Hi everyone,
I’m working on a project where I need to convert a PyAerial-based model into a TensorRT (TRT) engine for integration with an OTA test bench. I came across the NVIDIA documentation here, which describes the process, but I have a few questions:
- Code Location: The documentation mentions examples for converting models, specifically referencing
torch_to_trt_chest_example.py. However, I can’t locate this file. Could anyone guide me on where to find it? - ARC-OTA Version Support: From which version of ARC-OTA does the system begin supporting TRT model engines? Is there a specific configuration or dependency I need to be aware of for compatibility?
Any insights, references, or guidance on how to get started with this would be highly appreciated. Thank you in advance for your help!