Rainfall Prediction Web App (Django + ML)

This project is a web-based rainfall prediction system built using Django and a trained Machine Learning model. The system takes in input features like temperature, precipitation, and wind speed to predict whether it will rain. The ML model is trained using logistic regression and integrated into the web app via a .pkl file.

🌐 Tech Stack

• Backend: Django (Python)
• Machine Learning: Scikit-learn (Logistic Regression)
• Frontend: HTML (Django templates), CSS (static styling)
• Model Deployment: joblib for loading the serialized model
• Database: SQLite3 (default for Django projects)

🧠 Model Training Notebook

The Jupyter notebook Rainfall Prediction.ipynb contains:

Data cleaning and preprocessing steps

Feature selection and target definition

Logistic Regression model training and evaluation

Exporting the final trained model using joblib

You can open this notebook to explore the entire model development process.

⚙️ Setup Instructions

1. Clone the Repository

git clone https://github.com/notRamish/rainfall-predictor.git
cd rainfall-predictor

2. Create Virtual Environment (Optional but Recommended)

python -m venv venv
source venv/bin/activate  # On Windows use: venv\Scripts\activate

3. Install Requirements

pip install -r requirements.txt

4. Run the Django App

python manage.py runserver

Then open your browser and visit:

http://127.0.0.1:8000/

5. Use the Interface

Fill in the form fields (temp, precip, wind, etc.)

Click "Predict"

View prediction result (Rain / No Rain)

🧠 ML Model Overview

Trained using logistic regression on historical weather data.

Feature engineering and training steps are documented in rainfall_prediction.ipynb.

The final model is saved as model/model.pkl.

🛠️ Notes

This project was built for educational/demo purposes.

You can improve it by replacing the model, adding feature scaling, or connecting to a real-time weather API.

👨‍💻 Author

Mohammad Ramish Anwar

• Feel free to fork, use, or improve this project!