At Build, Microsoft released ML.Net 1.0 which aims to bring the power of machine learning to .Net developers without having to learn Python or become a full-fledged data scientist. This runs in C# or F# code, not on a separate service in Azure.
In this session, we will create our own custom machine learning model using these components:
- ML.Net - framework to train, evaluate, and consume machine learning models
- AutoML - library that tries many combinations of algorithms and hyperparameters to find a good fit for the given data and task
- ML.Net Model Builder - Visual Studio extension to generate the code to do the above
Our speaker for tonight is Darrell Speck, a senior developer at Proliant and a longtime developer on the Microsoft stack, starting with VB3. An avid Go player, he was amazed by the sudden world-class strength of AlphaGo, the Go playing AI from Google's DeepMind.