Microsoft's official ML sampler platter for .NET devs

What it does This repository is Microsoft’s official sample collection for ML.NET, housing two tiers of examples. “Getting Started” samples are stripped-down console apps that isolate specific ML tasks—sentiment analysis, fraud detection, demand prediction—while “End-to-End” apps demonstrate how to embed trained models into real web APIs and desktop software. Most tasks are implemented in both C# and F#.

The interesting bit The repo functions like a Rosetta stone for .NET developers, mapping classic ML problems onto familiar patterns. It often shows the same scenario through different lenses—image classification via high-level ML.NET APIs, TensorFlow featurizers, and ONNX scoring—so you can compare integration strategies without leaving the ecosystem.

Key highlights

• Covers the standard ML spectrum: binary and multi-class classification, regression, recommendation, time series, anomaly detection, clustering, ranking, and computer vision
• Most getting-started samples ship in both C# and F#
• Includes deep-learning bridges: TensorFlow model scoring and ONNX object detection
• Distinguishes between minimal console demos and production-ish web/desktop applications
• Cross-cutting scenarios address deployment patterns like scalable WebAPI integration

Caveats

• This is strictly a samples repo; framework bugs belong in the main ML.NET repository
• The README is essentially a large table of contents, so expect to dig into individual project folders for implementation details

Verdict Useful if you’re a .NET developer trying to map ML concepts onto familiar tooling. Skip it if you want a single polished reference architecture or already live in Python.