Microsoft is adding a new AI platform to the next major Windows 10 upgrade. Windows ML will let developers use pre-trained machine learning models in their Apps on Windows devices.

Windows ML is described as evaluating trained machine  learning models locally on Windows 10 devices, allowing developers to use pre-trained models within their applications. The platform provides hardware-accelerated performance by making use of the device's CPU or GPU to compute evaluations for both classical Machine Learning algorithms and Deep Learning.

The GPU will be used on devices with DirectX12 capability. CPU optimizations will also be used to enable high-performance evaluation of both classical ML and Deep Learning algorithms. If the model includes the use of  image, video, and camera data, Windows ML handles frame pre-processing and provides the camera pipeline setup for model input.

The models that Windows ML can run have to be pre-trained in the Open Neural Network Exchange (ONNX) format. However, Windows ML supports the v1.0 release of the ONNX format, which allows you to use models produced by different training frameworks so long as they've been passed through a converter tool.

You can also train your own ONNX models to use with Windows ML. This is possible using the Azure Machine Learning Workbench, and the Azure Custom Vision Service will soon support creating ONNX models for Windows. For developers using Visual Studio, it will be possible from Visual Studio Preview 15.7 to add an ONNX file to a UWP project and automatically generate a model interface in the project.

Developers using older versions of Visual Studio can use the MLGen tool to generate the code interface and then manually add it to their projects. This capability will be coming soon to Visual Studio tools for AI as well.

Windows ML supports 100+ ONNX operators on the CPU and accelerates computation on DirectX12 compatible GPUs. Given an ONNX model file, Windows ML's code generator can generate wrapper classes that call the Windows ML API for you, providing an interface to interact with the model in your app. The generated classes represent the model, inputs, and outputs, allowing you to load, bind, and evaluate the model in your project.

The plan moving forward is that Windows ML will add support for AI-specific processors, such as Intel's Movidius VPU, in the future.

More Information

Windows Machine Learning

ONNX

Visual Studio Code Tools for AI

Azure Machine Learning Workbench

Related Articles

Microsoft Q# Quantum Language And Toolkit 

More AI Tools From Microsoft

Azure Machine Learning Enhancements

Microsoft Cognitive Toolkit Version 2.0

Microsoft Expands Cognitive Services APIs

To be informed about new articles on I Programmer, sign up for our weekly newsletter, subscribe to the RSS feed and follow us on Twitter, Facebook or Linkedin.

Comments

Make a Comment or View Existing Comments Using Disqus

or email your comment to: comments@i-programmer.info