Machine Learning and AI

A video frame showing Ankit Asthana presenting. Select this image to start playing the video.

Infuse Machine Learning and AI in your .NET apps

.NET is already equipped with everything you need to build smarter apps by infusing Machine Learning and AI both for on-device and cloud scenarios.

You can use pre-built models with Cognitive Services, Core ML for Xamarin or generate and consume your owns models built with Azure Machine Learning, deep learning libraries like CNTK, Tensorflow and Accord.NET. The following sections cover these different technologies in detail.

You can get started with the popular MNIST for ML beginners model (Helloworld for Machine Learning) code example built in C#, using CNTK and Tensors by clicking the button below.

Explore MNIST for ML Beginners

An illustration of an eye, magnifying glass, text input, microphone, and speaker.

Cognitive Services

Easily add intelligent features to your .NET apps—such as emotion and sentiment detection, vision and speech recognition, language understanding, knowledge, and intelligent search.

Try the Intelligent Kiosk sample app. It illustrates how Cognitive Services can be incorporated in a .NET app.

Explore Cognitive Services

An illustration of a person emptying data from a cloud into a complex machine.

Azure Machine Learning

Azure Machine Learning is a fully-managed cloud service that enables you to easily build, deploy, and share predictive analytics solutions.

The Recommendation Engine sample app shows Azure Machine Learning being used in a .NET app.

Explore Azure Machine Learning

An illustration of books next to a beaker containing green fluid.

Machine Learning and AI Libraries

The Microsoft Cognitive Toolkit is a free, open source and commercial-grade toolkit that allows you to easily train deep learning algorithms.

The Microsoft Cognitive Toolkit Examples demonstrate how to use the toolkit in your .NET apps.

Explore Microsoft Cognitive Toolkit

For classic machine learning, you can also use Accord.NET, which comes with audio and image processing libraries completely written in C#.

An illustration of a person working on a laptop. Wires connect the laptop to a beaker, test tubes, and a bar graph.

F# for Data Science and ML

F# (pronounced F sharp) is a functional programming language that runs on .NET and supports object-oriented programming.

F#'s type system and unique features such as Type Providers and Units of Measure are a great fit for data science and ML.

Get started with F#

Ready to Get Started?

Our step-by-step tutorial will help you get .NET running on your computer.

Supported on Windows, Linux, and macOS

Get Started