Summary

In this chapter, we covered training, evaluating, and exporting a multi-class classification model designed to classify data into one of three or more different categories.

We saw how ML.NET supports saving and loading its models to .zip files and touched on how ML.NET also supports ONNX files to save some of its models and import pre-trained models from other technologies. Further, we were able to take our multi-class classification model and add it to an ASP .NET Web API. We did this with only a few steps by relying on the thread-safe PredictionEnginePool class from the Microsoft.Extensions.ML NuGet package. Finally, we touched on the many other machine learning tasks that ML.NET is capable of handling.

In the first part of this book, we saw how Polyglot Notebooks can serve as an interactive tool for analyzing, visualizing, and manipulating data.

In this part of the book, we discovered how Polyglot Notebooks pairs well with ML.NET and its machine learning capabilities...