🌿 VayuSense - Carbon Emission Prediction Platform

VayuSense is an AI-powered platform for predicting carbon emissions at both national and vehicle levels. Utilizing advanced machine learning techniques, VayuSense aims to provide accurate data-driven insights to combat climate change effectively.

🌐 Live Demo: vayusense.streamlit.app

✨ Features

• Country-wise Predictions: AI-based forecasting of CO₂ emissions for countries until 2050.
• Vehicle Estimator: Calculate lifetime emissions and trip-specific carbon footprints.
• Data Visualization: Interactive charts and graphs for better insight.
• Eco Score: Vehicle emission efficiency rating.
• Export Results: Download predictions in CSV format.
• Glass Morphism UI: Modern, visually appealing interface with glass morphism effects.
• Neon Glow Effects: Enhanced visual aesthetics with neon accents.

📂 Project Structure

VayuSense/
├── data/                   # Data visualizations and datasets
├── models/                 # Trained ML models
├── notebooks/              # Jupyter notebooks for analysis
├──screenshots              # EDA out put screenshots
├── src/                    # Source code modules
│   └── vehicle_estimator.py
├── streamlit_app/          # Streamlit application
│   ├── app.py              # Main application file
│   ├── favicon.ico         # Browser favicon
│   └── requirement.txt     # Required Package
└── README.md               # Project documentation

🛠️ Technologies Used

• Machine Learning: Random Forest, XGBoost
• Frontend: Streamlit with custom CSS
• Visualization: Plotly
• Data Processing: Pandas, NumPy
• Model Serialization: Joblib

🚀 Getting Started

📝 Prerequisites

• Python 3.8 or above
• Streamlit
• Plotly
• scikit-learn
• joblib

⚙️ Installation

1. Clone the repository:

   git clone https://github.com/itz-nirmal/carbon-emission.git

2. Navigate to the project directory:

   cd carbon-emission/streamlit_app

3. Install the required packages:

   pip install -r requirements.txt

📈 Usage

1. Run the Streamlit app:

   streamlit run app.py

2. Follow the on-screen instructions to navigate through different prediction modules.

📊 Data Visualizations \u0026 Analysis

Global CO2 Emissions Over Time

CO2 Emission Trend for Top 5 Emitters

Top 10 CO2 Emitters (2011)

GDP vs CO2 Emissions Correlation

Energy Use vs CO2 Emissions

CO2 Emissions Distribution by Income Group

Correlation Matrix of Features

Average CO2 Emission Growth Rates by Decade

🌐 Deployment

Deploy on Streamlit Cloud

1. Fork this repository
2. Sign up at streamlit.io
3. Click "New app" and select your forked repository
4. Set the main file path to streamlit_app/app.py
5. Click "Deploy"

Local Deployment

# Clone the repository
git clone https://github.com/itz-nirmal/carbon-emission.git
cd carbon-emission/streamlit_app

# Install dependencies
pip install -r requirements.txt

# Run the app
streamlit run app.py

❓ FAQ

Q: Why am I getting a "model not found" error? A: Make sure you've run the Jupyter notebooks in the notebooks/ folder to train and save the models.

Q: Can I add more countries to the prediction list? A: Yes! You can modify the countries list in app.py and ensure you have corresponding data.

Q: How accurate are the predictions? A: The model achieves approximately 92% accuracy based on historical data patterns.

🤝 Contributing

We welcome contributions to enhance the platform, improve predictions or usability. Feel free to submit pull requests or open issues.

📝 License

This project is licensed under the MIT License.

---

Made with ❤️ for a sustainable future 🌍