Header file of wrapper classes for Triton Inference Server server instance, inference request, response.
This file declares the wrapper classes used for inference processing using the Triton Inference Server C-API mode.
Definition in file infer_trtis_server.h.
Go to the source code of this file.
Data Structures | |
| class | nvdsinferserver::TrtServerRequest |
| Wrapper class for Triton inference request. More... | |
| class | nvdsinferserver::TrtServerResponse |
| Wrapper class for Triton output parsing. More... | |
| class | nvdsinferserver::TrtServerAllocator |
| Wrapper class for Triton server output memory allocator. More... | |
| struct | nvdsinferserver::triton::BackendConfig |
| The backend configuration settings. More... | |
| struct | nvdsinferserver::triton::RepoSettings |
| Model repository settings for the Triton Inference Server. More... | |
| class | nvdsinferserver::TrtISServer |
| Wrapper class for creating Triton Inference Server instance. More... | |
Namespaces | |
| nvdsinferserver | |
| Copyright (c) 2021, NVIDIA CORPORATION. | |
| nvdsinferserver::triton | |
Macros | |
| #define | TRITON_DEFAULT_MINIMUM_COMPUTE_CAPABILITY 6.0 |
| #define | TRITON_DEFAULT_PINNED_MEMORY_BYTES (1 << 28) |
| #define | TRITON_DEFAULT_BACKEND_DIR "/opt/tritonserver/backends" |
Typedefs | |
| using | nvdsinferserver::SharedRequest = std::shared_ptr< TrtServerRequest > |
| using | nvdsinferserver::UniqResponse = std::unique_ptr< TrtServerResponse > |
| using | nvdsinferserver::SharedResponse = std::shared_ptr< TrtServerResponse > |
| using | nvdsinferserver::TritonInferAsyncDone = std::function< void(SharedRequest, UniqResponse)> |
| #define TRITON_DEFAULT_BACKEND_DIR "/opt/tritonserver/backends" |
Definition at line 46 of file infer_trtis_server.h.
| #define TRITON_DEFAULT_MINIMUM_COMPUTE_CAPABILITY 6.0 |
Definition at line 43 of file infer_trtis_server.h.
| #define TRITON_DEFAULT_PINNED_MEMORY_BYTES (1 << 28) |
Definition at line 44 of file infer_trtis_server.h.