This repository contains the code related to model quantisation with TensorRT for TensorFlow 1.X, 2.X (highly recommend migrating to 2.X to avoid migraines) and PyTorch (to be created)
These scripts work on Nvidia GPUs and Nvidia Jetsons, (GTX1060 6Gb, Jetson Xavier and Jetson Nano 2Gb and 4Gb tested)
These scripts have been tested with Python 3.6.9 and: