Question about the projector backpropogation

[jinggqu]

Hello and thank you for your excellent work.

I have a question about the projector. From the code of this repository, I noticed that the parameters of the project are not included in the optimizer (shown in below), which is also evidenced in the paper Algorithm 1. I would like to know the reason for not training the projector and classifier as this is not mentioned in the paper.

Min_Max_Similarity/train_mms.py

Lines 87 to 89 in 07571ea

	params = list(self.model_1.parameters()) + list(self.model_2.parameters())
	optimizer = torch.optim.Adam(params,lr=opt.lr)

Thanks again for your contribution to the open source community.