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
Hi, in your paper, you say "After training the supernet, we randomly sample 1,000 architectures from the supernet and finetune each of them for 3 epochs", how to do this? Constructing a network according to a randomly generated genotype_list and evaluate this network?
The text was updated successfully, but these errors were encountered:
The released train codes are not for training super-net, right? I find it is hard to train super-net without detail setting. Could you please release the codes or configs for training super-net?
Hi, in your paper, you say "After training the supernet, we randomly sample 1,000 architectures from the supernet and finetune each of them for 3 epochs", how to do this? Constructing a network according to a randomly generated genotype_list and evaluate this network?
The text was updated successfully, but these errors were encountered: