DLInference#

Macros#

PVA_DL_INFERENCE_MAX_NUM_INPUTS

Maximum number of input tensors supported by the DL inference operator.

PVA_DL_INFERENCE_MAX_NUM_OUTPUTS

Maximum number of output tensors supported by the DL inference operator.

Functions#

NVCVStatus pvaDLInferenceCreate(NVCVOperatorHandle *handle, PvaDLInferenceCreateParams *params)

Creates an instance of the DL inference operator.

NVCVStatus pvaDLInferenceSubmit(NVCVOperatorHandle handle, cudaStream_t stream, PvaDLInferenceSubmitParams *params)

Submits the DLInference operator to a CUDA stream.

NVCVStatus pvaDLInferenceSubmit(NVCVOperatorHandle handle, cupvaStream_t stream, PvaDLInferenceSubmitParams *params)

Submits the DLInference operator to a cuPVA stream.

Structs#

PvaDLInferenceCreateParamRec

Parameters for creating an DL inference operator.

PvaDLInferenceSubmitParamRec

Parameters for executing a DL inference operation.

Functions#

NVCVStatus pvaDLInferenceCreate(
NVCVOperatorHandle *handle,
PvaDLInferenceCreateParams *params,
)#

Creates an instance of the DL inference operator.

Parameters:

  • handle – [out] Where the operator instance handle will be written to. Must not be NULL.

  • params – [in] Pointer to the PvaDLInferenceCreateParams structure containing tensor requirements for inputs and outputs.

Return values:

  • NVCV_ERROR_INVALID_ARGUMENT – Handle or params is null.

  • NVCV_ERROR_OUT_OF_MEMORY – Not enough memory to create the operator.

  • NVCV_SUCCESS – Operation executed successfully.

NVCVStatus pvaDLInferenceSubmit(
NVCVOperatorHandle handle,
cudaStream_t stream,
PvaDLInferenceSubmitParams *params,
)#

Submits the DLInference operator to a CUDA stream.

Note

CUDA stream support requirements:

  • PVA SDK 2.7.0 or later

  • Jetpack 7 or later

  • DriveOS 7 or later

  • x86 Emulator is not supported

Parameters:

  • handle – [in] Handle to the operator.

  • stream – [in] Handle to a valid CUDA stream.

  • params – [in] Pointer to the PvaDLInferenceSubmitParams structure containing input and output tensor handles.

Return values:

  • NVCV_ERROR_INVALID_ARGUMENT – Handle or params is null.

  • NVCV_ERROR_INTERNAL – Internal error in the operator.

  • NVCV_SUCCESS – Operation executed successfully.

NVCVStatus pvaDLInferenceSubmit(
NVCVOperatorHandle handle,
cupvaStream_t stream,
PvaDLInferenceSubmitParams *params,
)#

Submits the DLInference operator to a cuPVA stream.

Parameters:

  • handle – [in] Handle to the operator. Must not be NULL.

  • stream – [in] Handle to a valid cuPVA stream.

  • params – [in] Pointer to the PvaDLInferenceSubmitParams structure containing input and output tensor handles.

Return values:

  • NVCV_ERROR_INVALID_ARGUMENT – Handle or params is null.

  • NVCV_ERROR_INTERNAL – Internal error in the operator.

  • NVCV_SUCCESS – Operation executed successfully.