Marchons

Results 10 comments of Marchons

> Good to see the proposal. The architecture format looks great to me. This version is using a single node instead of multiple nodes simulation. > > Remain works include:...

> > Good to see the proposal. The architecture format looks great to me. This version is using a single node instead of multiple nodes simulation. > > Remain works...

Sicne 8.22 weekly meeting is suspended, i would like to discuss the implement detail in github. To implement the Paradigm of federated learning and federated class incremental learning in Ianvs,...

> The formula is clarified. Now the acc is a per task / per class metric, instead of a per sample one. That is revised in the proposal now. >...

> Great to see the updated version. As mentioned, the architecture can be revised accordingly. It will be appreciated if the developed part can be highlighted in the architecture. >...

After weekly meeting, the architecture for federated learning paradigm will be shown as follow: ![image](https://github.com/user-attachments/assets/68601042-f83f-4e0c-9607-8dc1c01d63b3) And the ferederated class incremental learning paradigm will be shown as follow: ![image](https://github.com/user-attachments/assets/3b7ec87b-c495-40b7-9254-c31fc919c752)

> There are three CI issues that remain to be resolved, mainly for Pylint, see https://github.com/kubeedge/ianvs/actions/runs/11268242806 Thanks for comment!, but i think the linked action is not my pr's action....

> The EGG-INFO file should not be uploaded to the github repository. Sorry about that, i remove it right now

> This pull request contains 8 commits, which might make maintenance difficult, considering the number of contributors, pull requests, and their commits in KubeEdge Ianvs recently. In the final stage,...