Learn how to build machine learning applications in C# with Microsoft’s new ML.NET library, using feature engineering, regression and classification
This training course will introduce you to Microsoft’s MLNET machine learning library. You’ll gain a solid understanding of machine learning and artificial intelligence, including key concepts such as feature engineering, regression and classification.
As you progress through the course material, you’ll design, train, and evaluate sophisticated machine learning models on your computer using C# and ML.NET.
I'll provide you with all required datasets, source code, and libraries to help you get started and build your own machine learning applications with confidence.
The training covers the following topics:
Artificial Intelligence
Machine Learning
AI Bias
AI Risks
AI Privacy & Security
AI Transparency
AI Accountability
Principles of Responsible AI
Supervised Machine Learning
Loading Numeric Data
Data Normalization
Processing Outliers
Processing Missing Values
Feature Histograms
The Pearson Correlation Matrix
Bonferonni Thresholds
Loading Text Data
One-Hot Encoding
Sparse Vector Encoding
Binning Data
Vector Cross-Products
Single Linear Regression
Gradient Descent
Multiple Linear Regression
ML.NET Training Pipelines
Evaluating Models
Loading & Saving Models
Root Mean Square Error
Mean Square Error
Mean Absolute Error
Binary Classification
Logistic Regression
Platt Calibration
Accuracy
Precision
Recall
The ROC Curve
Area Under Curve
Multiclass Classification
The Confusion Matrix
Micro Average Accuracy
Macro Average Accuracy
What You’ll Get Inside This Course
This course includes everything you need to build machine learning applications with C# and ML.NET, including video lectures to master the core concepts and guided labs that use real datasets to create production-ready code.
84 Course Lectures
118 Lab Lessons
5 Knowledge Quizzes
Yes, Python leads the AI world. But that doesn’t mean it’s the right choice for you.
If you're a .NET developer working for a business organization, your entire code base will have already been written in C#. Bolting a Python machine learning app wrapped in a Flask service API on top of C# code is very brittle and it completely undercuts the type safety and performance provided by the C# compiler.
In terms of systems architecture, it's much better to host your machine learning model directly in C# and never leave the .NET ecosystem at all. Why switch stacks if you don't have to?
You can stay in your preferred environment, train custom models and deploy them directly, without having to rewrite your backend, bolting on Python, or learning a whole new tooling ecosystem.
C# just makes sense for .NET developers:
No context-switching, stay productive in your native stack
Streamlined deployment: integrate ML directly into your production apps
Robust architecture: no brittle network links between your app and the ML model
Type safety and performance: strong typing and compiled performance across the board
Enterprise-friendly: your code is much easier to maintain and audit
The Microsoft ML.NET library is an open source and cross-platform machine learning framework for the .NET ecosystem. With ML.NET, you can easily create custom machine learning apps using C# or F#.
ML.NET lets you re-use all the knowledge, skills, code, and libraries you already have as a .NET developer so that you can easily integrate machine learning into your web, mobile, desktop, games, and IoT apps.
ML.NET also offers Model Builder, a simple UI tool for training regression and classification models on a dataset, and ML.NET CLI, a command line tool that makes it super easy to build custom models.
Both these tools use AutoML, a cutting edge technology that automates the process of building best performing models for your machine learning scenario. All you have to do is load your data, and AutoML takes care of the rest of the model building process.
ML.NET has been designed as an extensible platform so that you can easily consume other popular ML frameworks like TensorFlow, ONNX, Infer.NET and more, and have access to even more machine learning scenarios, like image classification and object detection.
Check out this sample lab lesson from the course, in which I show you how to calculate the Pearson correlation matrix from the features in the California Housing dataset.
This preview covers:
How to calculate the Pearson correlation matrix with the MathNet.Numerics library
Printing the matrix to the console with BetterConsoleTables
Plotting the matrix as a heatmap with ScottPlot
This is just one of the 118 lab lessons available in the full course.
⭐️⭐️⭐️⭐️⭐️ Mark is a superb teacher! I had the chance to attend his C# courses before and they are excellent, and I'm happy to see that his Machine Learning teaching skills are as great as his C# skills. The Machine Learning course is mind-blowing.
⭐️⭐️⭐️⭐️⭐️ Mark's Machine Learning course is the unique source of knowledge about creating Machine Learning solutions using .NET and C#. It shows a wide picture of the whole technology and is a great starting point into the amazing world of AI.
Watch this sample lesson from the course, where I break down one of the most common metrics in machine learning: the accuracy.
This preview covers:
How accuracy can be unreliable when the dataset is biased
The meaning of false positives and false negatives
How to interpret the confusion matrix in real-world scenarios
This is just one of the 84 lessons in this course.
To follow along with the course and complete the labs, you’ll need a few tools installed. It’s all free, cross-platform, and easy to set up.
Requirements:
A computer running Windows, macOS, or Linux (no GPU required)
Visual Studio Code: the free, lightweight code editor from Microsoft
.NET SDK: the runtime and build tools for C#
ML.NET: Microsoft’s machine learning library for .NET
I will walk you through the setup of each tool in the course.
The lab modules you'll complete in this course will guide you through building real C# applications that train machine learning models on well-known public datasets from tech, healthcare, and other domains.
You'll be working with:
This real-world housing dataset from Google contains census data from 17,000 housing blocks across the state of California
This well-known dataset contains trip details of all 8.1 million taxi trips conducted in New York City in December 2018.
This dataset contains clinical data of 303 real-life patients who may or may not be suffering from cardiovascular disease.
The MNIST dataset contains 60,000 images of handwritten numbers and can be used to train handwriting recognition systems
Choose the plan that works best for you. Buy this course and dive into MLNET machine learning, or unlock unlimited access to every course on the site.
Want the best learning experience?
Members get access to the full course library, all labs and community pages, receive priority support and sneak previews of future course releases.
Prices shown exclude VAT. EU businesses can defer VAT during checkout with a valid VAT ID number.
Buy this course and get lifetime access to all lectures and knowledge quizzes
✔️ This course
✔️ Included quizzes only
✔️ Included labs only
🚫 No priority support
🚫 No community access
🚫 No future courses
€95 one time
Get access to all courses, lectures, labs, quizzes, and future releases
✔️ All courses
✔️ All quizzes
✔️ All labs
✔️ Priority support
✔️ Community pages
✔️ Access to future courses
€35 p/mo or €350 p/yr
In-company or online team training with guided labs and live support
🪙 Onsite or online
🪙 Conducted live
🪙 For teams of 5-12
🪙 3-day training
🪙 Guided labs
🪙 Can be customized
€1250 p/day
Live Classroom Delivery of Supervised Machine Learning with ML.NET and C#
In 2020, I was invited to Budapest by GLC Europe to deliver a 3-day, in-person version of this very course. The event marked the first time a European training agency offered a training course on Microsoft’s new ML.NET library.
At the time, I had been working closely with Cesar De La Torre, Microsoft’s AI Program Manager, to develop this content and ensure it aligned with Microsoft's vision of the ML.NET library.
This photo was taken right after I finished setting up my laptop and training materials, and just before the students arrived. Over the following three days, my students learned the theory of machine learning and worked through the labs, using the same content structure that now powers the online version you're about to join.
You're almost ready to dive in! If there's anything you're unsure about, take a moment to explore the answers below. I’ve covered the most common questions about this course, the platform, and what you can expect. But if you still need help, don’t hesitate to send me a quick message and I’ll be happy to help you.
You get lifetime access to the full training: all lectures, knowledge quizzes and your membership in the online discussion forums.
Yes! You should be comfortable with basic AI concepts, but no prior experience with machine learning is required.
You'll get access to a private discussion forum for this course where you can ask questions and share insights. Members also get priority support and access to a private community space.
Absolutely. This course is part of a growing library. With the All-Access Membership, you get everything—including future releases—and a curated gold learning path to guide your progress.
Forever. When you buy this course, it’s yours for life, including any future updates.
Yes. Any improvements or updates to this course will be included automatically at no extra charge.
Yes. Team and enterprise licenses are available. Contact me here for group pricing and custom onboarding options.
Sign up for the newsletter and get notified when I publish new posts and articles online.
Let's stay in touch!