README.md

metadata

title: AI Research Hub - Complete HuggingFace Demo
emoji: 🚀
colorFrom: blue
colorTo: purple
sdk: gradio
sdk_version: 5.32.0
app_file: app.py
pinned: false
license: mit
short_description: Complete HuggingFace Inference API demo platform
tags:
  - nlp
  - computer-vision
  - text-generation
  - question-answering
  - text-classification
  - zero-shot-classification
  - named-entity-recognition
  - text-to-image
  - sentiment-analysis
  - translation
  - summarization
  - feature-extraction
  - huggingface
  - inference-api
  - gradio
  - ai-demo

🚀 AI Research Hub - Complete HuggingFace Demo

A comprehensive demonstration of HuggingFace's Inference API capabilities, optimized for Hugging Face Spaces.

✨ Features

📝 Text Processing

• 💬 Chat - Conversational AI with advanced language models
• 🎭 Fill Mask - Text completion and prediction
• ❓ Question Answering - Extract answers from context
• 📝 Summarization - Automatic text summarization

🏷️ Classification & Analysis

• 🏷️ Sentiment Analysis - Emotion and sentiment detection
• 🎯 Zero-Shot Classification - Classify with custom labels
• 🏷️ Named Entity Recognition - Extract people, places, organizations
• 🧮 Text Similarity - Compare semantic similarity between texts

🎨 Multimodal Capabilities

• 🌐 Translation - English to French translation
• 🖼️ Image Classification - Identify objects and scenes in images
• 🎨 Text-to-Image - Generate images from text descriptions

🔧 Setup Instructions

For Hugging Face Spaces:

1. Fork/Duplicate this Space

2. Set up your HuggingFace Token:

   • Go to Settings → Repository secrets
   • Add a new secret: HF_TOKEN
   • Value: Your HuggingFace token from here

3. That's it! The Space will automatically restart and all features will be available.

For Local Development:

# Clone the repository
git clone <your-repo-url>
cd <repo-name>

# Install dependencies
pip install -r requirements.txt

# Set environment variable
export HF_TOKEN="your_token_here"

# Run the application
python app.py

🎯 Key Improvements

✅ Fixed Issues

• Question Answering: Proper input format for API calls
• Text Classification: Working models with proper error handling
• Zero-Shot Classification: Correct API method usage
• Named Entity Recognition: Fixed entity extraction and labeling
• Image Classification: Better error handling for uploads

🚀 Enhanced Features

• Spaces Optimization: Memory and performance optimized for HF Spaces
• Robust Error Handling: Clear error messages and fallback strategies
• Modern UI: Clean, responsive interface with organized tabs
• Token Management: Multiple token source detection for Spaces
• Text Similarity: Semantic comparison with cosine similarity scores

📊 Models Used

Task	Model	Description
Chat	microsoft/DialoGPT-medium	Conversational AI
Fill Mask	distilbert-base-uncased	Lightweight BERT model
Q&A	distilbert-base-cased-distilled-squad	SQuAD-trained model
Summarization	facebook/bart-large-cnn	CNN-trained BART
Sentiment	cardiffnlp/twitter-roberta-base-sentiment-latest	Twitter sentiment
Zero-Shot	facebook/bart-large-mnli	MNLI-trained BART
NER	dslim/bert-base-NER	CoNLL-trained BERT
Translation	Helsinki-NLP/opus-mt-en-fr	English-French translator
Embeddings	sentence-transformers/all-MiniLM-L6-v2	Sentence embeddings
Image Classification	google/vit-base-patch16-224	Vision Transformer
Text-to-Image	runwayml/stable-diffusion-v1-5	Stable Diffusion

🔒 Privacy & Security

• No Data Storage: All processing happens in real-time, no data is stored
• Secure Token Handling: Tokens are handled securely through Spaces secrets
• API Rate Limiting: Built-in handling for API rate limits and quotas

🚨 Troubleshooting

Common Issues:

1. "API client not available"

   • Solution: Set HF_TOKEN in Spaces settings

2. "Rate limit reached"

   • Solution: Wait a moment and try again

3. "Model loading"

   • Solution: Some models need time to load, retry after a few seconds

4. "Service unavailable"

   • Solution: Temporary HuggingFace service issue, try again later

🛠️ Technical Details

Architecture

• Frontend: Gradio with custom CSS
• Backend: HuggingFace Inference API
• Deployment: Optimized for HF Spaces environment
• Error Handling: Comprehensive error catching and user feedback

Performance Optimizations

• Memory Efficient: Minimal memory footprint for Spaces
• Fast Models: Selected for quick response times
• Graceful Degradation: Features degrade gracefully if dependencies missing
• Connection Pooling: Efficient API client management

📈 Future Enhancements

• Advanced Visualization: Interactive embedding plots
• Batch Processing: Multiple text processing
• Custom Model Support: User-specified models
• Audio Processing: Speech-to-text capabilities
• Real-time Streaming: Live text generation

🤝 Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Test thoroughly
5. Submit a pull request

📄 License

This project is open source and available under the MIT License.

🙏 Acknowledgments

• HuggingFace for the amazing Inference API and Spaces platform
• Gradio for the intuitive interface framework
• Open Source Community for the incredible model ecosystem

---

Built with ❤️ for the AI research community