This repo is on the analysis of the Bay Area Bike Share data. The project has three parts:

• Data preprocessing including data cleaning, handling missing data, data sampling, integration of different data sets, and preparing the data for machine learning models
• Data exploration to understand trends in the different data sets
• Predicting future bike availability at different stations

Jupitor notebooks are included in each folder for summarising the main results.

1. git clone the current repo
2. Install the dependencies in the requirements.yml file by running conda env create -f environment.yml

Data files were not inlcuded in this repository due to their large sizes. They are freely available at https://s3.amazonaws.com/fordgobike-data/index.html