notebook.community

Edit and run

Homework 1: Data basics

Due: 1/12/2016

Problem:

  • Obtain the CSV (comma separated variable) file containing the counts of bicycles crossing the Fremont Bridge since 2012 (as described here).
  • Create a project directory with subdirectories for data and analysis, and create a README file.
  • Download the data from here. Put the it into the data directory.
  • Create an iPython Notebook to analyze these data.

In the notebook:

  1. Isolate the 2015 data, creating fields for date, hour, and count;
  2. use python matplotlib to plot the counts by hour;
  3. compute the hourly average counts;
  4. determine what is the busiest hour of the day.




In [1]:



    


%matplotlib inline

import matplotlib
import matplotlib.pyplot as plt
import seaborn
seaborn.set()
matplotlib.rcParams['figure.figsize'] = (15, 8)

import numpy as np
import pandas as pd



    











In [2]:



    


data = pd.read_csv("../data/4xy5-26gy.csv", parse_dates=['date'], index_col=['date'])



    











In [3]:



    


data.head()



    


















    
Out[3]:










  
    

      
      fremont_bridge_nb
      fremont_bridge_sb
    

    

      date
      
      
    

  
  
    

      2012-10-03 02:00:00
      1
      1
    

    

      2012-10-05 01:00:00
      4
      4
    

    

      2012-10-05 03:00:00
      2
      2
    

    

      2012-10-05 23:00:00
      16
      16
    

    

      2012-10-07 22:00:00
      9
      9

1. Isolate the 2015 data, creating fields for date, hour, and count;





In [4]:



    


data['datetime'] = data.index.copy()



    











In [5]:



    


data.head()



    


















    
Out[5]:










  
    

      
      fremont_bridge_nb
      fremont_bridge_sb
      datetime
    

    

      date
      
      
      
    

  
  
    

      2012-10-03 02:00:00
      1
      1
      2012-10-03 02:00:00
    

    

      2012-10-05 01:00:00
      4
      4
      2012-10-05 01:00:00
    

    

      2012-10-05 03:00:00
      2
      2
      2012-10-05 03:00:00
    

    

      2012-10-05 23:00:00
      16
      16
      2012-10-05 23:00:00
    

    

      2012-10-07 22:00:00
      9
      9
      2012-10-07 22:00:00
    

  




















In [6]:



    


data['year'] = data['datetime'].apply(lambda x: x.year)



    











In [7]:



    


data.head()



    


















    
Out[7]:










  
    

      
      fremont_bridge_nb
      fremont_bridge_sb
      datetime
      year
    

    

      date
      
      
      
      
    

  
  
    

      2012-10-03 02:00:00
      1
      1
      2012-10-03 02:00:00
      2012
    

    

      2012-10-05 01:00:00
      4
      4
      2012-10-05 01:00:00
      2012
    

    

      2012-10-05 03:00:00
      2
      2
      2012-10-05 03:00:00
      2012
    

    

      2012-10-05 23:00:00
      16
      16
      2012-10-05 23:00:00
      2012
    

    

      2012-10-07 22:00:00
      9
      9
      2012-10-07 22:00:00
      2012
    

  




















In [8]:



    


data = data[data['year'] == 2015]



    











In [9]:



    


data = data.drop('year',1)



    











In [10]:



    


data['date'] = data['datetime'].apply(lambda x: x.date())



    











In [11]:



    


data.head()



    


















    
Out[11]:










  
    

      
      fremont_bridge_nb
      fremont_bridge_sb
      datetime
      date
    

    

      date
      
      
      
      
    

  
  
    

      2015-07-30 05:00:00
      20
      22
      2015-07-30 05:00:00
      2015-07-30
    

    

      2015-07-11 20:00:00
      29
      36
      2015-07-11 20:00:00
      2015-07-11
    

    

      2015-07-31 19:00:00
      88
      103
      2015-07-31 19:00:00
      2015-07-31
    

    

      2015-07-14 08:00:00
      448
      190
      2015-07-14 08:00:00
      2015-07-14
    

    

      2015-07-17 20:00:00
      32
      66
      2015-07-17 20:00:00
      2015-07-17
    

  




















In [12]:



    


data['hour'] = data['datetime'].apply(lambda x: x.time().hour)



    











In [13]:



    


data.head()



    


















    
Out[13]:










  
    

      
      fremont_bridge_nb
      fremont_bridge_sb
      datetime
      date
      hour
    

    

      date
      
      
      
      
      
    

  
  
    

      2015-07-30 05:00:00
      20
      22
      2015-07-30 05:00:00
      2015-07-30
      5
    

    

      2015-07-11 20:00:00
      29
      36
      2015-07-11 20:00:00
      2015-07-11
      20
    

    

      2015-07-31 19:00:00
      88
      103
      2015-07-31 19:00:00
      2015-07-31
      19
    

    

      2015-07-14 08:00:00
      448
      190
      2015-07-14 08:00:00
      2015-07-14
      8
    

    

      2015-07-17 20:00:00
      32
      66
      2015-07-17 20:00:00
      2015-07-17
      20
    

  




















In [14]:



    


date_hour_data = data.groupby(['date','hour']).sum()



    











In [15]:



    


date_hour_data.head()



    


















    
Out[15]:










  
    

      
      
      fremont_bridge_nb
      fremont_bridge_sb
    

    

      date
      hour
      
      
    

  
  
    

      2015-07-01
      0
      7
      7
    

    

      1
      5
      4
    

    

      2
      2
      7
    

    

      3
      4
      7
    

    

      4
      9
      4

2. use python matplotlib to plot the counts by hour





In [16]:



    


hour_data = data.groupby('hour').sum()



    











In [17]:



    


ax = hour_data.plot(kind='bar',title='Rides/Hour', width=0.8)
ax.set_ylabel("Bicycle counts")



    


















    
Out[17]:








<matplotlib.text.Text at 0x10decda50>

3. compute the hourly average counts





In [18]:



    


hour_mean = data.groupby('hour').mean()



    











In [19]:



    


hour_mean



    


















    
Out[19]:










  
    

      
      fremont_bridge_nb
      fremont_bridge_sb
    

    

      hour
      
      
    

  
  
    

      0
      8.297297
      12.351351
    

    

      1
      4.685714
      5.828571
    

    

      2
      3.161290
      3.677419
    

    

      3
      2.531250
      2.468750
    

    

      4
      5.111111
      4.694444
    

    

      5
      20.744186
      14.790698
    

    

      6
      86.076923
      44.102564
    

    

      7
      224.282051
      98.461538
    

    

      8
      279.205128
      146.589744
    

    

      9
      134.432432
      92.513514
    

    

      10
      70.263158
      54.973684
    

    

      11
      58.512821
      51.666667
    

    

      12
      60.763158
      56.236842
    

    

      13
      64.750000
      58.950000
    

    

      14
      62.388889
      65.361111
    

    

      15
      77.404762
      85.357143
    

    

      16
      114.057143
      174.771429
    

    

      17
      154.384615
      355.717949
    

    

      18
      122.500000
      212.323529
    

    

      19
      78.828571
      108.114286
    

    

      20
      56.026316
      68.973684
    

    

      21
      39.540541
      41.432432
    

    

      22
      27.606061
      34.454545
    

    

      23
      15.975610
      21.000000

4. Determine what is the busiest hour of the day





In [20]:



    


hour_mean.sort('fremont_bridge_nb', ascending=False).head(5)



    


















    
Out[20]:










  
    

      
      fremont_bridge_nb
      fremont_bridge_sb
    

    

      hour
      
      
    

  
  
    

      8
      279.205128
      146.589744
    

    

      7
      224.282051
      98.461538
    

    

      17
      154.384615
      355.717949
    

    

      9
      134.432432
      92.513514
    

    

      18
      122.500000
      212.323529
    

  




















In [21]:



    


hour_mean.sort('fremont_bridge_sb', ascending=False).head(5)



    


















    
Out[21]:










  
    

      
      fremont_bridge_nb
      fremont_bridge_sb
    

    

      hour
      
      
    

  
  
    

      17
      154.384615
      355.717949
    

    

      18
      122.500000
      212.323529
    

    

      16
      114.057143
      174.771429
    

    

      8
      279.205128
      146.589744
    

    

      19
      78.828571
      108.114286

Content source: gshguru/uwseds

Similar notebooks:

notebook.community | gallery | about