Ride Sharing Bike Project

• There is an increase in cycling culture around the world and automated cycle renting systems are being set up. Cycles are taken from one stop and dropped at any other stop. Cycles are used depending on many features such as weather, month, year. Being able to predict the requirement of cycles will optimise the cycle renting systems. They will help to use the right number of cycles and make the most out of them.

• This is the problem that we have solved. We have build a model that will predict the number of cycles required/consumed each day. We have used a SVR model and made a webapp which will show the prediction for the given input of date and a few more parameters.

Setup

• Download or clone the repository. In case you are cloning the repository, run the following command:

git clone https://github.com/IEEE-NITK/Ride-Sharing-Bike-Project/

• To run the Flask app you need to have Python 3 pre-installed in your system. After that create a virtual environment and run the following commands.

virtualenv venv
pip install -r requirements.txt

• To run the Flask app, run the following commands:

python server.py

Supported Platforms:

• Windows
• Linux
• MacOS

Copyright and license

• Code released under the MIT license.

Project Members

Third Years

• Salman Shah
• Hrishi Hiraskar
• Aditya A
• Mohita Chowdhury
• Jeshventh Raja

Second Years

• Anumeha Agrawal
• Adil Ashique
• Rosa George

References

• Modeling bike availability in a bike-sharing system using machine learning
• Bench marking of classification algorithms: Decision Trees and Random Forests - a case study using R
• Station segmentation of Hangzhou public free-bicycle system based on improved randomized algorithm