November 20, 2018 8:15 am

Getting Started with Windows Machine Learning

By / Program Manager, Partners, Analytics, and Essential Experiences Team

With the Windows 10 October 2018 Update, Windows Machine Learning is out of preview and ready to become part of your AI toolbox. The Windows ML inference engine evaluates trained models locally on Windows devices, removing concerns of connectivity, bandwidth, and data privacy. This video series will help you understand how you can get started with Windows ML and explore the new APIs through the lens of Rufus the robot.

Overview of Windows Machine Learning

What is Windows Machine Learning and why should you care? Killian and Rosane walk through these questions and clarify the positioning and capabilities of this powerful AI inference engine.

Windows Machine Learning: Hello World (MNIST Edition)

Let’s jump to the code! Killian and Rosane work through the Windows ML MNIST tutorial and answer questions developers commonly ask along the way.

Windows Machine Learning: Models and Features

Killian and Rosane (with guest star Rufus!) discuss loading models and understanding how to determine a given model’s expected inputs and output features.

Windows Machine Learning: Device Selection

Killian and Paul (and our robot friend Rufus!) go over how to select a device to run Windows ML model evaluation and uncover what the default option really means.

Windows Machine Learning: Sessions and Bindings

Killian and Paul (and ROS-powered Rufus!) are ready to show you how to evaluate a model with Windows ML, but not before they cover the importance of sessions and bindings to get there.

Providing Feedback & Staying Informed

What Windows ML topics should we cover next? Let us know what you think by tweeting (and following) @killianqueue!

Join the conversation

  1. Can’t find the Windows Machine Learning Github repo.
    A link please, so that I can find the sample project? I am just starting and this would be great.