Bike Sharing Demand — AutoGluon

Predicting bike rental demand using tabular AutoML (AutoGluon) on the classic Kaggle "Bike Sharing Demand" dataset.

Project goal

The goal is to predict the number of rented bikes (count) for each timestamp based on weather, calendar, and time-of-day features.
This is useful for real-world demand forecasting (mobility, delivery, fleet planning).

Competition: Kaggle "Bike Sharing Demand".

What I did

• Loaded the provided train.csv / test.csv
• Engineered time features from datetime (hour, weekday, etc.)
• Trained several AutoGluon Tabular models
• Generated predictions for count on the test set
• Submitted the results to Kaggle and recorded the score

Tech stack

• Python
• Jupyter Notebook
• AutoGluon Tabular
• pandas / numpy / matplotlib

Reproduce

1. Create and activate a virtual environment
2. Download the Kaggle data (train.csv, test.csv) and put them in data/
3. Open the notebook in notebooks/ and run all cells to train and create a submission file

Result

Best public leaderboard score: 0.46182 RMSLE.

AutoGluon builds an ensemble of different models and ranks them. I compared multiple presets / training time limits and selected the best run for submission.

---

This project was done as part of the ML coursework to practice demand forecasting on tabular data and to build a full Kaggle workflow (training → inference → submission).