Introduction to Azure Machine Learning

Democratizing Data Science

Azure Machine Learning (AML) is a cloud service for accelerating and managing the machine learning project lifecycle. So once the task is defined, it allows users to:

• Explore and prepare your dataset
• Develop, train and validate your model
• Deploy models
• Monitor and manage models and dataset lifecycle with MLops

One of the greatest features of AML is its versatility in terms of tools to develop models. Indeed, it is not compulsory to be able to code in Python or other languages to develop ML models, it might not even be necessary to have data science skills.

Indeed, AML democratizes the use of Machine Learning techniques via a set of tools encountering a wide range of skills. Those tools are:

• Notebook
• Designer
• Automated Machine Learning (AutoML)

Now let’s go deeper into each of the steps of a model deployment with some examples, using the AML UI. To access the UI, you need to create an AML resource via the Azure Portal (instruction here). Once initialized the resource, you can jump directly into the portal and start working on your ML projects.