Metodo ITensorNative.GetBuffer
Ottiene il buffer del tensore come matrice di byte.
HRESULT GetBuffer(
[out, size_is(, *capacity)] BYTE **value,
[out] UINT32 *capacity);
Parametri
Nome | Digita | Descrizione |
---|---|---|
value | BYTE** | Buffer del tensore. |
capacità | UINT32* | Capacità del buffer. |
Resi
HRESULT Il risultato dell'operazione.
Esempi
TensorFloat SoftwareBitmapToSoftwareTensor(SoftwareBitmap softwareBitmap)
{
// 1. Get access to the buffer of softwareBitmap
BYTE* pData = nullptr;
UINT32 size = 0;
BitmapBuffer spBitmapBuffer(softwareBitmap.LockBuffer(BitmapBufferAccessMode::Read));
winrt::Windows::Foundation::IMemoryBufferReference reference = spBitmapBuffer.CreateReference();
auto spByteAccess = reference.as<::Windows::Foundation::IMemoryBufferByteAccess>();
CHECK_HRESULT(spByteAccess->GetBuffer(&pData, &size));
// ...
}
Vedi anche
Requisiti
Requisito | |
---|---|
Client minimo supportato | Windows 10, build 17763 |
Server minimo supportato | Windows Server 2019 con Esperienza desktop |
Intestazione | windows.ai.machinelearning.native.h |
Nota
Per informazioni su Windows Machine Learning, usa le risorse seguenti:
- Per porre domande tecniche o rispondere a domande tecniche su Windows Machine Learning, usa il tag windows-machine-learning in Stack Overflow.
- Per segnalare un bug, registra il problema in GitHub.