In this notebook I do some initial data exploration with the metro delay data



In [58]:

    
import pandas as pd
import matplotlib.pyplot as plt
import matplotlib
from geopy.geocoders import Nominatim
from geopy.distance import vincenty
from __future__ import division
import pickle
matplotlib.style.use('ggplot')
%matplotlib inline

Metro delays are from http://www.opendatadc.org/dataset/wmata-disruption-reports, a open file of daily disruption reports from WMATA's api



In [52]:

    
metro_delays = pd.read_csv("We'll be Moving Momentarily - Incidents.csv")



In [53]:

    
metro_delays.head()









    Out[53]:






  
    
      
      Date
      Time
      Incident
      Line
      Direction
      Cause
      Delay
    
  
  
    
      0
      12/31/2012
      5:00 a.m.
      A Fort Totten-bound Yellow Line train at Brad...
      Yellow
      Fort Totten
      an operational problem
      13.0
    
    
      1
      12/31/2012
      10:30 a.m.
      A Grosvenor-bound Red Line train at Gallery P...
      Red
      Grosvenor
      an equipment problem
      6.0
    
    
      2
      12/31/2012
      4:11 p.m.
      A Greenbelt-bound Green Line train at Branch ...
      Green
      Greenbelt
      an operational problem
      4.0
    
    
      3
      12/31/2012
      4:50 p.m.
      A Greenbelt-bound Green Line train at Shaw-Ho...
      Green
      Greenbelt
      an equipment problem
      3.0
    
    
      4
      12/31/2012
      5:55 p.m.
      A Largo Town Center-bound Blue Line train at ...
      Blue
      Largo Town Center
      a brake problem
      9.0

How many delays are there of greater than 30 minutes?



In [54]:

    
long_delays = metro_delays[metro_delays['Delay'] >= 30]
len(long_delays)









    Out[54]:





461



In [55]:

    
long_delays.iloc[0]["Date"]









    Out[55]:





'12/10/2012'

I need to convert the 'Date' and 'Time' columns into a timestamp for each event



In [56]:

    
metro_delays['Time'] = metro_delays['Time'].str.replace('.','')
metro_delays['time_stamp'] = metro_delays['Date'] + ' ' + metro_delays['Time']



In [57]:

    
metro_delays['Time'] = pd.to_datetime(metro_delays['time_stamp'], format='%m/%d/%Y %I:%M %p')
metro_delays.head()









    Out[57]:






  
    
      
      Date
      Time
      Incident
      Line
      Direction
      Cause
      Delay
      time_stamp
    
  
  
    
      0
      12/31/2012
      2012-12-31 05:00:00
      A Fort Totten-bound Yellow Line train at Brad...
      Yellow
      Fort Totten
      an operational problem
      13.0
      12/31/2012 5:00 am
    
    
      1
      12/31/2012
      2012-12-31 10:30:00
      A Grosvenor-bound Red Line train at Gallery P...
      Red
      Grosvenor
      an equipment problem
      6.0
      12/31/2012 10:30 am
    
    
      2
      12/31/2012
      2012-12-31 16:11:00
      A Greenbelt-bound Green Line train at Branch ...
      Green
      Greenbelt
      an operational problem
      4.0
      12/31/2012 4:11 pm
    
    
      3
      12/31/2012
      2012-12-31 16:50:00
      A Greenbelt-bound Green Line train at Shaw-Ho...
      Green
      Greenbelt
      an equipment problem
      3.0
      12/31/2012 4:50 pm
    
    
      4
      12/31/2012
      2012-12-31 17:55:00
      A Largo Town Center-bound Blue Line train at ...
      Blue
      Largo Town Center
      a brake problem
      9.0
      12/31/2012 5:55 pm

export dataframe with pickle



In [59]:

    
pickle.dump( metro_delays, open( "metro_delays.p", "wb" ) )



In [ ]:

	Date	Time	Incident	Line	Direction	Cause	Delay
0	12/31/2012	5:00 a.m.	A Fort Totten-bound Yellow Line train at Brad...	Yellow	Fort Totten	an operational problem	13.0
1	12/31/2012	10:30 a.m.	A Grosvenor-bound Red Line train at Gallery P...	Red	Grosvenor	an equipment problem	6.0
2	12/31/2012	4:11 p.m.	A Greenbelt-bound Green Line train at Branch ...	Green	Greenbelt	an operational problem	4.0
3	12/31/2012	4:50 p.m.	A Greenbelt-bound Green Line train at Shaw-Ho...	Green	Greenbelt	an equipment problem	3.0
4	12/31/2012	5:55 p.m.	A Largo Town Center-bound Blue Line train at ...	Blue	Largo Town Center	a brake problem	9.0

	Date	Time	Incident	Line	Direction	Cause	Delay	time_stamp
0	12/31/2012	2012-12-31 05:00:00	A Fort Totten-bound Yellow Line train at Brad...	Yellow	Fort Totten	an operational problem	13.0	12/31/2012 5:00 am
1	12/31/2012	2012-12-31 10:30:00	A Grosvenor-bound Red Line train at Gallery P...	Red	Grosvenor	an equipment problem	6.0	12/31/2012 10:30 am
2	12/31/2012	2012-12-31 16:11:00	A Greenbelt-bound Green Line train at Branch ...	Green	Greenbelt	an operational problem	4.0	12/31/2012 4:11 pm
3	12/31/2012	2012-12-31 16:50:00	A Greenbelt-bound Green Line train at Shaw-Ho...	Green	Greenbelt	an equipment problem	3.0	12/31/2012 4:50 pm
4	12/31/2012	2012-12-31 17:55:00	A Largo Town Center-bound Blue Line train at ...	Blue	Largo Town Center	a brake problem	9.0	12/31/2012 5:55 pm