Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/ireneflorez/bikeshare

Exploring US Bike-Share data of 3 major cities - Using Jupyter Notebooks and Python scripts
https://github.com/ireneflorez/bikeshare

bike-share chicago jupyter-notebook python-script

Last synced: 6 days ago
JSON representation

Exploring US Bike-Share data of 3 major cities - Using Jupyter Notebooks and Python scripts

Awesome Lists containing this project

README 

        
# BikeShare



# Python Script to Explore US Bikeshare Data

This Python 3 script is written to explore data related to bike share systems for Chicago, New York City, and Washington. It imports data from csv files and computes descriptive statistics from the data. It is run from the terminal (or from Jupyter notebooks).



## How to run the script

This script is written in Python 3 and uses pandas, datetime, time, and csv packages. You will need all of these packages and Python 3 to be able to launch the script. Launch the bikeshare.py file from your terminal and you will see a prompt that will guide your experience reviewing the bikeshare data and reading relevant data analysis. 

You can also open the Data_Review_Chicago.ipynb file to view much of the data analysis that is provided in the .py file, all relative to Chicago data.



1) download the bikehsare folder onto your computer

2) open terminal

3) navigate to the location of your bikeshare folder

4) type the following into terminal 

/ python bikeshare.py /



## Datasets

The datasets used for this script contain bike share data for the first six months of 2017.  You can access the original data files here [Chicago](https://www.divvybikes.com/system-data), [New York City](https://www.citibikenyc.com/system-data), [Washington](https://www.capitalbikeshare.com/system-data). Some data wrangling, to reduce columns and reformat, has been performed to condense these files to the core six columns used in this project. This makes the analysis and the evaluation in this project more straightforward.



The data is provided by [Motivate](https://www.motivateco.com/), which is a bike share system provider for many cities in the United States. The data files for all three cities contain the same six columns:

* Start Time

* End Time

* Trip Duration (in seconds)

* Start Station

* End Station

* User Type (Subscriber or Customer)



The Chicago and New York City files also contain the following two columns:

* Gender

* Birth Year



## Questions explored

The script answers the following questions about the bike share data:



Popular times of travel

* What is the most popular month for start time?

* What is the most popular day of week (Monday, Tuesday, etc.) for start time?

* What is the most popular hour of day for start time?



Popular stations and trip

* What is the most popular start station and most popular end station?

* What is the most popular trip?



Trip duration

* What is the total trip duration and average trip duration?



User information

* What are the counts of each user type?

* What are the counts of gender?  (only available for NYC and Chicago)

* What are the earliest (i.e. oldest person), most recent (i.e. youngest person), and most popular birth years?



## Additional Business Analysis Questions to Consider

* How many bikes should each bikeshare have available?

* What is the optimal driving time between stations?

* Which users are more likely to pay for yearly subscription? 

* Which users are more likely to buy short-term, such as 3-day or 24 hour passes? 

* How does the use of bikes vary by cities? 

* How does the season impact bikeshare use?

* Are certain user age groups more likely to use the bikeshare system than others?



## Resources

* https://pandas.pydata.org/pandas-docs/stable/api.html#datetimelike-properties

* https://pandas.pydata.org/pandas-docs/stable/generated/pandas.to_datetime.html



The datetime module contains many useful methods for handling date filtering and extraction:

strftime converts date objects into readable strings: https://www.programiz.com/python-programming/datetime/strftime

strptime takes strings and converts them into date objects that Python can understand: https://www.programiz.com/python-programming/datetime/strptime

strf & strp time: https://docs.python.org/3/library/datetime.html#strftime-and-strptime-behavior

This stackoverflow answer explains perfectly how to extract day, month, year from a Date: https://stackoverflow.com/questions/21954197/which-is-the-fastest-way-to-extract-day-month-and-year-from-a-given-date/21954923#21954923