KarhouTam / FL-bench

Benchmark of federated learning. Dedicated to the community. 🤗

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Confusion about test_before and test_after

RogerRogerUSC opened this issue · comments

Hi karhouTam,

I am just confused about understanding the "after" and "before" of test_acc(before) and test_acc(after) in your code. Could you explain them a little bit? Thank you.

Hi, @RogerRogerUSC . First, thanks for your attention to this project.

The test_before is the weighted accuracy of models that distribute to all test clients. Those models could be different from each other in some pFL methods.

The test_after is the models that mentioned above plus local fine-tuning (by client local training data). The fine-tuning epoch is set by argument --finetune_epoch.

This issue is closed due to long time no response.