ExecuteNeuralNetwork (MISO)

Execute a neural network with multiple inputs and single output (MISO) using the provided images. Ensure all model input names are listed in sInputNames and that the images and names are in the same order. The result of the specified output layer is returned. The function doesn't perform any pre-processing or transformation.
Can use available TwinCAT Job Tasks for executing parallel code regions.

Syntax

Definition:

HRESULT ExecuteNeuralNetwork(
    HRESULT             hrPrev,
    ITcVnNeuralNetwork* ipNeuralNetwork,
    PVOID               pSrcImages,
    USHORT              nSrcArraySize,
    PCCH                sInputNames,
    ITcVnImage*&        ipDestImage,
    PCCH                sOutputName
)

Parameters

Name	Type	Description
hrPrev	HRESULT	HRESULT indicating the result of previous operations (If SUCCEEDED(hrPrev) equals false, no operation is executed.)
ipNeuralNetwork	ITcVnNeuralNetwork*	Neural network model
pSrcImages	PVOID	Pointer to an array of source input images
nSrcArraySize	USHORT	The size of the array referenced by pSrcImages. pSrcImages must reference an array with a size equal to the number of model inputs.
sInputNames	PCCH	Input names separated by ";"
ipDestImage	ITcVnImage*&	Destination image (An appropriate destination image based on the output layer is created. To process the destination image, use e.g. F_VN_ConvertDataLayout or F_VN_ReshapeImage if necessary.
sOutputName	PCCH	Specify the name of the layer for which the output is required. If empty, the output of one output layer is provided.

Return value

HRESULT

Required License

TC3 Neural Network Realtime Inference

System Requirements

Development environment	Target platform	PLC libraries to include
TwinCAT V3.1.4024.59 or later	PC or CX (x64) with min. PL50, e.g. Intel 4-core Atom CPU	Tc3_Vision