Function TRTSERVER_InferenceRequestProviderNew

Function Documentation

TRTSERVER_Error *TRTSERVER_InferenceRequestProviderNew(TRTSERVER_InferenceRequestProvider **request_provider, TRTSERVER_Server *server, const char *model_name, int64_t model_version, const char *request_header_base, size_t request_header_byte_size)

TRTSERVER_InferenceRequestProvider.

Object representing the request provider for an inference request. The request provider provides the meta-data and input tensor values needed for an inference.[DEPRECATED] Please use TRTSERVER_InferenceRequestProviderNewV2() API Create a new inference request provider object. The request header protobuf must be serialized and provided as a base address and a size, in bytes.

Return

a TRTSERVER_Error indicating success or failure.

Parameters

  • request_provider: Returns the new request provider object.

  • server: the inference server object.

  • model_name: The name of the model that the inference request is for.

  • model_version: The version of the model that the inference request is for, or -1 to select the latest (highest numbered) version.

  • request_header_base: Pointer to the serialized request header protobuf.

  • request_header_byte_size: The size of the serialized request header in bytes.