Zhen Wang

@DavdGao hi dawei, we are unfamiliar with this algorithm. Could you give an explanation for us about what is the (\epsilon, \delta) each single query (training round) satisfies, when we...

This pr includes a new trainer, which is designed for conducting contrastive learning. @DavdGao could you have a look at that part for us? Thanks!

1. please answer our questions about the code; 2. do not push the changes that obviously cannot pass the UnitTest.

provide a unit test (see `test` folder) for basic fed contrastive learning and fine-tuning.

> we have suggested to add UT, but there is still nothing

> DO NOT merge this PR ahead of #365 . As that pr has been merged, please resolve the conflicts so that we can close this thread.

Thanks for your contribution! There seems to be a conflict, which has to be resolved before we could merge your commit.

I downgrade my pyg_lib from 0.4.0 to 0.3.1 according to another issue when I encountered another similar issue. However, I encounter this one. If this issue cannot be resolved now,...

Even thought I changed `disjoint` to False, it is still such an error there. It seems that pyg_lib causes the issues. `torch_sparse` works well for `disjoint=false`.