In [4]:

    

import pandas as pd
from pylab import *
import matplotlib.pyplot as plt


    

In [7]:

    

data = pd.read_csv("../../data/cleaned/UCB_dept_merge.csv")


    

In [ ]:

    




    

In [9]:

    

#Changing data type to date
data.creation_date = pd.to_datetime(data.creation_date)
data.po_closed_date = pd.to_datetime(data.po_closed_date)


    

In [10]:

    

#Creating Time to completion column
data["days_to_close"] = data.po_closed_date - data.creation_date
data['days_to_close'] = data['days_to_close'].astype('timedelta64[D]')


    

In [13]:

    

#Shows that data.po_id is not unique
len(data.po_id) == len(data.po_id.unique())


    

    
Out[13]:





False

In [14]:

    

#total number of po_id
len(data.po_id)


    

    
Out[14]:





611110

In [20]:

    

#i need to pull out those data duplicate
data[data.po_id.duplicated()]['po_id'].unique().shape[0]


    

    
Out[20]:





103668

In [26]:

    

po_counts = data.groupby('po_id').count().iloc[:, 0].sort(inplace=False)
po_counts.index = po_counts.index.astype('str')


    

In [30]:

    

%matplotlib inline


    

In [38]:

    

po_counts[po_counts > 50].plot(figsize=(15, 5))


    

    
Out[38]:





<matplotlib.axes._subplots.AxesSubplot at 0x11470e690>






    

In [2]:

    

100000/600000


    

    
Out[2]:





0

In [ ]:

    

h


    

In [ ]:

    




    

In [ ]:

    




    

In [ ]:

    




    

In [ ]: