README.md

metadata

title: My Ai Twin
emoji: 🤖
colorFrom: blue
colorTo: indigo
sdk: gradio
sdk_version: 5.16.0
app_file: app.py
pinned: false
license: mit

AI Twin RAG Application

This application is an AI Twin of yourself, built using Retrieval-Augmented Generation (RAG). It answers questions based on your personal data stored in the knowledge_base/ directory.

Features

• Personal Knowledge Base: Uses your real and synthetic data to answer questions.
• RAG Pipeline: Utilizes LangChain, ChromaDB, and Hugging Face embeddings for accurate retrieval.
• Interactive UI: Built with Gradio for easy interaction.

Setup & Installation

1. Install Dependencies:

   pip install -r requirements.txt
   

2. Set Up Hugging Face Token:

   • You need a Hugging Face API token to use the inference endpoint.
   • Set it as an environment variable:

     export HUGGINGFACEHUB_API_TOKEN="your_token_here"
     # Windows PowerShell:
     $env:HUGGINGFACEHUB_API_TOKEN="your_token_here"
     

   • Or uncomment the line in app.py to set it directly.

3. Run Verification (Optional):

   python verify_rag.py
   

4. Launch the App:

   python app.py
   

   The app will run locally at http://127.0.0.1:7860.

Deployment to Hugging Face Spaces

1. Create a new Space on Hugging Face.
2. Select Gradio as the SDK.
3. Upload the following files:
   • app.py
   • requirements.txt
   • knowledge_base/ (entire folder)
   • chroma_db/ (optional, can be generated on the fly, but better to let it generate)
4. Add your HUGGINGFACEHUB_API_TOKEN in the Space settings (Settings -> Variables and secrets).

Files

• app.py: Main application logic.
• knowledge_base/: Directory containing your profile data.
• requirements.txt: Python dependencies.
• verify_rag.py: Script to test RAG logic without UI.