liuyugeng / baadd Goto Github PK

Your work is very fascinating, if it is convenient for you, could you release the code in your free time?

Hello author!
I found that the net_norm in the original source code is net_norm, but here it change to 'none'
in CIFAR10 and other datasets, it's hard to convergence, the condensed dataset is poor.
Thanks

Hello author!
when I run this code, I found that the number of the tigger is out of 0~255, Is this normal? thanks!

When I run backdoor on DD, I cannot return the weights of the models.
And when I use the saved dataset to evaluate it again, I cannot get the same results as the training log.