tutorials

📚 EAGLE Tutorials

Welcome to the EAGLE tutorials! These guides will help you master the framework from basics to advanced usage.

🎯 Tutorial Overview

1. Getting Started

For beginners and first-time users

• Installation and setup
• Running your first analysis
• Understanding outputs
• Basic Python API usage
• Attribution analysis basics

Time: 20-30 minutes

2. Advanced Usage

For experienced users and researchers

• Custom dataset preparation
• Model architecture customization
• Advanced attribution techniques
• Performance optimization
• Ensemble methods
• Production deployment

Time: 45-60 minutes

🗺️ Learning Path

graph LR
    A[Installation] --> B[First Analysis]
    B --> C[Understanding Results]
    C --> D[Attribution Analysis]
    D --> E[Custom Datasets]
    E --> F[Model Customization]
    F --> G[Advanced Features]

Loading

💡 Quick Tips

1. Start with the provided datasets (GBM, IPMN, NSCLC) before using custom data
2. Always enable attribution analysis for clinical interpretability
3. Use the Python API for more control than command-line interface
4. Save your configurations for reproducible experiments
5. Check the visualizations - they provide intuitive insights

📊 Example Datasets

EAGLE comes with three pre-configured cancer datasets:

Dataset	Type	Features	Best For
GBM	Brain Cancer	MRI + Clinical + Reports	Learning basics
IPMN	Pancreatic	CT + Clinical + Reports	Risk stratification
NSCLC	Lung Cancer	CT + Clinical + Reports	Survival prediction

🛠️ Common Workflows

Research Workflow

1. Start with baseline comparison (--mode all)
2. Analyze attribution for best model
3. Iterate on model configuration
4. Generate publication-ready figures

Clinical Workflow

1. Train on your cohort data
2. Enable attribution analysis
3. Examine high-risk patients
4. Validate with clinical outcomes

Development Workflow

1. Create custom dataset configuration
2. Implement custom encoders if needed
3. Optimize hyperparameters
4. Deploy with batch inference

📚 Additional Resources

• API Reference - Detailed function documentation
• FAQ - Common questions answered
• Troubleshooting - Solutions to common issues
• GitHub Examples - Code examples

🤝 Getting Help

If you need help:

1. Check the relevant tutorial section
2. Review the FAQ
3. Search GitHub Issues
4. Open a new issue with a minimal example

🎓 Tutorial Notebooks

Interactive Jupyter notebooks are coming soon:

• 01_basic_usage.ipynb - Interactive getting started
• 02_custom_data.ipynb - Preparing your own data
• 03_attribution_deep_dive.ipynb - Advanced interpretability
• 04_model_optimization.ipynb - Hyperparameter tuning

Stay tuned for updates!

---

Happy learning! The EAGLE framework is designed to be both powerful and accessible. These tutorials will help you unlock its full potential for your survival prediction tasks.