I am using trtexec to benchmark a tensorRT engine. The engine has fixed size input. I am
wondering if there is a way to get the input and output shapes. This information should be available since the engine performs inference on dummy inputs to benchmark the engine, but I don’t see it in the logs.