[常見問題] Matplotlib - How to avoid overlapping of labels & autopct in a matplotlib pie chart?

 Source From Here

Question
My Python code is:

view plaincopy to clipboardprint?

1. import matplotlib.pyplot as plt  
2.   
3. values = [234, 64, 54,10, 0, 1, 0, 9, 2, 1, 7, 7]  
4. labels = ['Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul','Aug','Sep','Oct', 'Nov','Dec']  
5.   
6. colors = ['yellowgreen', 'red', 'gold', 'lightskyblue',   
7.           'white','lightcoral','blue','pink', 'darkgreen',   
8.           'yellow','grey','violet','magenta','cyan']  
9.   
10. plt.pie(values, labels=labels, autopct='%1.1f%%', shadow=True,   
11.         colors=colors, startangle=90, radius=1.2)  
12.   
13. plt.show()  

Is it possible to show the labels "Jan", "Feb", "Mar", etc. and the percentages, either:

* Without overlapping, or
* Using an arrow mark?

HowTo
Alternatively you can put the legends beside the pie graph:

view plaincopy to clipboardprint?

1. import matplotlib.pyplot as plt  
2. import numpy as np  
3.   
4. x = np.char.array(['Jan','Feb','Mar','Apr','May','Jun','Jul','Aug','Sep','Oct', 'Nov','Dec'])  
5. y = np.array([234, 64, 54,10, 0, 1, 0, 9, 2, 1, 7, 7])  
6. colors = ['yellowgreen','red','gold','lightskyblue','white','lightcoral','blue','pink', 'darkgreen','yellow','grey','violet','magenta','cyan']  
7. porcent = 100.*y/y.sum()  
8.   
9. patches, texts = plt.pie(y, colors=colors, startangle=90, radius=1.2)  
10. labels = ['{0} - {1:1.2f} %'.format(i,j) for i,j in zip(x, porcent)]  
11.   
12. sort_legend = True  
13. if sort_legend:  
14.     patches, labels, dummy =  zip(*sorted(zip(patches, labels, y),  
15.                                           key=lambda x: x[2],  
16.                                           reverse=True))  
17.   
18. plt.legend(patches, labels, loc='best', bbox_to_anchor=(-0.1, 1.), fontsize=8)  
19. # plt.savefig('piechart.png', bbox_inches='tight')