Downloading datasets through torchvision can be quite slow or can even fail. Therefore, it is recommended that you download the datasets prior to preparing the data.
- Add MLP
- Add CNN
- Write client side training -[ ] Dynamic regularization
- Server side update
- Update only parameters & not entire state dict?
- Main FedDyn loop
- Proper communication between server and clients
- Prepare dataset
- IID
- Non-IID Balanced
- None-IID Unbalanced
- Download datasets
- Metrics (bits transferred, accuracy etc)
- Parallel training