ninefyi / global-ai-conf-23

Build Semantic Search Application with Azure OpenAI service and MongoDB Atlas Search

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Build Semantic Search Application with Azure OpenAI service and MongoDB Atlas Search

  • Global AI Conference 2023 - Tue, 12 Dec 2023

  • Stack: Azure OpenAI Service, FastAPI, VueJS, LangChain, MongoDB Atlas Search (Atlas Vector Search)

  • Slide

  • Reference 1:

  • Reference 2:

  • Reference 3:

  • Create Atlas Vector Search Index

{
  "name": "vector_index",
  "type": "vectorSearch",
  "fields": [
    {
      "numDimensions": 1536,
      "path": "plot_embedding_azureopenai",
      "similarity": "euclidean",
      "type": "vector"
    },
    {
      "path": "genres",
      "type": "filter"
    },
    {
      "path": "year",
      "type": "filter"
    }
  ]
}
  • Create .env.test
MONGODB_URI=""
AZURE_OPENAI_ENDPOINT=""
OPENAI_API_TYPE="azure"
OPENAI_API_KEY=""
OPENAI_API_VERSION ="2023-05-15"
  • Run FastAPI uvicorn main:app --host 127.0.0.1 --port 8000 --reload
  • Run VueJS npm run dev

Create Vector Search Index

image

FastAPI

image

VueJS

image

Azure AI Studio

image

<style> .custom-image { width: 800px; height: auto; } </style>

About

Build Semantic Search Application with Azure OpenAI service and MongoDB Atlas Search

License:Apache License 2.0


Languages

Language:Vue 51.5%Language:Python 20.3%Language:CSS 16.3%Language:Shell 3.6%Language:Dockerfile 3.4%Language:JavaScript 2.7%Language:HTML 2.1%