Matplotlib中pie()无法调整饼图尺寸的问题求助

你的原始代码：

import pandas as pd 
import numpy as np 
df = pd.read_csv("Medals.csv")
import matplotlib.pyplot as plt
fig=plt.figure()
plt.suptitle("Medals data")
x = sum(df.Gold)
y = sum(df.Silver)
z = sum(df.Bronze)
y = np.array([x,y,z])
mylabels = ["Gold", "Silver", "Bronze"]
plt.pie(y,labels = mylabels,startangle=0,autopct='%1.1f%%')
plt.axis('equal')
plt.legend(["Gold","Silver","Bronze"],loc ="lower right")
fig.savefig("plot7.png")

下面提供两种可行的解决方法：

方法1：用radius参数缩小饼图+调整图例位置 #

你之前用radius没生效，大概率是图例和饼图重叠导致视觉上没变化，修改后代码如下：

import pandas as pd 
import numpy as np 
df = pd.read_csv("Medals.csv")
import matplotlib.pyplot as plt
fig=plt.figure()
plt.suptitle("Medals data")
x = sum(df.Gold)
y = sum(df.Silver)
z = sum(df.Bronze)
y = np.array([x,y,z])
mylabels = ["Gold", "Silver", "Bronze"]
# 设置radius参数缩小饼图，数值越小饼图越小
plt.pie(y,labels = mylabels,startangle=0,autopct='%1.1f%%', radius=0.7)
plt.axis('equal')
# 将图例移到饼图外侧，避免重叠
plt.legend(mylabels, loc="upper right", bbox_to_anchor=(1.3, 0.9))
# 保存时加上bbox_inches='tight'，防止图例被截断
fig.savefig("plot7.png", bbox_inches='tight')

方法2：自定义Axes区域控制饼图大小 #

通过创建指定尺寸的Axes，精准控制饼图在Figure中的占比，完全不改变Figure本身的尺寸：

import pandas as pd 
import numpy as np 
df = pd.read_csv("Medals.csv")
import matplotlib.pyplot as plt
fig=plt.figure()
plt.suptitle("Medals data")
# 创建自定义Axes，参数为[left, bottom, width, height]，取值范围0-1
# 例如[0.2,0.2,0.6,0.6]表示Axes从Figure的20%位置开始，宽度和高度各占60%
ax = fig.add_axes([0.2, 0.2, 0.6, 0.6])
x = sum(df.Gold)
y = sum(df.Silver)
z = sum(df.Bronze)
y = np.array([x,y,z])
mylabels = ["Gold", "Silver", "Bronze"]
ax.pie(y,labels = mylabels,startangle=0,autopct='%1.1f%%')
ax.axis('equal')
ax.legend(mylabels, loc="lower right")
fig.savefig("plot7.png")

内容的提问来源于stack exchange，提问作者Vineet Pashine