/custom-vision-azure-functions

Source code for the learning series "Serverless image classification with Azure Functions and Custom Vision". Published on my blog.

Primary LanguagePythonMIT LicenseMIT

Serverless image classification with Azure Functions and Custom Vision

Welcome to the new series focused on Azure Custom Vision and Azure Functions! In the "Serverless image classification with Azure Functions and Custom Vision" series, you are going to train a TensorFlow image classification model in Azure Custom Vision and run the model in an Azure Function.

You will learn how to build a serverless HTTP API with Azure Functions and Azure Custom Vision. By the end o this series you will be able to:

  • Train and publish an image classifier using the Custom Vision SDK for Python.
  • Export a TensorFlow model using the Python SDK and run the model locally to classify images.
  • Import a TensorFlow machine learning model into an Azure Function.
  • Build a serverless HTTP API for classifying images.

Part 1

Tutorial Tutorial Source code

In the first article of this series, you will build and deploy an image classification model using the Custom Vision SDK for Python.


Part 2

Tutorial Tutorial Source code

This article will show you how to export your classifier using the Custom Vision SDK for Python and use the model to classify images.


Part 3

Tutorial Tutorial Source code

In this article, you will create a Python Azure Function with HTTP trigger to consume a TensorFlow machine learning model.


Part 4

Tutorial Tutorial Source code

In this article, you will deploy a function project to Azure using Visual Studio Code to create a serverless HTTP API.