Wrapper class for Triton output parsing.
Definition at line 219 of file infer_trtis_server.h.
◆ TrtServerResponse()
nvdsinferserver::TrtServerResponse::TrtServerResponse |
( |
TrtServerPtr |
server, |
|
|
UniqTritonT< TRITONSERVER_InferenceResponse > |
data, |
|
|
uint64_t |
id |
|
) |
| |
|
protected |
Constructor.
- Parameters
-
server | Handle to the server class instance. |
data | Pointer to the inference result. |
id | The corresponding request ID. |
◆ getStatus()
◆ model()
const std::string& nvdsinferserver::TrtServerResponse::model |
( |
| ) |
const |
|
inline |
◆ mutableOutputs()
std::vector<SharedBatchBuf>& nvdsinferserver::TrtServerResponse::mutableOutputs |
( |
| ) |
|
|
inline |
◆ parse()
Check for error and parse the inference output.
This functions checks the error status of the response and gets the model name and version from the response. It then call the parseParams() and parseOutputData() functions.
- Parameters
-
req | Pointer to the corresponding Triton request. |
- Returns
- NVDSINFER_SUCCESS or NVDSINFER_TRITON_ERROR.
◆ takeoverOptions()
SharedOptions nvdsinferserver::TrtServerResponse::takeoverOptions |
( |
| ) |
|
|
inline |
◆ TrtISServer
The documentation for this class was generated from the following file: