ML.NET

ML.NET, Machine Learning for .NET Developers

Cameron Vetter • June 17, 2019

Machine Learning is quickly becoming a ubiquitous technology and an expected skill of development teams. Python has dominated this space with all of the best tools as part of the Python stack. As a .NET developer, you need to not only understand the terminology, basic techniques but also learn a new language. ML.NET provides an alternative allowing you to do Machine Learning in C#.

In the image above (borrowed from the ML.NET home page) a simple logistic regression is performed using ML.NET. All of this is completed in C# without a single line of Python. In addition, there is full support for F# as well. Microsoft has provided some great examples on their samples page https://github.com/dotnet/machinelearning-samples. These examples include:

  • Binary Classification
  • Multi-class Classification
  • Regression
  • Clustering
  • Anomaly Detection
  • Computer Vision

In addition ML.NET has been designed to be an extensible platform and currently can consume other popular ML frameworks including:

  • TensorFlow
  • ONNX
  • Infer .NET

These examples make it look very easy, but the basic steps are still the same as they are with a Python framework:

  1. Data Collection
  2. Data Preparation and Transformation
  3. Model Creation
  4. Model Training
  5. Model Deployment

Similar to other ML tools, be prepared to spend a significant amount of your ML development working on the data; collecting it and transforming it into the right format.

These practical examples show how to solve some of the most common problems that developers are trying to solve with real world examples like these:

ML .Net is production ready today! What will you build?

Reach out to us by clicking here and we can get something going with you!

Share
back to blog