Machine Learning – ML.NET: The Machine Learning Framework for .NET Developers (MSDN Magazine)
The ML.NET library is a new open source collection of machine learning (ML) code that can be used to create powerful prediction systems. Many ML libraries are written in C++ with a Python API for easier programming. Examples include scikit-learn, TensorFlow, CNTK and PyTorch. However, if you use a Python-based ML library to create a prediction model, it’s not so easy for a .NET application to use the trained model. Fortunately, the ML.NET library can be used directly in .NET applications. And because ML.NET can run on .NET Core, you can create predictive systems for macOS and Linux, too.