/documents-retrieval-agent

Chat With Multiple PDF Documents using Conversational RAG on CPU with LLAMA2, Langchain ChromaDB

Primary LanguagePythonApache License 2.0Apache-2.0

Chat With Multiple PDF Documents using Conversational RAG on CPU with LLAMA2, Langchain ChromaDB

This is a Conversational Retrieval Augmented Generation (RAG) Knowledge Base Chat built on top of LLAMA2 (Embeddings & Model), Langchain and ChromaDB and orchestrated by FastAPI framework to provide and Endpoint for easy communication.

Quickstart

Conversational RAG runs offline on local CPU

  1. Setup a virtual environement & Install the requirements:
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt`

  1. Copy your PDF files to the documents folder.

  2. Run the FastAPI server, to process and ingest your data on start with the LLM RAG and return the answer:

python main.py "What is the invoice number value?"