NVIDIA TensorRT Platform for High-Performance DL Inference
Published in
Jun 24, 2021
TensorRT is a framework for high-performance neural network inference and deep learning application deployment. With just a few lines of code, you can convert a TensorFlow computational graph into one that’s optimized and works much faster. Optimization affects not only the capacity of the scales, but also includes adaptation to the computing hardware. Therefore, it is important to carry it out on the machine that will be used in forecast mode.
Original article written by Rinat S.
