更改 matplotlib 饼图上的 autopct 标签位置

Question

我正在 matplotlib 中绘制一些饼图，我的一些图表上的百分比标签相互重叠，看起来很乱。有没有办法改变文本的位置，所以它都是可读的？ 下面是我得到的示例-“其他”和“ALD2_OH”类别重叠且不可读。

我的绘图代码在这里：

matplotlib.rcParams.update({'font.size': 18})
plt.figure(figsize=(11,11))

labels = ['ALD2 + OH','PAN + $therm$','ALD2 + NO$_3$','ATOOH + $hv$',
          'Others',]

colours = ['BlueViolet','DarkMagenta','DarkOrchid','DarkViolet','Purple'
           ]

patches, texts,autotexts = plt.pie(main_producers, labels=labels, colors = colours,
        autopct='%1.1f%%', startangle = 90)

plt.title('Contribution to MCO$_3$ yeild')

希望有人能帮忙！

谢谢

Answer 1

您可能希望从楔形中心沿半径移动autotexts 窄楔形：

for patch, txt in zip(patches, autotexts):
    # the angle at which the text is located
    ang = (patch.theta2 + patch.theta1) / 2.
    # new coordinates of the text, 0.7 is the distance from the center 
    x = patch.r * 0.7 * np.cos(ang*np.pi/180)
    y = patch.r * 0.7 * np.sin(ang*np.pi/180)
    # if patch is narrow enough, move text to new coordinates
    if (patch.theta2 - patch.theta1) < 10.:
        txt.set_position((x, y))

这会产生（我在某种程度上模拟了您的数据）：