Solving 3 business problems with ML.NET in 45 minutes
summary
Shows concrete use of ML.NET to build three production-ready pipelines: a regression model for demand/sales prediction (trained with normalized features, evaluated by R² and loss), a recommendation system using matrix factorization whose weekly CI job precomputes and stores the top‑20 pairwise product recommendations in a database, and customer grouping via k-means clustering for targeted offers. Demonstrates training vs cheap inference, hyperparameter tuning (alpha), AutoML/Visual Studio export, and a real client deployment that worked.
AI-written summary · about
chapters (16)
- 0:00Start
- 4:22Ai Definition
- 6:53Types of Algorithms in Machine Learning
- 9:10Unsupervised Learning
- 9:58Clusterization
- 11:52Linear Function
- 14:26Cost Function
- 16:28Ml Net
- 18:27The Model of Working with Ml Net
- 21:09Load from Text File
- 21:23Prediction Model
- 25:57Normalization
- 28:39Product Recommendation
- 29:28Matrix Factorization
- 31:08Hyper Parameters
- 35:02Demand Prediction
related talks
39mSolving 3 business problems in 45 minutes using ML.NET
01h 03mMachine learning for C# developers: Introducing ML
50mA Pragmatic Deep Dive into ML.NET
45mMachine Learning for .NET developers
52mML.NET - Machine Learning made for .NET
57mMachine Learning Made Easy – An Introduction to ML.NET
59mMachine Learning in .NET (ML.NET)
47mBuilding ML practices to address the top four use cases