notebook.community

Edit and run

Topic: Challenge Set 1 Subject: Explore MTA turnstile data Date: 01/23/2017 Name: Prashant Tatineni

Challenge 1





In [1]:



    


import pandas as pd



    











In [2]:



    


wk1 = pd.read_csv('http://web.mta.info/developers/data/nyct/turnstile/turnstile_161224.txt')
wk2 = pd.read_csv('http://web.mta.info/developers/data/nyct/turnstile/turnstile_161231.txt')
wk3 = pd.read_csv('http://web.mta.info/developers/data/nyct/turnstile/turnstile_170107.txt')
wk4 = pd.read_csv('http://web.mta.info/developers/data/nyct/turnstile/turnstile_170114.txt')



    











In [3]:



    


wk1.columns = ['C/A','UNIT','SCP','STATION','LINENAME','DIVISION','DATE','TIME','DESC','ENTRIES','EXITS']
wk2.columns = ['C/A','UNIT','SCP','STATION','LINENAME','DIVISION','DATE','TIME','DESC','ENTRIES','EXITS']
wk3.columns = ['C/A','UNIT','SCP','STATION','LINENAME','DIVISION','DATE','TIME','DESC','ENTRIES','EXITS']
wk4.columns = ['C/A','UNIT','SCP','STATION','LINENAME','DIVISION','DATE','TIME','DESC','ENTRIES','EXITS']



    











In [9]:



    


MTA_dict = {}
for index, row in wk1.iterrows():
    MTA_dict.setdefault((row['C/A'],row['UNIT'],row['SCP'],row['STATION']),[]).append([row['LINENAME'],row['DIVISION'],row['DATE'],row['TIME'],row['DESC'],row['ENTRIES'],row['EXITS']])



    











In [14]:



    


MTA_dict['N134','R385','00-00-00','ROCKAWAY BLVD'][0:3]



    


















    
Out[14]:








[['A', 'IND', '12/17/2016', '00:00:00', 'REGULAR', 7586070, 2968637],
 ['A', 'IND', '12/17/2016', '04:00:00', 'REGULAR', 7586115, 2968663],
 ['A', 'IND', '12/17/2016', '08:00:00', 'REGULAR', 7586266, 2968726]]

Challenge 2





In [15]:



    


import dateutil



    











In [16]:



    


MTA_time = {}
for index, row in wk1.iterrows():
    MTA_time.setdefault((row['C/A'],row['UNIT'],row['SCP'],row['STATION']),[]).append([dateutil.parser.parse(row['DATE'] + ' ' + row['TIME']),row['ENTRIES']])



    











In [29]:



    


MTA_time['N134','R385','00-00-00','ROCKAWAY BLVD'][0:3]



    


















    
Out[29]:








[[datetime.datetime(2016, 12, 17, 0, 0), 7586070],
 [datetime.datetime(2016, 12, 17, 4, 0), 7586115],
 [datetime.datetime(2016, 12, 17, 8, 0), 7586266]]

Challenge 3





In [26]:



    


import datetime

MTA_date = {}
for key in MTA_time:
    newList = []
    for i in range(0,len(MTA_time[key]),6):
        if i+5 < len(MTA_time[key]):
            newList.append([MTA_time[key][i][0].date(), MTA_time[key][i+5][1] - MTA_time[key][i][1]])
    MTA_date[key] = newList



    











In [27]:



    


MTA_date['N134','R385','00-00-00','ROCKAWAY BLVD'][0:3]



    


















    
Out[27]:








[[datetime.date(2016, 12, 17), 781],
 [datetime.date(2016, 12, 18), 605],
 [datetime.date(2016, 12, 19), 1339]]

Challenge 4





In [30]:



    


%matplotlib inline
import matplotlib.pyplot as plt



    











In [31]:



    


dates = [i[0] for i in MTA_date['N134','R385','00-00-00','ROCKAWAY BLVD']]
counts = [i[1] for i in MTA_date['N134','R385','00-00-00','ROCKAWAY BLVD']]



    











In [32]:



    


plt.figure(figsize=(10,3))
plt.plot(dates,counts)



    


















    
Out[32]:








[<matplotlib.lines.Line2D at 0x11f3ce190>]

Challenge 5





In [162]:



    


MTA_date



    


















    
Out[162]:








{('A002', 'R051', '02-00-00', '59 ST'): [[datetime.date(2016, 12, 31), 1092],
  [datetime.date(2017, 1, 1), 797]]}

















In [33]:



    


MTA_date_station = {}
for key in MTA_date:
    MTA_date_station.setdefault((key[0],key[1],key[3]),[]).append([key[2],MTA_date[key]])



    











In [35]:



    


MTA_date_station['N134','R385','ROCKAWAY BLVD'][0]



    


















    
Out[35]:








['00-00-00',
 [[datetime.date(2016, 12, 17), 781],
  [datetime.date(2016, 12, 18), 605],
  [datetime.date(2016, 12, 19), 1339],
  [datetime.date(2016, 12, 20), 1397],
  [datetime.date(2016, 12, 21), 1381],
  [datetime.date(2016, 12, 22), 1366],
  [datetime.date(2016, 12, 23), 1250]]]

Challenge 6





In [36]:



    


MTA_date_station_ts = {}
for key in MTA_date_station:
    ts = {}
    for turnstile in MTA_date_station[key]:
        for i in turnstile[1]:
            ts.setdefault(i[0],0)
            ts[i[0]] += i[1]
    MTA_date_station_ts[key] = ts



    











In [40]:



    


MTA_date_station_ts['N134','R385','ROCKAWAY BLVD']



    


















    
Out[40]:








{datetime.date(2016, 12, 17): 2467,
 datetime.date(2016, 12, 18): 1634,
 datetime.date(2016, 12, 19): 4024,
 datetime.date(2016, 12, 20): 4068,
 datetime.date(2016, 12, 21): 4131,
 datetime.date(2016, 12, 22): 3949,
 datetime.date(2016, 12, 23): 3666}

Challenge 7





In [49]:



    


dates = sorted(MTA_date_station_ts['N134','R385','ROCKAWAY BLVD'].keys())



    











In [51]:



    


counts = []
for date in dates:
    counts += [MTA_date_station_ts['N134','R385','ROCKAWAY BLVD'][date]]



    











In [52]:



    


plt.figure(figsize=(10,3))
plt.plot(dates,counts)



    


















    
Out[52]:








[<matplotlib.lines.Line2D at 0x11fceea50>]

Challenge 8





In [58]:



    


# Week 1
MTA_wk1 = MTA_date_station_ts
week1 = []
for key in MTA_wk1['N134','R385','ROCKAWAY BLVD']:
    week1 += [MTA_wk1['N134','R385','ROCKAWAY BLVD'][key]]
    
# Week 2
MTA_time = {}
for index, row in wk2.iterrows():
    MTA_time.setdefault((row['C/A'],row['UNIT'],row['SCP'],row['STATION']),[]).append([dateutil.parser.parse(row['DATE'] + ' ' + row['TIME']),row['ENTRIES']])
MTA_date = {}
for key in MTA_time:
    newList = []
    for i in range(0,len(MTA_time[key]),6):
        if i+5 < len(MTA_time[key]):
            newList.append([MTA_time[key][i][0].date(), MTA_time[key][i+5][1] - MTA_time[key][i][1]])
    MTA_date[key] = newList
MTA_date_station = {}
for key in MTA_date:
    MTA_date_station.setdefault((key[0],key[1],key[3]),[]).append([key[2],MTA_date[key]])
MTA_wk2 = {}
for key in MTA_date_station:
    ts = {}
    for turnstile in MTA_date_station[key]:
        for i in turnstile[1]:
            ts.setdefault(i[0],0)
            ts[i[0]] += i[1]
    MTA_wk2[key] = ts
week2 = []
for key in MTA_wk2['N134','R385','ROCKAWAY BLVD']:
    week2 += [MTA_wk2['N134','R385','ROCKAWAY BLVD'][key]]

# Week 3
MTA_time = {}
for index, row in wk3.iterrows():
    MTA_time.setdefault((row['C/A'],row['UNIT'],row['SCP'],row['STATION']),[]).append([dateutil.parser.parse(row['DATE'] + ' ' + row['TIME']),row['ENTRIES']])
MTA_date = {}
for key in MTA_time:
    newList = []
    for i in range(0,len(MTA_time[key]),6):
        if i+5 < len(MTA_time[key]):
            newList.append([MTA_time[key][i][0].date(), MTA_time[key][i+5][1] - MTA_time[key][i][1]])
    MTA_date[key] = newList
MTA_date_station = {}
for key in MTA_date:
    MTA_date_station.setdefault((key[0],key[1],key[3]),[]).append([key[2],MTA_date[key]])
MTA_wk3 = {}
for key in MTA_date_station:
    ts = {}
    for turnstile in MTA_date_station[key]:
        for i in turnstile[1]:
            ts.setdefault(i[0],0)
            ts[i[0]] += i[1]
    MTA_wk3[key] = ts
week3 = []
for key in MTA_wk3['N134','R385','ROCKAWAY BLVD']:
    week3 += [MTA_wk3['N134','R385','ROCKAWAY BLVD'][key]]

# Week 4
MTA_time = {}
for index, row in wk4.iterrows():
    MTA_time.setdefault((row['C/A'],row['UNIT'],row['SCP'],row['STATION']),[]).append([dateutil.parser.parse(row['DATE'] + ' ' + row['TIME']),row['ENTRIES']])
MTA_date = {}
for key in MTA_time:
    newList = []
    for i in range(0,len(MTA_time[key]),6):
        if i+5 < len(MTA_time[key]):
            newList.append([MTA_time[key][i][0].date(), MTA_time[key][i+5][1] - MTA_time[key][i][1]])
    MTA_date[key] = newList
MTA_date_station = {}
for key in MTA_date:
    MTA_date_station.setdefault((key[0],key[1],key[3]),[]).append([key[2],MTA_date[key]])
MTA_wk4 = {}
for key in MTA_date_station:
    ts = {}
    for turnstile in MTA_date_station[key]:
        for i in turnstile[1]:
            ts.setdefault(i[0],0)
            ts[i[0]] += i[1]
    MTA_wk4[key] = ts
week4 = []
for key in MTA_wk4['N134','R385','ROCKAWAY BLVD']:
    week4 += [MTA_wk4['N134','R385','ROCKAWAY BLVD'][key]]



    











In [60]:



    


plt.plot(week1)
plt.plot(week2)
plt.plot(week3)
plt.plot(week4)



    


















    
Out[60]:








[<matplotlib.lines.Line2D at 0x123e45e90>]

Challenge 9





In [64]:



    


total_ridership = {}
for station in MTA_wk1:
    total_ridership[station] = sum(MTA_wk1[station].values()) + sum(MTA_wk2[station].values()) + sum(MTA_wk3[station].values()) + sum(MTA_wk4[station].values())

Challenge 10





In [81]:



    


stations = total_ridership.keys()



    











In [82]:



    


counts = total_ridership.values()



    











In [118]:



    


total_ridership_counts = sorted(total_ridership.values())[19:-9]  # remove errors: negative and large positive values



    











In [119]:



    


plt.hist(total_ridership_counts)



    


















    
Out[119]:








(array([ 340.,  223.,   80.,   31.,   10.,    5.,    2.,    2.,    2.,    1.]),
 array([       0. ,   113678.8,   227357.6,   341036.4,   454715.2,
          568394. ,   682072.8,   795751.6,   909430.4,  1023109.2,
         1136788. ]),
 <a list of 10 Patch objects>)

Content source: ptpro3/ptpro3.github.io

Similar notebooks:

notebook.community | gallery | about