更改Matplotlib散点图的默认边缘颜色

Question

I'm looking for a way to change the default edge color for matplotlib scatter plots. Typically, things like that would be set through rcParams , and my understanding is that I should set patch.edgecolor . However, that doesn't seem to work. Here is an example:

import numpy as np
from matplotlib import pyplot as plt

x = np.random.randn(50)
y = np.random.randn(50)

with plt.rc_context({'patch.edgecolor': 'white'}):
    plt.subplot(121)
    plt.scatter(x, y)

plt.subplot(122)
plt.scatter(x, y, edgecolors='white')

In the result, I would like to have both subplots look the same, but instead they look like this: Note that I don't want to change the code within the with statement, but instead configure matplotlib such that it uses white edges as a fallback if I don't specify anything else. Confusingly, the documentation uses a default argument edgecolors=None in the function signature, but states that the default value for edgecolors is 'face' . How can I change this behaviour?

Answer 1

The rc parameter you're looking for is called

'scatter.edgecolors'

Eg plt.rcParams['scatter.edgecolors'] = "white" or
with plt.rc_context({'scatter.edgecolors': 'white'}): .

This is a new feature introduced in #12992 and available from matplotlib 3.1 onwards, which will be released very soon. As of today, you can install the release candidate via pip install --pre --upgrade matplotlib to get this feature.