Purposes

The primary aim of our project was to predict the number of bike riders in a specific Bay Area city on a given day by analyzing various factors, including time, weather conditions, and rider behavior.

To achieve this, we focused on understanding travel trends for selected cities by thoroughly exploring and preparing the dataset. This process involved data manipulation, extraction, and preprocessing, followed by the application of regression algorithms and techniques to forecast ridership patterns.

Dataset Source: Bay Area Bike Share Dataset

Algorithms Selected to Train and Evaluate

• RandomForestRegression
• ExtraTreesRegressor
• XGBoostRegressor
• GradientBoostingRegressor

Running and Comparing all Algorithms

Model Performance Summary

Built With

• python