed(42)
df = pd.DataFrame(np.random.randint(5, 10, (7, 3)), columns=['a', 'b', 'c'])

fig, axes = plt.subplots(1,3, figsize=(10,5))
for ax, col in zip(axes, df.columns):
    ax.pie(df[col], wedgeprops=dict(width=.3))
    ax.set(ylabel='', title=col, aspect='equal')

---

Note that you even do not have to loop over your columns but could create a pie chart directly from the pandas dataframe:

df.plot(kind='pie', subplots=True, wedgeprops=dict(width=.3), legend=False, labels=None)

result: