Skip to content

Latest commit

 

History

History
64 lines (41 loc) · 4.96 KB

readme.md

File metadata and controls

64 lines (41 loc) · 4.96 KB

Vector search in Python (Azure AI Search)

This repository contains multiple notebooks that demonstrate how to use Azure AI Search for vector and non-vector content in RAG patterns and in traditional search solutions.

Start with azure-search-vector-python-sample.ipynb for the basic steps. The code reads the data/text-sample.json file, which contains the input strings for which embeddings are generated. Output is a combination of human-readable text and embeddings that's pushed into a search index.

Python Vector Video

Once you understand the basics, continue with the following notebooks for more exploration:

Sample Description
azure-search-backup-and-restore.ipynb Backup retrievable index fields and restore it on a different search service.
azure-search-custom-vectorization-sample.ipynb Integrated data chunking and vectorization using custom skills and open source models.
azure-search-integrated-vectorization-sample.ipynb Integrated data chunking and vectorization (preview) using a skills to split text and call an Azure OpenAI embedding model.
azure-search-vector-image-index-creation-python-sample.ipynb Vectorization using Azure AI Vision image embedding.
azure-search-vector-image-python-sample.ipynb Vectorize images using Azure AI Vision image retrieval.
azure-search-vector-python-huggingface-model-sample.ipynb Vectorize using Hugging Face E5-small-V2 embedding model.
azure-search-vector-python-langchain-sample.ipynb LangChain integration.
azure-search-vector-python-llamaindex-sample.ipynb LlamaIndex integration.
azure-search-vector-python-sample.ipynb Basic vector indexing and queries. Start here.

Prerequisites

To run the Python samples in this folder, you will need the following:

  • An Azure subscription, with access to Azure OpenAI.
  • Azure AI Search, any tier, but choose a service that can handle the workload. We recommend Basic or higher.
  • A deployment of the text-embedding-ada-002 embedding model on Azure OpenAI.
  • Azure OpenAI connection and model information:
    • Azure OpenAI API key
    • Azure OpenAI embedding model deployment name (we name deployments after the model name: "text-embedding-ada-002")
    • Azure OpenAI REST API version (we recommend 2023-05-15)
  • Python (these instructions were tested with version 3.11.x)

You can use Visual Studio Code with the Python extension for these demos.

Set up your environment

  1. Clone this repository.

  2. Create a .env based on the code/.env-sample file. Copy your new .env file to the folder containing your notebook and update the variables.

  3. If you're using Visual Studio Code with the Python extension, make sure you also have the Jupyter extension.

Run the code

  1. Open the code folder and sample subfolder. Open a ipynb file in Visual Studio Code.

  2. Optionally, create a virtual environment so that you can control which package versions are used. Use Ctrl+shift+P to open a command palette. Search for Python: Create environment. Select Venv to create an environment within the current workspace.

  3. Copy the .env file to the subfolder containing the notebook.

  4. Execute the cells one by one, or select Run or Shift+Enter.

Troubleshoot errors

If you get error 429 from Azure OpenAI, it means the resource is over capacity:

  • Check the Activity Log of the Azure OpenAI service to see what else might be running.

  • Check the Tokens Per Minute (TPM) on the deployed model. On a system that isn't running other jobs, a TPM of 33K or higher should be sufficient to generate vectors for the sample data. You can try a model with more capacity if 429 errors persist.

  • Review these articles for information on rate limits: Understanding rate limits and A Guide to Azure OpenAI Service's Rate Limits and Monitoring.