ML.NET 2.0 is an open-source, cross-platform machine learning framework for .NET developers that enables integration of custom machine learning models into .NET apps. With this new release, ML.NET Text Classification API has been refined, enabling you to train custom text classification models using the latest deep learning techniques from Microsoft Research inside of Model Builder. Additionally, ML.NET 2.0 introduces a new API for sentence similarity that uses the same underlying TorchSharp NAS-BERT model as the Text Classification API. To get started with these new features, install or upgrade to the latest versions of the ML.NET 2.0.0 and 0.20.0 packages as well as Model Builder 16.4.0 or later. For more details on setting up your GPU, see the ML.NET GPU guide.
