🧠## 🌐 Live Demo

Try the app now: https://casual-competitor-susbk7aqfls5uxswyrp2qz.streamlit.app/

No installation required! Just upload your CSV file and start discovering causal relationships.

🎥 Demo Video

Watch the app in action! See how to upload data, configure analysis, and discover causal relationships:

Streamlit.-.Brave.2025-07-31.14-10-02.mp4

🔧 How to add your Google Drive video

1. Upload your demo video to Google Drive
2. Right-click the video → "Get shareable link"
3. Make sure it's set to "Anyone with the link can view"
4. Copy the file ID from the URL (the long string between /d/ and /view)
5. Replace YOUR_GOOGLE_DRIVE_FILE_ID in the link above with your actual file ID

Example: If your Google Drive link is: https://drive.google.com/file/d/1ABC123xyz789DEF456/view?usp=sharing

Then your file ID is: 1ABC123xyz789DEF456

Alternative display option (if you prefer the video to play directly in GitHub):

<div align="center">
  <iframe src="https://drive.google.com/file/d/1xcQMgNFicwbHpBuE5N1mBUsf5P5VdoEJ/view?usp=sharing" 
          width="640" height="360" 
          allow="autoplay">
  </iframe>
</div>

🎯 What the Demo Shows:

• Data Upload: How to upload and preview your CSV dataset
• Variable Selection: Choosing treatment and outcome variables
• Causal Graph: Viewing the generated causal relationship graph
• Inference Results: Understanding the causal effect estimates
• Error Handling: What to do when issues arise
• Real Example: Complete walkthrough with sample data

🌟 Features Discovery & Inference

A Streamlit web application for discovering causal relationships in your data using Microsoft's DoWhy library. This tool helps you identify and quantify causal effects between variables in your datasets through correlation-based graph discovery and rigorous causal inference.

� Live Demo

Try the app now: https://casual-competitor-susbk7aqfls5uxswyrp2qz.streamlit.app/

No installation required! Just upload your CSV file and start discovering causal relationships.

�🌟 Features

• Causal Graph Discovery: Automatically generates causal graphs using correlation-based analysis
• Causal Inference: Estimates causal effects using DoWhy's backdoor identification methods
• Interactive Web Interface: User-friendly Streamlit interface with real-time analysis
• Data Preprocessing: Automatic data cleaning, encoding, and preprocessing pipeline
• Fast Mode: Optimized performance for large datasets
• Traditional Analysis: Optional statistical analysis with correlation matrices and visualizations
• Error Handling: Comprehensive error handling with helpful debugging information

📋 Requirements

• Python 3.8+
• Virtual environment (recommended)

🚀 Quick Start

🌐 Option 1: Use the Live Demo (Recommended)

Simply visit: https://casual-competitor-susbk7aqfls5uxswyrp2qz.streamlit.app/

Upload your CSV file and start analyzing causal relationships immediately!

💻 Option 2: Local Installation

1. Clone the repository:

git clone https://github.com/Krishnadev-cmd/Casual-Competitor.git
cd Casual-Competitor

2. Create and activate virtual environment:

# Windows
python -m venv venv_py
venv_py\Scripts\activate

# Linux/Mac
python -m venv venv_py
source venv_py/bin/activate

3. Install dependencies:

pip install -r requirements.txt

4. Run locally:

streamlit run src/app.py

📦 Dependencies

• streamlit: Web application framework
• dowhy: Microsoft's causal inference library
• pandas: Data manipulation and analysis
• numpy: Numerical computing
• matplotlib: Plotting library
• seaborn: Statistical data visualization
• networkx: Graph analysis
• scikit-learn: Machine learning utilities
• pydot: Graph visualization (requires Graphviz)

🎯 Usage

🌐 Using the Live Demo

1. Visit: https://casual-competitor-susbk7aqfls5uxswyrp2qz.streamlit.app/
2. Upload your CSV file using the sidebar
3. Configure your analysis (see steps below)
4. Run and explore the results!

💻 Using Local Installation

1. Start the application:

streamlit run src/app.py

2. Open your browser and navigate to http://localhost:8501

📊 Analysis Configuration (Both Options)

• Upload a CSV file
• Specify treatment variable (what you want to change)
• Specify outcome variable (what you want to predict)
• Choose whether to show dataset analysis
• Enable fast mode for large datasets
• Run the analysis and explore the results!

📊 How It Works

1. Data Preprocessing

• Automatic ID column removal
• Missing value imputation
• Categorical variable encoding
• Feature scaling and normalization
• Date parsing and feature extraction

2. Causal Graph Discovery

• Correlation-based edge detection
• Automatic DAG (Directed Acyclic Graph) construction
• Treatment → Outcome path guarantee
• Threshold-based edge filtering (correlation > 0.3)

3. Causal Inference

• DoWhy's backdoor identification method
• Linear regression estimation
• Placebo treatment refutation tests
• Statistical significance testing

📁 Project Structure

Casual_Competitor/
├── src/
│   ├── app.py              # Main Streamlit application
│   ├── doWhy_utils.py      # DoWhy integration utilities
│   ├── Preprocess.py       # Data preprocessing pipeline
│   ├── Utils.py            # Helper functions
│   └── __pycache__/        # Python cache files
├── data/                   # Sample datasets
│   ├── Auto Sales data.csv
│   ├── bakery_Sales.csv
│   ├── big_mart_sales.csv
│   ├── Electronic_sales_Sep2023-Sep2024.csv
│   ├── retail_data.csv
│   └── Video_Games_Sales_as_at_22_Dec_2016.csv
├── venv_py/               # Virtual environment
├── requirements.txt       # Python dependencies
└── README.md             # This file

🔧 Configuration Options

Sidebar Controls

• Dataset Analysis: Show/hide traditional statistical analysis
• Preprocessing: Specify if your dataset is already preprocessed
• Fast Mode: Enable for faster processing on large datasets
• Treatment Variable: The variable you want to manipulate
• Outcome Variable: The variable you want to predict/analyze

Performance Tips

• Fast Mode: Recommended for datasets with >500 rows or >15 columns
• Preprocessing: Skip if your data is already clean and encoded
• Variable Selection: Choose variables with clear causal relationships

📈 Example Use Cases

1. Marketing Analysis:

   • Treatment: Marketing spend
   • Outcome: Sales revenue
   • Discover: How marketing investment affects sales

2. Pricing Strategy:

   • Treatment: Product price
   • Outcome: Customer demand
   • Discover: Price elasticity effects

3. A/B Testing:

   • Treatment: Feature flag (0/1)
   • Outcome: User engagement
   • Discover: Feature impact on user behavior

⚠️ Important Notes

• Causal vs Correlation: This tool identifies potential causal relationships, but domain expertise is still required for interpretation
• Data Quality: Results are only as good as your input data - ensure clean, relevant datasets
• Sample Size: Larger datasets generally produce more reliable causal estimates
• Variable Selection: Choose treatment and outcome variables with theoretical causal relationships

🐛 Troubleshooting

Common Issues

1. "Variable not found": Check that your treatment/outcome variables exactly match column names
2. "Graph contains cycles": Enable fast mode or check for circular relationships in your data
3. Zero causal effect: May indicate no causal relationship or insufficient data signal
4. Visualization errors: Ensure Graphviz is installed for graph plotting

Debug Information

The app provides comprehensive debug information including:

• Graph structure visualization
• Available column names
• Raw inference results
• Error details and suggestions

🤝 Contributing

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• Microsoft DoWhy: For the powerful causal inference framework
• Streamlit: For the excellent web application framework
• NetworkX: For graph analysis capabilities
• Pandas & NumPy: For data manipulation foundations

📞 Support

If you encounter any issues or have questions:

1. Check the troubleshooting section above
2. Look at the debug information in the app
3. Open an issue on GitHub
4. Review the DoWhy documentation for advanced usage

🚀 Future Enhancements

• Support for more causal discovery algorithms
• Advanced visualization options
• Export functionality for results
• Batch processing capabilities
• Integration with more causal inference methods
• Real-time data streaming support

---

Happy Causal Discovery! 🎉