TensorRT INT8 Python Demo CUDA 8.0 CUDNN 7.2.1 TensorRT 4.0.1.4 Python 3.5.2 How to use Just run python trt_int8_mnist.py If I want to quant my model, What should I do? Create an issue at here