Window Machine Learning is a high-performance, reliable API for deploying hardware-accelerated ML inferences on Windows devices. Please visit the Windows ML documentation to learn more about Windows ML.
We design and optimize for all Windows devices.
Our goal is to provide developers with a platform that enables new experiences that run well on all Windows devices. Our design drives innovation in the DirectX ecosystem through DirectML and gives developers the confidence that their applications will work for all Windows customers.
We maintain and curate the Windows ML APIs.
The API is designed to ensure consistency of developer’s experience across the Windows platform. We provide long-term servicing and support, and we are committed to ensuring application’s compatibility as we evolve the API.
Windows ML is a core component of Windows.
The Windows ML code is packaged and distributed with each new release of Windows. To provide consumers with high-quality products, Microsoft is responsible for distributing Windows ML and related binaries as part of Windows or standalone distributable packages.
We encourage community contributions to Windows ML to enhance users’ experience on Windows. We use the principles above to guide how we look at and evaluate all contributions.
Ensure your feature request follows all these principles to help the review process and include information about the customer problem(s) the feature request addresses.
Note: minor issues or bugs can be addressed more quickly using the bug/performance issue request rather than feature request.
If you'd like to contribute to Windows ML and engage with the community to get feedback, please review to the contributing process details and submit a new feature request here.