I want to know if I can convert the TensorRT engine into a TensorFlow model(uff or pb). If so, what should I do?
I don’t think currently TensorRT engine to TensorFlow or any model conversion is currently supported.
Thank you for your reply. Is there an API that can get the calculation graph or tensorrt layer when inference, I am trying to optimize the performance of a TensorRT engine, but the TensorFlow model can’t be found now. So Does it currently support such user-defined optimizations?