Iman Tabrizian

Thanks @apokerce for filing this issue and creating detailed repro instructions. I'll file a ticket for the team to investigate.

I think It is not possible right now. @rmccorm4 do you think this is something we could add?

Python backend is not supported on Python 2.7 since Python 2 is considered EOL. > Can my python backend run in one container, and triton server run in another one?...

> Is shared memory in terms of NVIDIA triton difference from [CUDA shared memory](https://developer.nvidia.com/blog/using-shared-memory-cuda-cc/)? Yes. CUDA Shared memory is a Triton terminology for transferring CUDA tensors between client and server...

> Does pref_analyzer times also include transfering data back to CPU when cuda share memory is used? @matthewkotila / @tgerdesnv do know whether Perf Analyzer includes the time to copy...

I think this enhancement makes sense. @GuanLuo / @nnshah1 any additional thoughts?

Thanks for filing a detailed issue. I have filed an internal ticket for the team to look at it.

Thanks for filing this issue. I think we have checks for other input mismatches with the model configuration (https://github.com/triton-inference-server/core/blob/35555724612df29923007f1da45d2a58f928206c/src/infer_request.cc#L1066-L1175) looks like we need an additional check to make sure that...

@jbkyang-nvi has started taking a look at this bug. Looks like there are a few locations where we can update the checks to make sure the request has the correct...

@kbegiedza Thanks for reporting this issue. Can you share the code for `BBoxService` and `ImageService` as well so that we can repro this issue?