Question-Answer Chatbot with LangChain and Streamlit

This project is an end-to-end Question and Answer (QA) chatbot that accepts a PDF or paragraph, processes the input using LangChain, and allows users to ask questions based on the provided content. The model retrieves relevant sections and gives accurate answers accordingly.

Features

• PDF and Text Support: Upload PDFs or paste paragraphs for analysis.
• Interactive Q&A: Ask questions based on the provided input and receive relevant answers.
• Streamlit Web Interface: User-friendly web application powered by Streamlit for easy interaction.

Prerequisites

• Python: Version 3.10 or higher
• Conda: For environment management

Setup Instructions

Follow these steps to set up and run the application:

1. Clone the Repository

git clone https://github.com/alokranjan609/Question-Answer-Chatbot.git
cd Question-Answer-Chatbot

2. Create a Conda Environment

Create a Conda environment using Python 3.10:

conda create -p venv python==3.10

3. Activate the Environment

Activate the Conda environment:

conda activate ./venv

4. Install Required Dependencies

Install the necessary packages from the requirements.txt file:

pip install -r requirements.txt

5. Run the Application

Navigate to the folder containing the end_to_end.py file and run the Streamlit application:

python -m streamlit run end_to_end.py

6. Access the Application

The application will be running locally at:

http://localhost:8501

Usage

1. Upload a PDF file or paste a paragraph into the input area.
2. Ask questions about the provided content.
3. Get relevant answers based on the content you provided.

Project Structure

• end_to_end.py: The main Streamlit app file for running the Q&A system.
• requirements.txt: Lists the dependencies required to run the application.
• chain.ipynb: Jupyter notebook for experimenting with the QA system.