This project demonstrates the integration of multiple AI agents using Retrieval-Augmented Generation (RAG) and WikiSearch functionalities, along with Cassandra for data storage and retrieval. It leverages Llama 3.1 for natural language processing and "all-MiniLM-L6-v2" for generating embeddings.
This project, implemented in a Jupyter Notebook, showcases how to:
- Connect to Cassandra.
- Load and process data from web pages using
WebBaseLoader
. - Convert text to vectors using the "all-MiniLM-L6-v2" Hugging Face embedding model.
- Perform document retrieval and question routing using LangChain and related libraries.
- Generate human-like responses using the Llama 3.1 LLM via GROQ.
- Cassandra Integration: Connects to Cassandra for storing and retrieving document embeddings.
- LangChain: Utilizes LangChain for web page loading, text splitting, embedding, and vector store operations.
- WebBaseLoader: Uses
WebBaseLoader
to efficiently load content from web pages. - Hugging Face Embeddings: Employs the "all-MiniLM-L6-v2" model for generating high-quality text embeddings.
- Llama 3.1 with GROQ: Integrates the Llama 3.1 LLM using GROQ for refined, human-like responses.
- WikiSearch: Integrates the Wikipedia API for answering questions not related to stored documents.
- Query Routing: Routes user queries to either the Cassandra vector store or WikiSearch based on relevance.