README.md

metadata

title: AI Data Science Assistant
emoji: 🤖
colorFrom: blue
colorTo: purple
sdk: gradio
sdk_version: 5.33.2
app_file: app.py
pinned: false

🤖 AI Data Science Assistant

A powerful AI assistant for data analysis, document Q&A, and general conversation - completely free and open-source!

🌟 Features

📊 CSV Data Analysis

• Upload & Analyze: Drop your CSV files for instant analysis
• Smart Visualizations: Auto-generates 4 types of charts:
  • Data types distribution (pie chart)
  • Missing values analysis (bar chart)
  • Numeric distributions (histograms)
  • Correlation matrix (heatmap)
• Statistical Summary: Complete descriptive statistics for numeric columns
• Data Preview: Shows dataset shape, columns, and first 5 rows

📚 PDF Document Q&A (RAG)

• Upload PDFs: Process any PDF document for question answering
• Smart Retrieval: Uses RAG (Retrieval Augmented Generation) with FAISS vector database
• Context-Aware: Answers questions based on actual document content
• Source Attribution: References the source document in responses
• Persistent Memory: Remembers uploaded documents during your session

💬 General AI Chat

• Claude-Style Responses: Clear, helpful, and honest communication
• No Hallucination: Says "I don't know" when uncertain
• Multi-Turn Conversations: Maintains conversation context
• Wide Knowledge: Helps with coding, explanations, creative tasks, and more

🚀 Try It Now

Live Demo: https://huggingface.co/spaces/mrradix/ai-ds-assistant

🎯 Use Cases

For Data Scientists & Analysts

• Quick CSV exploration and visualization
• Statistical analysis without writing code
• Data quality assessment (missing values, distributions)
• Correlation analysis between variables

For Researchers & Students

• PDF document analysis and Q&A
• Extract insights from research papers
• Ask questions about uploaded documents
• Get explanations of complex concepts

For General Users

• AI-powered conversations
• Help with various tasks and questions
• Document analysis and summarization
• Data visualization assistance

🛠️ Technical Stack

Models (100% Open Source)

• Language Model: google/flan-t5-base - Google's instruction-tuned T5
• Embeddings: sentence-transformers/all-MiniLM-L6-v2 - For document retrieval
• Vector Store: FAISS - Efficient similarity search

Libraries & Frameworks

• Frontend: Gradio - Interactive web interface
• Data Processing: Pandas, NumPy - Data manipulation
• Visualization: Matplotlib - Chart generation
• AI/ML: Transformers, PyTorch - Model inference
• RAG Pipeline: LangChain - Document processing and Q&A
• PDF Processing: PDFMiner - Text extraction

📖 How to Use

1. CSV Analysis

1. Go to the "📊 CSV Analysis" tab
2. Upload your CSV file using the file uploader
3. Click "📈 Analyze CSV"
4. View the automatic analysis, statistics, and charts

2. PDF Q&A

1. Switch to the "📚 PDF Q&A" tab
2. Upload a PDF document
3. Click "📤 Process PDF" and wait for confirmation
4. Ask questions about the document content
5. Get AI-powered answers based on the document

3. General Chat

1. Visit the "💬 General Chat" tab
2. Type your question or message
3. Get helpful, Claude-style responses
4. Continue the conversation naturally

🎨 Screenshots

CSV Analysis Interface

The CSV tab provides instant data insights with professional visualizations and statistical summaries.

PDF Q&A Interface

Upload any PDF and ask questions - the AI will answer based on the actual document content.

General Chat Interface

Natural conversation interface for any topic or question.

🔧 Local Installation

Want to run this locally? Here's how:

# Clone the repository
git clone https://huggingface.co/spaces/mrradix/ai-ds-assistant
cd ai-ds-assistant

# Install dependencies
pip install -r requirements.txt

# Run the application
python app.py

Requirements

• Python 3.8+
• 4GB+ RAM recommended
• GPU optional (will use CPU if not available)

🌟 Key Advantages

✅ Completely Free

• No API keys required
• No usage limits
• Open-source models only

✅ Privacy-First

• All processing happens locally (in the Space)
• No data sent to external APIs
• Your documents stay private

✅ Production-Ready

• Robust error handling
• Professional UI/UX
• Mobile-responsive design

✅ Educational

• Learn about RAG systems
• Understand AI model deployment
• Explore data science workflows

🤝 Contributing

This is an open-source project! Contributions are welcome:

1. Report Issues: Found a bug? Open an issue
2. Feature Requests: Have an idea? Let's discuss it
3. Pull Requests: Improvements and fixes are appreciated
4. Documentation: Help improve the docs

📄 License

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

🙏 Acknowledgments

• Google: For the Flan-T5 model
• Hugging Face: For the amazing model hub and Spaces platform
• LangChain: For the RAG framework
• Gradio: For the intuitive interface framework
• Open Source Community: For all the incredible libraries used

📧 Contact & Support

• Creator: @mrradix
• Issues: GitHub Issues
• Discussions: Space Discussions

---

⭐ If you find this helpful, please give it a star and share with others!

Built with ❤️ using open-source AI models