How to train a Feature Encoder of Nodes

[Reaivs]

According to the publication a ResNet50 can be trained for the Appearance Embedding.

How are this parameters trained?

[Newdxz]

According to the publication a ResNet50 can be trained for the Appearance Embedding.

How are this parameters trained?
I have same question
I try to use other cnn to reduce ids
but how to do it ?