MyML.NetApp is a sample ML.NET application that demonstrates how to train, evaluate, and consume a machine learning model using .NET 9.0.

• MyML.NetApp
  • Table of Contents
  • Overview
  • Project Structure
  • Setup
    • Prerequisites
    • Build
    • First Run
  • Usage
    • Train and Evaluate Model
    • Consume Model
  • Testing
  • Localization
  • Troubleshooting
  • Changelog
  • Contributing
    • Submitting Issues
    • Code Style
  • License
  • Contact

This project shows how to use ML.NET to build, train, evaluate, and consume a machine learning model. It uses a sample dataset (wikipedia-detox-250-line-data.tsv) for demonstration purposes. All model-building and utility logic is centralized in a shared library for maintainability and testability.

• MyML.NetApp/ - Main application for consuming the trained model
• MyML.NetAppML.ConsoleApp/ - Console app for making predictions using the trained model
• MyML.NetAppML.Model/ - Model input/output classes and model consumption logic
• MyML.NetAppML.Lib/ - Shared library containing all model-building, training, and utility logic
• MyML.NetAppML.Tests/ - xUnit test project for all core logic
• Data/ - Sample data files

• .NET 9.0 SDK
• Visual Studio or VS Code

dotnet build MyML.NetApp.sln

1. Clone this repository:

   git clone https://github.com/yourusername/MyML.NetApp.git
   cd MyML.NetApp

2. Restore dependencies:

   dotnet restore

3. Build the solution:

   dotnet build MyML.NetApp.sln

4. (Optional) Train a new model by calling ModelBuilder.CreateModel() from your own script or app.
5. Ensure MLModel.zip exists in MyML.NetAppML.Model (retrain if needed).

• See the shared library's ModelBuilder.CreateModel() for training logic.
• The trained model is saved as MLModel.zip in the MyML.NetAppML.Model directory.

• Run the console app to make predictions using the trained model:

  dotnet run --project MyML.NetAppML.ConsoleApp

• Or run the main application:

  dotnet run --project MyML.NetApp

Run all unit tests with:

dotnet test MyML.NetAppML.Tests/MyML.NetAppML.Tests.csproj

All core logic is covered by xUnit tests, including model input/output, prediction, and utility methods.

All user-facing strings are managed via .resx resource files for easy localization and code quality compliance.

• Model file not found: Ensure MLModel.zip exists in the MyML.NetAppML.Model directory. Retrain if needed.
• ML.NET version issues: All projects target .NET 9.0 and use ML.NET 1.5.0. Ensure your environment matches.
• Test failures or hangs: Ensure your test data includes at least two distinct class labels for multiclass classification.

See CHANGELOG.md for a detailed list of changes and release history.

We welcome contributions from the community! To get started:

1. Fork this repository and clone your fork locally.
2. Create a new branch for your feature or bugfix:

   git checkout -b my-feature-branch

3. Make your changes and add tests as appropriate.
4. Run the test suite to ensure nothing is broken:

   dotnet test MyML.NetAppML.Tests/MyML.NetAppML.Tests.csproj

5. Commit and push your changes to your fork.
6. Open a Pull Request (PR) against the master branch with a clear description of your changes.

• If you find a bug or have a feature request, please open an issue with as much detail as possible.
• Include steps to reproduce, expected behavior, and screenshots/logs if relevant.

• Follow the existing code style and conventions enforced by .editorconfig and analyzers.
• Write clear, concise commit messages and PR descriptions.

This project is licensed under the MIT License.

For questions or feedback, please contact the project maintainer:

• Neil Mascarenhas