In [7]:

    

import pandas as pd
%matplotlib inline
import matplotlib.pyplot as plt
import seaborn as sns
import tables


    

In [2]:

    

df = pd.read_csv('../parkingdata/clean_parking_violations.tsv', delimiter='\t', parse_dates=True)
df = df.sort_values(by="ticket_issue_datetime")
df['ticket_issue_datetime'] = pd.to_datetime(df.ticket_issue_datetime)
df['total'] = 1 # makes counting cleaner


    

In [3]:

    

df['weekday'] = df['ticket_issue_datetime'].apply(lambda x: x.weekday())
df['year'] = df['ticket_issue_datetime'].apply(lambda x: x.year)
df['month'] = df['ticket_issue_datetime'].apply(lambda x: x.month)


    

In [4]:

    

df['year-month'] = df.year.astype(str) + '-' + df.month.astype(str)


    

In [8]:

    

groupby_year_month = df.groupby(['year','month']).total.count().reset_index()
average_per_month = groupby_year_month.groupby('month').total.mean().reset_index()
average_per_month


    

    
Out[8]:






  

    

      

      
month
      
total
    
  
  

    

      
0
      
1
      
141732
    
    

      
1
      
2
      
115615
    
    

      
2
      
3
      
176299
    
    

      
3
      
4
      
193942
    
    

      
4
      
5
      
172649
    
    

      
5
      
6
      
175203
    
    

      
6
      
7
      
173120
    
    

      
7
      
8
      
177289
    
    

      
8
      
9
      
169749
    
    

      
9
      
10
      
151573
    
    

      
10
      
11
      
131478
    
    

      
11
      
12
      
127044
    
  

In [132]:

    

sns.lmplot('month','total', data=average_per_month, fit_reg=False)


    

    
Out[132]:





<seaborn.axisgrid.FacetGrid at 0x13df6e630>






    

In [134]:

    

sns.kdeplot(average_per_month.total)


    

    
Out[134]:





<matplotlib.axes._subplots.AxesSubplot at 0x1f20f2d68>






    

In [135]:

    

sns.clustermap(average_per_month)


    

    
Out[135]:





<seaborn.matrix.ClusterGrid at 0x11de242e8>






    

In [140]:

    

sns.heatmap([average_per_month.total, average_per_month.month], annot=True, fmt="d")


    

    
Out[140]:





<matplotlib.axes._subplots.AxesSubplot at 0x1c2ed2ef0>






    

In [137]:

    

sns.distplot(average_per_month.total)


    

    
Out[137]:





<matplotlib.axes._subplots.AxesSubplot at 0x1d8deae10>






    

In [139]:

    

plt.hist(average_per_month.total, alpha=.3)
sns.rugplot(average_per_month.total);


    

In [143]:

    

sns.boxplot([average_per_month.total, average_per_month.month])


    

    
Out[143]:





<matplotlib.axes._subplots.AxesSubplot at 0x192ed2278>






    

In [145]:

    

sns.violinplot([average_per_month.month, average_per_month.total])


    

    
Out[145]:





<matplotlib.axes._subplots.AxesSubplot at 0x13d0e0cc0>






    

In [15]:

    

sns.distplot(average_per_month.total, bins=5, kde=True, rug=True);


    

In [16]:

    

sns.regplot(x="total", y="month", data=average_per_month);


    

In [17]:

    

sns.lmplot(x="total", y="month", data=average_per_month);


    

In [ ]: