You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Please submit PR or issue in leaf-pytorch repo. It is confusing to submit an unrelated issue in this self-reminder repo.
Meanwhile, please provide detailed information about your problem, such as your debugging log, outputs and the specific code lines about your issue.
在使用您提供的leaf-pytorch版本中,使用updates列表存储客户端更新,但是在实际过程中,self.updates.append((num_samples, update)),update会将之前的其他客户端的update覆盖,最后使用聚合参数的时候实际上是使用都是一个客户端的参数在计算全局参数,导致了无论在哪一轮训练,全局参数始终与一个客户端的相同,希望排查一下
The text was updated successfully, but these errors were encountered: