Introduction

Completed

As a data scientist, you want to write code that works in any development environment. Whether you're using local or cloud compute, the code should successfully execute to train a machine learning model for example.

To run code, you need to ensure necessary packages, libraries, and dependencies are installed on the compute you use to run the code. In Azure Machine Learning, environments list and store the necessary packages that you can reuse across compute targets.

Note

In this module, we refer to Azure Machine Learning's interpretation of environments. Note that the term environments is also used to describe other technical concepts. For example, in DevOps, environments refer to the collection of resources used for a specific phase in the application deployment, like the development or production environment. Learn more about continuous deployment for machine learning.