No documentation on using infer_request proto?

I am building a java client for Tensorrt inference server.
All the example given are using infer_context which is supported in python and c++ libraries.
There is no protobuf file for infer_context. The protobuf that is shared in the repository is infer_request.proto. And there is no proper documentation to use infer_request.proto
Also there is no good documentation to use Rest APIs as well.
Can someone please help me in passing the right inputs and header for InferRequest?

1 Like

“infer_context” is a C++ or python class/object that is documented in the API section of the documentation. Or you can look in request.h or init.

https://docs.nvidia.com/deeplearning/sdk/tensorrt-inference-server-master-branch-guide/docs/cpp_api/classnvidia_1_1inferenceserver_1_1client_1_1InferContext.html

https://docs.nvidia.com/deeplearning/sdk/tensorrt-inference-server-master-branch-guide/docs/python_api.html#module-tensorrtserver.api

The protobuf file that defines the GRPC interface is: https://github.com/NVIDIA/tensorrt-inference-server/blob/master/src/core/grpc_service.proto

There is documentation in that protobuf file and you should follow standard GRPC documentation on how to compile that file into a client implementation. We show python example here: https://github.com/NVIDIA/tensorrt-inference-server/blob/master/src/clients/python/grpc_image_client.py
and go example here: https://github.com/NVIDIA/tensorrt-inference-server/tree/master/src/clients/go

@David Goodwin Thanks for the update.
I was able to create the Java client.