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
{{ message }}
This repository has been archived by the owner on Jan 20, 2024. It is now read-only.
When I reproduce the performance of plain10 and plain2 on cifar100, I did a lot of experiments and couldn't reach the accuracy of Figure 4(b) in your paper. Therefore, I carefully read the experimental setup and found that We also used weight decay with the value of 0.0001 for training ResNets in your paper, so, can you tell me the specific method,for example weight-decay,learning-rate and crop=true or false.
The text was updated successfully, but these errors were encountered:
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
When I reproduce the performance of plain10 and plain2 on cifar100, I did a lot of experiments and couldn't reach the accuracy of Figure 4(b) in your paper. Therefore, I carefully read the experimental setup and found that We also used weight decay with the value of 0.0001 for training ResNets in your paper, so, can you tell me the specific method,for example weight-decay,learning-rate and crop=true or false.
The text was updated successfully, but these errors were encountered: