Skip to content

VectorPDFChat: This part emphasizes the use of vector embeddings for PDF documents and the chatbot nature of the system. RAGKnowledgeBase: This part clearly indicates the use of RAG and the knowledge base aspect.

Notifications You must be signed in to change notification settings

Himel-Tasrif/VectorPDFChat-RAGKnowledgeBase

Repository files navigation

Haor PDF Chatbot

A custom knowledge-based PDF query chatbot using RAG, a VectorDB, and the Mistral-7B-Instruct-v0.2 LLM. Built with Gradio for an interactive web interface.

Live Demo: https://huggingface.co/spaces/himel06/Haor_PDF_Chatbot


Features

  • Extracts knowledge from PDFs using Retrieval-Augmented Generation (RAG).
  • Stores embeddings in a VectorDB (e.g.Chromadb).
  • Interfaces with mistralai/Mistral-7B-Instruct-v0.2 LLM.
  • Multiple approaches: different .py modules (e.g., app.py, brain.py) offering alternative workflows.
  • Gradio-based web UI for an easy interactive experience.

Requirements

Make sure you have Python 3.10+ installed, then clone and install dependencies:

git clone https://huggingface.co/spaces/himel06/Haor_PDF_Chatbot.git
cd Haor_PDF_Chatbot
pip install -r requirements.txt

All Rights Reserved by @Tasrif Nur Himel

About

VectorPDFChat: This part emphasizes the use of vector embeddings for PDF documents and the chatbot nature of the system. RAGKnowledgeBase: This part clearly indicates the use of RAG and the knowledge base aspect.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published