This is an ever-evolving page where samples and content from the ML.NET community are highlighted, so anyone in the community can also take advantage of these additional samples.
However, note that Microsoft does not maintain the samples in the list below.
| Name | Description | ML Tasks or area of focus | API status | Owner |
|---|---|---|---|---|
| Photo-Search | Sample WPF app running an ONNX model which was previously built with Keras and exported to ONNX model format. | Deep Learning, Image classification | ML.NET v0.9 and .NET Core 3.0 | Kudos for Tak-Au |
| ML.NET Custom Transform | Show a custom transform implementation in ML.NET | Transform | Updated to ML.NET v0.9 | Kudos for endintiers |
| VB.NET samples for ML.NET | VB.NET samples, similar to some of the Getting Started samples, but implemented with VB.NET | Most in ML.NET | Updated to ML.NET v0.9 | Kudos for Nukepayload2 |
| ML.NET in visual desktop UWP app | Demonstrates how to use ML.NET to implement some Machine Learning use cases in UWP. Blog Post here | Multiple ML tasks | ML.NET v0.6 | Kudos for @diederikkrols |
| Multi-Output Regression with ML.NET & TensorFlow | Multi-Output Regression with ML.Net and TensorFlow | Regression with TF | Updated to ML.NET v0.9 | Kudos for zeahmed |
| DataFrame for ML.NET | It implements a subsample of pandas's dataframes API, on top of ML.NET | Data wrangling | TBD | Xavier Dupré xadupre [MSFT] |
| Beer-ML | Demo project uses Systembolaget database using four types of ML Tasks | Binary Classification, Multi-Class Classification, Regression, Clustering | Deprecated LearningPipeline API Needs to be updated to the "Dynamic API" | Alexander Dragunov adrag239 |
| Simple Linear Regression | Predicts salary based off years of experience. | Regression | Uses Static API. Needs to migrate to Dynamic API | Jon Wood, jwood803 |
| ONNX model scoring | Apache MXNet MLP model exported to ONNX and used in ML.NET | Regression | Cosmin Catalin, cosmincatalin |
Do you have any cool ML.NET sample that you'd like to share with the community? If so, add it to the following issue and eventually, it'll be showcased in this page.
Thanks for your contribution! :)