Skip to content

Latest commit

 

History

History

demo_loader

README.md

Project Description: Loading Documents into Cosmos DB for MongoDB VCore Vector Store

Overview

This project aims to demonstrate the process of loading documents into a Cosmos DB for MongoDB VCore Vector Store using LangChain Loaders, while also loading images into an Azure Storage Account.

Features

  • Integration of LangChain Loaders for seamless document loading into Cosmos DB for MongoDB VCore Vector Store.
  • Utilization of Azure Storage Account for efficient storage and retrieval of images associated with the documents.
  • Demonstrates how to set up and configure the environment for document and image loading tasks.

Requirements

  • Azure subscription for deploying Cosmos DB for MongoDB VCore and Azure Storage Account.
  • Python environment with LangChain and Azure SDK installed.
  • Basic knowledge of MongoDB, Azure Cosmos DB, and Azure Storage concepts.

Usage

  1. Set up Cosmos DB for MongoDB VCore and Azure Storage Account in your Azure subscription.
  2. Clone the repository to your local machine.
  3. Create .env file and populate:
  • OPENAI_API_KEY=''
  • MONGO_CONNECTION_STRING=''
  • AZURE_STORAGE_CONNECTION_STRING=''
  1. Create pythonn env:
python -m venv venv
  1. Install Requirements:
venv\Scripts\activate
python -m pip install -r requirements.txt
  1. Load sample docuemtns vectorstoreloader.py
python vectorstoreloader.py

License

This project is licensed under the MIT License, granting permission for commercial and non-commercial use with proper attribution.

Support

For any questions or issues, please open an issue on GitHub or reach out to the project maintainers.

Disclaimer

This project is provided for educational and demonstration purposes only. Use at your own risk.