Deploying a Machine Learning Model with Azure ML Pipelines

Machine learning pipelines are a way to describe your machine learning process as a series of steps such as data extraction and pre-processing, but also training, deploying, and running models.

In this article, I’ll show you how you can use Azure ML Pipelines to deploy an already trained model such as this one, and use it to generate batch predictions multiple times a day. But before we do that, let’s understand why pipelines are so important in machine learning.

Clean Code with Machine Learning Pipelines

Thinking about the model we’ve trained to predict house prices, we can easily imagine a scenario in which it only needs to run at certain times on a batch of new houses, generate price predictions, and store said predictions for easy access.

The good news is that all of this can be done in a single script containing everything and the kitchen sink, however this approach will not scale particularly well. This is because whenever you’ll need to add more functionality to the script (i.e. by retrieving more features from external APIs), it will grow and grow, acquiring more and more responsibilities, until it becomes a something akin to a big ball of mud.

Read more on Vlad's blog, Head of AI at Strongbytes: https://vladiliescu.net/deploying-models-with-azure-ml-pipelines/