Navigation
API > API/Runtime > API/Runtime/NNE
The interface of a model instance that can run synchronously from CPU memory.
Use UE::NNE::IModelXXX::CreateModelInstance() to get a model instance. Use UE::NNE::GetRuntime
| Name | IModelInstanceRunSync |
| Type | class |
| Header File | /Engine/Source/Runtime/NNE/Public/NNERuntimeRunSync.h |
| Include Path | #include "NNERuntimeRunSync.h" |
Syntax
class IModelInstanceRunSync
Derived Classes
Destructors
| Name | Remarks | Include Path | Unreal Specifiers |
|---|---|---|---|
virtual ~IModelInstanceRunSync() |
NNERuntimeRunSync.h |
Typedefs
| Name | Type | Remarks | Include Path |
|---|---|---|---|
| ERunSyncStatus | EResultStatus | NNERuntimeRunSync.h | |
| ESetInputTensorShapesStatus | EResultStatus | NNERuntimeRunSync.h |
Functions
Public
| Name | Remarks | Include Path | Unreal Specifiers |
|---|---|---|---|
TConstArrayView< FTensorDesc > GetInputTensorDescs() |
Get the input tensor descriptions as defined by the model, potentially with variable dimensions. | NNERuntimeRunSync.h | |
TConstArrayView< FTensorShape > GetInputTensorShapes() |
Get the input shapes.SetInputTensorShapes must be called prior of running a model. | NNERuntimeRunSync.h | |
TConstArrayView< FTensorDesc > GetOutputTensorDescs() |
Get the output tensor descriptions as defined by the model, potentially with variable dimensions. | NNERuntimeRunSync.h | |
TConstArrayView< FTensorShape > GetOutputTensorShapes () |
Getters for outputs shapes if they were already resolved. | NNERuntimeRunSync.h | |
ERunSyncStatus RunSync
(
TConstArrayView< FTensorBindingCPU > InInputTensors, |
Evaluate the model synchronously. | NNERuntimeRunSync.h | |
ESetInputTensorShapesStatus SetInputTensorShapes
(
TConstArrayView< FTensorShape > InInputShapes |
Prepare the model to be run with the given input shape. | NNERuntimeRunSync.h |