RAG System with Ollama - Python Documentation Retrieval

This project implements a Retrieval-Augmented Generation (RAG) system that leverages Ollama for answering questions based on Python documentation. The system fetches relevant documentation chunks from a database, retrieves them, and generates an accurate response by leveraging both retrieval and generative capabilities.

Overview

This project aims to help users retrieve information and documentation from Python's official documentation and answer questions related to Python libraries, modules, or code practices using a retrieval-augmented approach.

Key Features:

• RAG (Retrieval-Augmented Generation): Combines document retrieval and language generation to provide more accurate and context-aware answers.

• Ollama Local API: Uses the Ollama API locally to query and retrieve documentation, allowing questions to be answered based on specific pieces of information extracted from Python docs.

• Python Docs-based Knowledge: The system is specifically designed to interact with Python’s official documentation.

• Chroma Database: Chroma is used to persist and query chunks of Python documentation.

RAG System with LangChain, Chroma, and Ollama

This project implements a Retrieval-Augmented Generation (RAG) system using LangChain, Chroma, and Ollama for local document-based question answering. It uses document chunks stored in Chroma's vector database and generates answers based on the retrieved context using the Ollama model.

Requirements

To get started, you'll need to install the necessary Python libraries. You can install the required packages with the following:

pip  install  langchain-community  langchain-chroma  langchain-ollama

Requirements

• Required libraries:

• langchain_community

• langchain-chroma

• langchain-ollama

Setup Instructions

1. Clone the repository:

git clone https://github.com/lynah2/Python-Documentation-Query-System-with-RAG.git
cd Python-Documentation-Query-System-with-RAG

2. Install dependencies:

python -m venv env
source env/bin/activate  # macOS/Linux
.\env\Scripts\activate   # Windows

3. Download models:

ollama pull mistral  # Replace with the model you want to use

4. Prepare Chroma Database:

python populate_database.py

Usage

To query the system, run the following command:

python query_system.py

Example: