Northeastern University DS 2000 — Programming with Data, Fall 2022, Profs. Muzny and Rachlin 
Fall 2022 
Homework 7 
Released: October 28, 2022 
Due: November 4th, 2022 @ 9pm ET 
Late Due Date (1-48 hours late): November 6th, 2022 @ 9pm ET 
Homework 7: ! 
Introduction 
The city of Boston and Blue Bikes provide remarkably complete data (https://www.bluebikes.com/system-data) 
about how Blue Bikes are used from month to month. We've downloaded the data for the month of September 
and filtered and cleaned it for you. (We made it a much smaller data set and kept only Sept 23 - 30th). 
Blue Bikes are a vital piece of public transit infrastructure and it's important that Boston's amateur data science 
team (that's you!) evaluates how they are being used in order to make> 
expand and upgrade the network. 
Expectations 
We expect you to write your code in a modular style using a well-defined collection of functions. The functions 
you implement are totally at your discretion. As always, you will be graded on a) correctness, b) code design 
and efficiency, and c) documentation and style. You may not use Pandas, numpy, or any machine learning 
library such as Scikit-Learn that would trivialize some parts of the implementation. 
What you're practicing in this assignment: 
- function/program design 
- dictionaries (yes, you must use dictionaries everywhere we tell you that you must) 
- lists, etc 
Requirements: 
1. Read the data. Load the trips.csv file, which contains all blue bikes trips in the last week of 
September, into a list of dictionaries where, for each dictionary, the column headers are the keys 
(strings) and the values are the values for the corresponding row. (You may use the code we wrote in 
class on 10/28 to create this list of dictionaries.) Read stations.csv file into a separate dictionary 
that maps station names to the GPS coordinates (latitude and longitude) for that station. Do not use the 
csv library. Write your own functions to read in both csv files. 
Warning! These are large files with many lines of data! To check if you've properly loaded them, test the 
length and take a look at the first few rows. You should have 445 stations and 141609 trips (yes, more 
than one hundred thousand). 
Here are the first two rows of the list of dictionaries from trips.csv: 
[{'duration': 602, 'start_day': 23, 'start_day_name': 'Friday', 'start_station': 
'MIT Vassar St', 'end_station': "Packard's Corner - Commonwealth Ave at Brighton 
Ave", 'bike_id': 8250}, {'duration': 466, 'start_day': 23, 'start_day_name': 
Northeastern University DS 2000 — Programming with Data, Fall 2022, Profs. Muzny and Rachlin 
'Friday', 'start_station': 'Ruggles T Stop - Columbus Ave at Melnea Cass Blvd', 
'end_station': 'St. Alphonsus St at Tremont St', 'bike_id': 8335}] 
Here are two examples of station entries you would have in your stations dictionary: 
{'1200 Beacon St': [42.34414899, -71.11467361], 
'160 Arsenal': [42.36466403, -71.17569387], 
… } 
2. Calculate the distance and speed of each trip. Using the provided haversine distance function to 
calculate the distance between two stations. With the distance and trip time you can determine the trip 
speed in miles per hour. Note that the given trip durations are in seconds. 
Warning! Some trips are associated with stations that are not in your stations.csv file. Make sure that 
these trips do not cause your program to stop (for instance, by throwing an error). We recommend filling 
in a special value (e.g., None or "N/A") for these trips. 
import math 
# Earth's radius varies because the earth is not a perfect sphere. 
# Use this number as it is considered the global average. 
EARTH_RADIUS = 3959 
def haversine_distance(start, end): 
""" 
Calculates the distance in miles between two points on the earth's surface 
described by latitude and longitude. 
Parameters: 
start: list 
list of two floats—latitude and longitude 
end: list 
list of two floats—latitude and longitude 
Return: 
float - distance in miles between the two points 
""" 
lat1 = start[0] 
long1 = start[1] 
lat2 = end[0] 
long2 = end[1] 
lat1 = math.radians(lat1) 
long1 = math.radians(long1) 
lat2 = math.radians(lat2) 
long2 = math.radians(long2) 
delta_lat = lat2 - lat1 
delta_long = long2 - long1 
# the earth's radius is a constant value 
a = math.sin(delta_lat / 2)**2 + math.cos(lat1) * math.cos(lat2) * math.sin(delta_long / 2)**2 
haversine = EARTH_RADIUS * 2 * math.asin(math.sqrt(a)) 
return haversine 
Hint: you may find it useful to add distance and speed information to each row in your data by inserting 
new keys into each dictionary with this information. 
Northeastern University DS 2000 — Programming with Data, Fall 2022, Profs. Muzny and Rachlin 
Hint 2: Here are examples of what your first two rows in your list of dictionaries might look like after this 
operation. (We've highlighted the added information for you). 
[{'duration': 602, 'start_day': 23, 'start_day_name': 'Friday', 'start_station': 
'MIT Vassar St', 'end_station': "Packard's Corner - Commonwealth Ave at Brighton 
Ave", 'bike_id': 8250, 'dist': 1.0413475448513339, 'mph': 6.227327510738874}, 
{'duration': 466, 'start_day': 23, 'start_day_name': 'Friday', 'start_station': 
'Ruggles T Stop - Columbus Ave at Melnea Cass Blvd', 'end_station': 'St. Alphonsus 
St at Tremont St', 'bike_id': 8335, 'dist': 0.7073886058314272, 'mph': 
5.464804680242785}] 
And here's an example of what a row that we were unable to calculate distance and mph for might look 
like: 
{'duration': 1231, 'start_day': 23, 'start_day_name': 'Friday', 'start_station': 
'Graham and Parks School – Linnaean St at Walker St', 'end_station': 'Harvard 
University Radcliffe Quadrangle at Shepard St / Garden St', 'bike_id': 6442, 
'dist': None, 'mph': None} 
3. Visualize the distribution of distances and speeds for the trips (distances.pdf and speeds.pdf). 
Create two well-labeled histograms using matplotlib's plt.hist() function. This function takes a 
single list of values, then calculates the frequencies of these values for you. Feel free to use additional 
optional parameters to make the graph visually appealing to your eye. 
plt.hist(values, bins = 100) 
matplotlib hist documentation 
Hint: think about how you might be able to use a get_column or extract_column function to help you 
here. (Both professors have written these functions in data_utils.py and dataproc.py.) 
Hint 2: you don't want to include data from trips that we couldn't calculate a valid distance or speed for, 
so make sure to do something to filter these out! (But you won't be filtering out trips with a calculated 
distance of 0, that's a different case!) 
4. Create a dictionary that maps day of the week to the count of trips associated with that day of 
the week. Using only trips that end at the station "Forsyth St at Huntington Ave". Print a report 
displaying these counts. This is the only required output of your program. 
For example, your output might look like: 
Number of trips ending at Forsyth St at Huntington Ave: 
Friday: ??? 
Saturday: ??? 
… 
Note that we do not expect these to be in "standard week day" order. 
5. Extra credit (up to +9). Create a visualization (singlebike.pdf) depicting the voyage of a single bike 
throughout the course of late September 2022. Your visualization should show the physical locations of 
Northeastern University DS 2000 — Programming with Data, Fall 2022, Profs. Muzny and Rachlin 
each start/end station that this bike came from/went to for each trip. Make sure to choose a bike that 
went on at least 50 trips in your trip data. Plot Northeastern and any other (0 to 3 other) geographic 
points in Boston/Cambridge/Somerville that will help your user understand where the stations are that 
the bike has traveled between. 
The data is in order, top-to-bottom, first trip on September 23rd to last trip on September 30th. 
All appearance choices are up to you! Make sure that your visualization is well-labeled and flex your 
creative muscles! 
Program Structure 
Reminders (we copy + pasted much of this from the style guide!) about what your program should look like: 
""" Your Name 
course 
assignment 
date 
description of the program (1 - 2 sentences) 
your copy+pasted program output here 
""" 
import matplotlib.pyplot as plt 
# other imports here 
# constants go here 
# NO other variables should be defined outside of functions 
def example_function1(example_parameter): 
""" 
Your function comment here. Must include all three of: 
what does it do, what parameters does it take, what value does it return 
""" 
# your function code here 
def example_function2(example_parameter1, example_parameter2): 
""" 
Your function comment here 
""" 
# your function code here 
def main(): 
# Your code goes here! 
# Your code goes here! 
if __name__ == "__main__": 
main() 
What to Submit 
Submit your program (bluebikes.py), your visualizations (distances.pdf, speeds.pdf, and, optionally, 
singlebike.pdf). In your program header, include the program text output produced when you run 
your program. 
Northeastern University DS 2000 — Programming with Data, Fall 2022, Profs. Muzny and Rachlin 
Please Read: It is your responsibility to verify that your program runs (by examining the autograder output), 
and that ALL required files have been submitted to gradescope. If you forget to submit your code, your 
visualization, or your program doesn’t run, your grade will be affected. If you resubmit your assignment for any 
reason, your resubmission must include ALL required files. Files from previous submissions are deleted from 
gradescope and not accessible to the graders! Resubmissions after the due date (but before the late due 
date) will receive the full late penalty. 
No homework files will be accepted for grading or regrading after the late due date!