Windows ML

Windows Machine Learning (ML) allows you to use trained machine learning models in your applications. The platform evaluates trained models locally on Windows 10 devices, providing hardware-accelerated performance by leveraging the device's CPU or GPU, and computes evaluations for both classical ML algorithms and Deep Learning.

Windows machine learning

In this section

Topic Description
Overview Learn about Windows ML and how to use trained models in your app.
Get started Create your first Windows ML app with this step-by-step tutorial.
Train a model Train a model for Windows ML using Visual Studio Tools for AI.
Convert a model Convert existing models to ONNX format to use with Windows ML.
Integrate a model Integrate a model into your app by following the load, bind, and evaluate pattern.
Sample apps See sample UWP apps that demonstrate how to use Windows ML.