haoyu

It's a wonderful text classification system! I wonder what the form should the input be. Would you like to help me?

about the average_weights function

1

In the original paper, it uses a weighted average here. However, the implementation in average_weights is the simple average. Is there a bug or do I misunderstand something? Thanks!

about the implementation of FedAvg

3

Why does the FedAvg use a simple average without weight?

Does the project compute the FLOPs or MACs?

1

I find the project ignores the add operations in matrix product. So does it compute MACs instead of FLOPs?

Hi, the long-tail hashing is a wonderful work! Would you like to provide the code about how to generate k prototypes?