Data Scientist and author Siraj Raval recently released a 12-minute video overview of Azure Machine Learning (embedded at the end of this post). The video begins with a overview of cloud computing and Microsoft Azure generally, before getting into the details of some specific Azure services for machine learning:
- [at 2:00] Hybrid cloud / on-premises architectures (see for example SQL Server and Azure Stack)
- [at 3:40] Overview of Azure Machine Learning
- [at 4:25] The MML Spark toolkit
- [at 5:25] Project Brainwave, inference with FPGAs (field-programmable gate array chips)
- [at 6:05] The Azure Machine Learning Workbench
- [at 7:15] Visual Studio Tools for AI (also available for open-source Visual Studio Code)
- [at 7:30] Azure Machine Learning Studio, including a demo of using the drag-and-drop UI to predict car prices.
For those of you familiar with similar services from AWS, or just generally baffled by some of the Microsoft service names (and if so you're not alone there — believe me!), Siraj has also produced a plain English codex for Microsoft Azure services, with a description, corresponding service name in other systems, and what the service maybe should have been called in the first place.
Take a look at Siraj's video, below: