some other question

[UestcJay]

Thank you very much for your great work. I am a novice. Generally, when judging whether the network architecture is effective enough, do you only train part of the data? Training imagenet is too big. looking for your reply!

[gasvn]

Use a small dataset such as cifar. Most of the time, if the model's performance is good enough, the performance on ImageNet should be not that bad.