LearningModelSessionOptions.OverrideNamedDimension(String, UInt32) Method
Definition
Important
Some information relates to prerelease product that may be substantially modified before it’s released. Microsoft makes no warranties, express or implied, with respect to the information provided here.
Important
For the latest documentation about Windows Machine Learning, see What is Windows ML. That documentation describes APIs that are in the Microsoft.Windows.AI.MachineLearning namespace, which ships in the Windows App SDK. Those APIs supersede the ones documented here, which are in the Windows.AI.MachineLearning namespace, and were shipped in 2018.
Provides the ability to override named input dimensions of a Learning Model to concrete values.
public:
virtual void OverrideNamedDimension(Platform::String ^ name, unsigned int dimension) = OverrideNamedDimension;void OverrideNamedDimension(winrt::hstring const& name, uint32_t const& dimension);public void OverrideNamedDimension(string name, uint dimension);function overrideNamedDimension(name, dimension)Public Sub OverrideNamedDimension (name As String, dimension As UInteger)Parameters
- name
-
String
Platform::String
winrt::hstring
The name of the input dimension to set.
- dimension
-
UInt32
unsigned int
uint32_t
The concrete value to assign to the specified dimension.
Windows requirements
| Requirements | Description |
|---|---|
| Device family |
Windows 10, version 2104 (introduced in 10.0.20348.0)
|
| API contract |
Windows.AI.MachineLearning.MachineLearningContract (introduced in v4.0)
|
Examples
void SetNamedDimensionOverrides(LearningModel model) {
// Create LearningModelSessionOptions
auto options = LearningModelSessionOptions();
// Override a named input dimension to a concrete value
options->OverrideNamedDimension(L"dimension_name", 256);
// Create session
LearningModelSession session = nullptr;
session = LearningModelSession(model, LearningModelDeviceKind::GPU, options);
}
Remarks
Using this API can yield performance improvements, as it allows for preallocation of tensors during session creation that would otherwise be allocated during model evaluation.