In [11]:

    

import os
os.listdir()
import warnings
warnings.filterwarnings('ignore')
import pandas as pd
import numpy as np
import matplotlib as plt
import seaborn as sns
import matplotlib.pyplot as pyp

pd.set_option('max_columns', 50)
#Inline imaging
%matplotlib inline


    

In [12]:

    

df = pd.read_csv('closed_issue_comments.csv', delimiter='|')
#df.set_index('issue_number', inplace=True)
df.head(3)


    

    
Out[12]:






  

    

      

      
issue_number
      
number_of_comments
      
repo_name
    
  
  

    

      
0
      
3390
      
1
      
solutions-engineering
    
    

      
1
      
3389
      
2
      
solutions-engineering
    
    

      
2
      
3364
      
1
      
solutions-engineering
    
  

In [48]:

    

comm_nums = df[(df['number_of_comments'] < 15)]


    

In [52]:

    

data = [comm_nums.index, comm_nums['number_of_comments']]
jp = sns.jointplot(x=comm_nums.index, y=(comm_nums['number_of_comments']), kind="reg", ylim=(0, 15), size=20)
jp.savefig('issue-comment-regression')
#jp = sns.jointplot(x="Issue Number", y="number_of_comments", data=data, kind="reg", ylim=(0, 10), size=20)


    

In [32]:

    

#import matplotlib.pyplot as pyp
plt.pyplot.style.use('ggplot')
new_df = df[['number_of_comments']]
cs=plt.cm.Set1(np.arange(40)/40)
#print(pyp.style.available)
pyp.figure(figsize=(13, 13))
a = new_df['number_of_comments'].value_counts()
b = pyp.pie(a, labels=a.index, colors=cs, autopct='%1.1f%%', shadow=True)
pyp.legend(a.index, bbox_to_anchor=(1.1, 1.05))
pyp.axis("equal")
pyp.tight_layout()
pyp.savefig('issue-comments-pie')


    

In [ ]:

    




    

In [ ]: