ChG
I'd like to work on this!
I'd like to work on this issue but I can't reproduce it. The tutorial works well on my machine and on the document website. And the error message for `non_iid`...
I'm interested in working on this!
Hi @August-murr , happy to see that you have already worked it out! However I noticed that your implementation only allows merge models in the disk after training, this could...
> @coding-famer The callback has an optional parameter called `merge_at_every_checkpoint`, which merges the saved checkpoint at either every step or at the end of each epoch during training. Sounds great!