chat-with-multiple-PDFs

1. Introduction

Building a chatbot that leverages Retrieval-Augmented Generation (RAG) with the Ensemble Retriever technique. It is engineered not only to summarize but also to facilitate Q&A interactions with the content of user-uploaded PDFs using the Langchain framework through a user-friendly Streamlit interface.

Streamlit Interface

Key features:

🌟 RAG with Ensemble Retriever: Utilizes the RAG technique with an Ensemble Retriever (Hybrid Search).

🌟 Interact with PDFs: The system allows interaction with multiple PDF files uploaded by users.

🌟 Summarization: Capable of summarizing and displaying the content of PDF files.

🌟 Streamlit Interface: The interface is designed with Streamlit, user-friendly and easy to use.

2. How to use?

Getting Started

git clone https://github.com/dinhquy-nguyen-1704/chat-with-multiple-PDFs.git
cd chat-with-multiple-PDFs

conda create -n chat-with-multiple-PDFs python=3.9.6
conda activate chat-with-multiple-PDFs

pip install -r requirements.txt

Using Streamlit

Add the OpenAI API KEY in the .env file.

OPENAI_API_KEY=sk-...

Then, you can run the app.py file and a Streamlit interface will appear.

python -m streamlit run app.py

I have prepared a Vietnamese PDF file in the sample folder for you to test.

Run on Colab

I have also provided you with a Google Colab Notebook for the convenience of running code, you just need to replace your OPEN_AI_API in the .env file.