This hands-on workshop will teach you how to create traditional non-neural machine learning prediction systems. You will learn what six fundamental ML techniques are and what types of problems they can (and cannot) solve, implementing them using only raw C# without any external libraries.
This workshop emphasizes practical techniques, but also gives you just enough theory so you can modify and customize your systems.
Techniques covered include: logistic regression binary classification, naive Bayes multiclass classification, k-nearest neighbors classification, Thompson Sampling and epsilon-greedy bandit algorithms, and k-means clustering.
You will learn:
- What types of problems machine learning can and cannot solve
- How to implement the six fundamental ML techniques using C#
- How to interpret the results of an machine learning model
Requirements for the "Implementing Machine Learning Using C# and Visual Studio" Lab at Vegas:
- A Windows 10 laptop.
- Access to a user account with full Administrator privileges.
- An accessible USB drive port.
- Basic familiarity with C# development using Visual Studio.
- Visual Studio 2019 (any edition will work but the Lab is based on the Community (free) edition of VS 2019) with the ".NET Core cross-platform development" workload installed. Click image below:
Note: This Lab doesn't assume you know anything about machine learning -- absolutely no ML background is assumed.
Note: All demos are intended to run on .NET Core but they will, in theory, run under the older .NET Framework (".NET Desktop Development" Workload), however, there will be no technical support for .NET Framework if you use it and something goes wrong.
Note: The Visual Studio Code tool will not be supported in the Lab.
Note: Mac machines running a Windows 10 emulator will not be supported.