How to do archchtecture search on supernet?

[TheSunWillRise]

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?

[gmh14]

I believe you have got the correct understanding of it.

[TheSunWillRise]

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?

[d12306]

right, btw, how you split the validation and training set during architecture search? @gmh14

[abhishektyaagi]

@815961618 @gmh14 Is there a way to obtain the trained super-net on any dataset?