Hugging Face's logo

Spaces:
CamiloVega
/
Easy_RAG
Sleeping

App Files Community
Easy_RAG / README.md
CamiloVega's picture
CamiloVega
Update README.md
dfdb4d0 verified
preview code
|
raw
history blame contribute delete
2.76 kB

A newer version of the Gradio SDK is available: 5.14.0

Upgrade
metadata
title: Easy RAG
emoji: 🐢
colorFrom: purple
colorTo: indigo
sdk: gradio
sdk_version: 5.8.0
app_file: app.py
pinned: false
license: mit
short_description: Chat with your docs very Easy

DocumentGPT - Advanced Document Analysis with RAG

Overview

DocumentGPT is a cutting-edge document analysis system that leverages the power of Retrieval Augmented Generation (RAG) to provide intelligent responses based on your documents. Built with advanced AI technologies, it allows users to upload multiple document types and get accurate, context-aware responses to their questions.

Key Features

  • Multi-Format Support: Process PDF, DOCX, CSV, and TXT files seamlessly
  • Advanced RAG Implementation: Using state-of-the-art LLM technology with Llama-2
  • GPU-Accelerated: Optimized performance with GPU acceleration
  • Real-Time Processing: Dynamic document processing and instant responses
  • Source Attribution: Every response includes references to source documents
  • Interactive Interface: User-friendly Gradio interface for easy interaction

Technical Stack

  • Large Language Model: Llama-2-7b-chat-hf
  • Embeddings: multilingual-e5-large
  • Vector Store: FAISS
  • Framework: Gradio
  • Processing: Langchain
  • Acceleration: HuggingFace Accelerate

Getting Started

Prerequisites

  • Python 3.8 or higher
  • GPU support (recommended)
  • HuggingFace account with access to Llama-2

Installation 

# Clone the repository
git clone https://github.com/yourusername/documentgpt.git
cd documentgpt

# Install dependencies
pip install -r requirements.txt

# Set up environment variables
export HUGGINGFACE_TOKEN=your_token_here

Running the Application 

python app.py

Usage

  1. Launch the application
  2. Upload your documents (PDF, DOCX, CSV, or TXT)
  3. Wait for the processing to complete
  4. Start asking questions about your documents
  5. View responses with source attributions

Advanced Features

  • Dynamic Knowledge Base: Updates in real-time as new documents are added
  • Memory Management: Efficient handling of document processing
  • Source Tracking: Transparent attribution of information sources
  • Optimized Performance: GPU acceleration for faster processing

Author

Camilo Vega

Acknowledgments

Special thanks to:

  • HuggingFace for providing GPU acceleration support
  • Meta AI for the Llama-2 model
  • The Langchain community for their excellent tools

Contact

For questions and support, please reach out through:

Made with by Camilo Vega