LearningModelBinding Class

Definition

public ref class LearningModelBinding sealed : IIterable<IKeyValuePair<Platform::String ^, Platform::Object ^> ^>, IMapView<Platform::String ^, Platform::Object ^>

/// [Windows.Foundation.Metadata.Activatable(Windows.AI.MachineLearning.ILearningModelBindingFactory, 65536, "Windows.AI.MachineLearning.MachineLearningContract")]
/// [Windows.Foundation.Metadata.ContractVersion(Windows.AI.MachineLearning.MachineLearningContract, 65536)]
/// [Windows.Foundation.Metadata.MarshalingBehavior(Windows.Foundation.Metadata.MarshalingType.Agile)]
/// [Windows.Foundation.Metadata.Threading(Windows.Foundation.Metadata.ThreadingModel.Both)]
class LearningModelBinding final : IIterable<IKeyValuePair<winrt::hstring, IInspectable const&>>, IMapView<winrt::hstring, IInspectable const&>

[Windows.Foundation.Metadata.Activatable(typeof(Windows.AI.MachineLearning.ILearningModelBindingFactory), 65536, "Windows.AI.MachineLearning.MachineLearningContract")]
[Windows.Foundation.Metadata.ContractVersion(typeof(Windows.AI.MachineLearning.MachineLearningContract), 65536)]
[Windows.Foundation.Metadata.MarshalingBehavior(Windows.Foundation.Metadata.MarshalingType.Agile)]
[Windows.Foundation.Metadata.Threading(Windows.Foundation.Metadata.ThreadingModel.Both)]
public sealed class LearningModelBinding : IEnumerable<KeyValuePair<string,object>>, IReadOnlyDictionary<string,object>

function LearningModelBinding(session)

Public NotInheritable Class LearningModelBinding
Implements IEnumerable(Of KeyValuePair(Of String, Object)), IReadOnlyDictionary(Of String, Object)

Inheritance

Object Platform::Object IInspectable LearningModelBinding

Attributes

ActivatableAttribute ContractVersionAttribute MarshalingBehaviorAttribute ThreadingAttribute

Implements

IIterable<IKeyValuePair<K,V>> IEnumerable<KeyValuePair<K,V>> IIterable<IKeyValuePair<String,Object>> IEnumerable<KeyValuePair<String,Object>> IIterable<IKeyValuePair<Platform::String,Platform::Object>> IIterable<IKeyValuePair<winrt::hstring,IInspectable>> IMapView<String,Object> IReadOnlyDictionary<String,Object> IMapView<Platform::String,Platform::Object> IMapView<winrt::hstring,IInspectable>

Windows requirements

Examples

The following example retrieves the first input and output features, creates an output frame, binds the input and output features, and evaulates.

private async Task EvaluateModelAsync(
    VideoFrame _inputFrame, 
    LearningModelSession _session, 
    IReadOnlyList<ILearningModelFeatureDescriptor> _inputFeatures, 
    IReadOnlyList<ILearningModelFeatureDescriptor> _outputFeatures,
    LearningModel _model)
{
    ImageFeatureDescriptor _inputImageDescription;
    TensorFeatureDescriptor _outputImageDescription;
    LearningModelBinding _binding = null;
    VideoFrame _outputFrame = null;

    try
    {
        // Retrieve the first input feature which is an image
        _inputImageDescription =
                        _inputFeatures.FirstOrDefault(feature => feature.Kind == LearningModelFeatureKind.Image)
                        as ImageFeatureDescriptor;

        // Retrieve the first output feature which is a tensor
        _outputImageDescription =
                        _outputFeatures.FirstOrDefault(feature => feature.Kind == LearningModelFeatureKind.Tensor)
                        as TensorFeatureDescriptor;

        // Create output frame based on expected image width and height
        _outputFrame = new VideoFrame(
            BitmapPixelFormat.Bgra8, 
            (int)_inputImageDescription.Width, 
            (int)_inputImageDescription.Height);

        // Create binding and then bind input/output features
        _binding = new LearningModelBinding(_session);

        _binding.Bind(_inputImageDescription.Name, _inputFrame);
        _binding.Bind(_outputImageDescription.Name, _outputFrame);

        // Evaluate and get the results
        var results = await _session.EvaluateAsync(_binding, "test");
    }
    catch (Exception ex)
    {
        StatusBlock.Text = $"error: {ex.Message}";
        _model = null;
    }
}

Remarks

Models specify their input and output features using unique string names. Before evaluating the model you can bind your inputs and outputs to the session using those names. To do this you use the LearningModelBinding object which you can create based on a session.

Windows Server

To use this API on Windows Server, you must use Windows Server 2019 with Desktop Experience.

Thread safety

This API is thread-safe.

See also

• Windows ML
• Windows ML samples (GitHub)