Welcome to the repository for the “Predicting the Air Quality Index of Indian Cities using Machine Learning” competition, where the goal is to forecast the Air Quality Index (AQI) for major Indian AQI stations for the upcoming 28 days using machine learning techniques.

This solution secured the second place in the H2O World India Hackathon held in April 2023.

Air quality monitoring is crucial for our well-being, and this competition utilizes historical daily averages of various air pollutants to build accurate models that can predict the AQI for each station. Our team has extensively analyzed the data and constructed models that we believe will help us achieve our objective.

To navigate the repo:

• All the code for EDA and modeling can be found in the notebooks directory.
• data directory contains the data used for training and the model pkl files. It also contains the final submission.csv file.
• Rest of the notebook contains code for H2O-wave app we built and deployed it on Heroku.
  • app.py is the main script to run the wave-app.
  • components, pages, and utils directory contain scripts in a modular fashion to support different tabs/sections in the app.

Link to our Dashboard: Link. You can find more details on our approach on the solution tab in our dashboard.

1. Sagar Thacker (LinkedIn | GitHub)
2. Shuchita Mishar (LinkedIn | GitHub)