CloudEye is a full-stack web application designed to predict the likelihood of cloudbursts across India using historical weather data and machine learning. It provides an intuitive user interface for entering weather parameters and receives predictions in real-time based on a trained CatBoost model.
-
🔮 Cloudburst Prediction
- Enter weather parameters like temperature, precipitation, humidity, wind, and more to receive a prediction using a CatBoost classification model.
-
📊 ML Model Powered
- Built on top of CatBoost, trained with features such as precipitation, humidity, wind gusts, cloud cover, etc.
-
🌐 State & District-Wide Coverage
- Designed to scale with regional data to analyze cloudburst chances across various Indian states and districts.
-
🖼️ User-Friendly Frontend
- Clean and responsive frontend using HTML, CSS, and JavaScript.
-
🗃️ PostgreSQL Database Integration
- Stores historical records, prediction logs, and user inputs securely for analysis and insights.
| Layer | Stack |
|---|---|
| Frontend | HTML5 · CSS3 · JavaScript |
| Backend | Python · CatBoost · Flask/FastAPI (optional) |
| Database | PostgreSQL |
| ML Model | CatBoost Classifier (trained via Google Colab) |
Trained on: Historical cloudburst-related weather data
Features Used:
Temperature, Precipitation, Wind Speed/Gusts, Cloud Cover
Relative Humidity, Atmospheric Pressure, Elevation, etc.
Output: Binary prediction — Cloudburst or No Cloudburst
Open Website: Open index.html from the Cloudburst Prediction folder in your browser.
Run the following in separate terminals:
Terminal 1:
cd frontend/RealTime
python app.py
Terminal 2:
cd frontend/Historical
python app1.py
Terminal 3:
cd /frotnend/Email
python cloudburst_checker.py
