Generative AI - Information Retrieval System Using LangChain and PaLM2

This repository contains an Information Retrieval System built using LangChain and Generative AI (GenAI) technologies. The system is designed to retrieve relevant information from a knowledge base or external documents using state-of-the-art natural language processing techniques. By leveraging LangChain's modular framework and the power of GenAI, this project enables efficient, context-aware information retrieval for a wide range of applications.

Features

• Contextual Search: Retrieves information based on natural language queries.
• Generative Responses: Uses GenAI to provide detailed, human-like responses.
• Document Parsing: Processes and indexes unstructured documents for efficient search.
• Integration with LangChain: Utilizes LangChain to manage retrieval pipelines.
• Extensibility: Easily customizable for different datasets and domains.

How to run?

STEPS:

Clone the repository

Project repo: https://github.com/vijaybalamahalingam/Information-Retrieval-System-GenAI-Using-LangChain.git

STEP 01- Create a conda environment after opening the repository

conda create -n llmapp python -y

conda activate llmapp

STEP 02- install the requirements

pip install -r requirements.txt

Create a .env file in the root directory and add your GOOGLE_API_KEY as follows:

GOOGLE_API_KEY= "xxxxxxxxxxxxxxxxxxxxxxxxxxxxx"

# Finally run the following command
streamlit run app.py

Now,

open up : http://localhost:8501

Techstack Used:

• Python
• LangChain
• Streamlit
• PaLM2
• FAISS