How to implement batchnorm with pytorch model

AI & Data Science Deep Learning (Training & Inference) TensorRT
1

I am importing a pytorch model for inference.But does not like import from caffe or tensorflow.It looks like there is no batchnorm layer here.

I follow the pytorch example of tensorrt 5.0

Can someone provide some advice?

2

Hello,

You can use the IScaleLayer to implement the function of batchnorm.
See https://docs.nvidia.com/deeplearning/sdk/tensorrt-api/python_api/infer/Graph/Layers.html#iscalelayer