Hello all,
Does anyone have experience adding a tensorflow model into another CUDA pipeline? Basically, I have a trained TF model that I want to load in TF C++ and run on the GPU in the middle of my own custom CUDA pipeline. Something like this:
Thanks!