HUGGINGFACE_SPACE_CONFIG.md

Hugging Face Space Configuration

This document contains the configuration needed to deploy this application as a Hugging Face Space.

Space Configuration

Basic Settings

• Space Name: chat-with-github-repo
• Space Type: Gradio
• Python Version: 3.11
• Visibility: Public

Environment Variables (Optional)

Set these in your Hugging Face Space settings for better performance:

HUGGINGFACE_API_KEY=your_hf_token_here
GITHUB_TOKEN=your_github_token_here

Hardware Requirements

• CPU: Basic (free tier works)
• RAM: 8GB+ recommended for larger repositories
• Storage: 10GB+ for model caching

Deployment Steps

1. Create a new Hugging Face Space:

   • Go to https://huggingface.co/new-space
   • Choose "Gradio" as the Space SDK
   • Set the space name and visibility

2. Upload files:

   • Upload all files from this directory to your space
   • Ensure the main app.py file is in the root directory

3. Configure environment variables (optional):

   • Go to your space settings
   • Add the environment variables listed above
   • This improves rate limits and enables private repo access

4. Deploy:

   • The space will automatically build and deploy
   • First deployment may take 5-10 minutes due to model downloads

File Structure for Hugging Face Space

your-space/
├── app.py                 # Main Gradio application
├── requirements.txt       # Python dependencies
├── README.md             # Space documentation
├── config.py             # Configuration settings
├── services/             # Service modules
│   ├── __init__.py
│   ├── github_service.py
│   ├── embedding_service.py
│   └── chat_service.py
├── utils/                # Utility modules
│   ├── __init__.py
│   └── file_processor.py
└── models/               # Data models
    ├── __init__.py
    └── schemas.py

Performance Optimization

For Free Tier:

• Uses lightweight embedding model (all-MiniLM-L6-v2)
• Processes files in batches
• Implements file size limits
• Caches models locally

For Better Performance:

• Upgrade to paid hardware
• Use larger embedding models
• Increase batch sizes
• Add Redis caching

Troubleshooting

Common Issues:

1. Out of Memory:

   • Reduce batch size in embedding service
   • Use smaller embedding model
   • Upgrade hardware

2. Slow Processing:

   • Add Hugging Face API token for better rate limits
   • Use GPU hardware
   • Optimize chunk sizes

3. Git Clone Failures:

   • Add GitHub token for private repos
   • Check repository URL format
   • Ensure repository is public

Debug Mode:

Set debug=True in demo.launch() for detailed error messages.

Monitoring

Monitor your space performance:

• Check space logs for errors
• Monitor memory usage
• Track processing times
• Review user feedback

Updates

To update your space:

1. Modify files locally
2. Upload changed files to your space
3. Space will automatically rebuild
4. Test functionality after deployment