PACKAGE

About

ML.NET is a cross-platform open-source machine learning framework which makes machine learning accessible to .NET developers.

For more information, see the ML.NET documentation.

This package adds the ability to load and run pre-trained ONNX models as part of your ML.NET pipeline.

Key Features

• Any type of ONNX model can be run with all the data interop handled automatically.
• Image Featurization - Automatically featurize an image using a pre-trained DNN model.

How to Use

See Machine Learning Samples for an assortment of samples that show how to get started using ML.NET.

Main Types

Some of the types provided by this library are:

• Microsoft.ML.Transforms.Onnx.OnnxOptions
• Microsoft.ML.Transforms.Onnx.OnnxScoringEstimator
• Microsoft.ML.Transforms.Onnx.OnnxTransformer

Additional Documentation

• Conceptual documentation
• API documentation

Related Packages

• Main Package: Microsoft.ML
• Core data abstraction: Microsoft.ML.DataView
• LightGBM Model Support: Microsoft.ML.LightGbm
• Fast Tree: Microsoft.ML.FastTree
• Image analytics: Microsoft.ML.ImageAnalytics
• Recommender: Microsoft.ML.Recommender
• Time series: Microsoft.ML.TimeSeries
• Automatic model selection / tuning: Microsoft.ML.AutoML
• Exporting Onnx Models: Microsoft.ML.OnnxConverter
• Text tokenizers: Microsoft.ML.Tokenizers

Feedback & Contributing

Microsoft.ML is released as open source under the MIT license. Bug reports and contributions are welcome at the GitHub repository.