No BatchNormlization layer when I build network manually?

For some reason, I have to parse a model manually.
I have referenced to the doc: https://docs.nvidia.com/deeplearning/sdk/tensorrt-developer-guide/index.html#create_network_python
and the Python API for Network: https://docs.nvidia.com/deeplearning/sdk/tensorrt-api/python_api/infer/Graph/pyGraph.html

I don’t found the “IBatchNormlizationLayer”,
so how can I add a BatchNormlization Layer to the network I create manually?

PS:when I use NvCaffeParser, it can parse BatchNorm layer!

Maybe it will help.

BatchNorm2D with weights from pytorch:

g0 = weights['conv1.bn.weight'].numpy().astype(ModelData.NPDTYPE).reshape(-1)
m0 = weights['conv1.bn.running_mean'].numpy().astype(ModelData.NPDTYPE).reshape(-1)
v0 = weights['conv1.bn.running_var'].numpy().astype(ModelData.NPDTYPE).reshape(-1)
scale0 = g0 / np.sqrt(v0 + 2e-5)
shift0 = -m0 / np.sqrt(v0 + 2e-5) * g0 + weights['conv1.bn.bias'].numpy().astype(ModelData.NPDTYPE).reshape(-1)
power0 = np.ones(len(g0), dtype=ModelData.NPDTYPE)
bn_1 = network.add_scale(conv1.get_output(0), trt.ScaleMode.CHANNEL,
                         shift0, scale0, power0)

much thanks!!your answer save my life!!