|
|
NVIDIA DeepStream SDK API Reference
|
9.0 Release
|
Go to the documentation of this file.
22 #ifndef __NVDSINFERSERVER_GRPC_BACKEND_H__
23 #define __NVDSINFERSERVER_GRPC_BACKEND_H__
40 m_RequestOutputs = names;
42 void setUrl (
const std::string &url) { m_Url = url;}
44 m_EnableCudaBufferSharing = enableSharing;
62 std::set<std::string> m_RequestOutputs;
63 std::shared_ptr<nvdsinferserver::InferGrpcClient> m_InferGrpcClient;
64 bool m_EnableCudaBufferSharing =
false;
This is a header file for pre-processing cuda kernels with normalization and mean subtraction require...
void setEnableCudaBufferSharing(const bool enableSharing)
NvDsInferStatus initialize() override
const std::string & model() const
std::function< void(SharedBatchArray)> InputsConsumed
Function wrapper called after the input buffer is consumed.
Header file of Triton Inference Server inference backend.
Triton backend processing class.
NvDsInferStatus ensureServerReady() override
void requestTritonOutputNames(std::set< std::string > &names) override
std::shared_ptr< BaseBatchArray > SharedBatchArray
std::function< void(NvDsInferStatus, SharedBatchArray)> AsyncDone
Asynchronous inference done function: AsyncDone(Status, outputs).
NvDsInferStatus ensureModelReady() override
NvDsInferStatus setupLayersInfo() override
NvDsInferStatus Run(SharedBatchArray inputs, InputsConsumed bufConsumed, AsyncDone asyncDone) override
void setOutputs(const std::set< std::string > &names)
Header file of the common declarations for the nvinferserver library.
std::function< void(NvDsInferStatus, SharedBatchArray)> InferenceDone
Function wrapper for post inference processing.
NvDsInferStatus enqueue(SharedBatchArray inputs, SharedCuStream stream, InputsConsumed bufConsumed, InferenceDone inferenceDone) override
std::shared_ptr< CudaStream > SharedCuStream
Cuda based pointers.
~TritonGrpcBackend() override
TritonGrpcBackend(std::string model, int64_t version)
Header file for the gRPC client and request class for inference using the Triton Inference Server....
void setUrl(const std::string &url)
Triton gRPC mode backend processing class.
NvDsInferStatus
Enum for the status codes returned by NvDsInferContext.