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
I noticed in your paper the training epoch is 20 and warmup is 5 epochs, but when I tried to reproduce the results with train.sh, it runs for 100 epochs with warmup_epochs set to 10,
Another disparity is the weight_decay, the paper indicate the end is 0.1 while the code use 0.4.
Do I need to change the code to get it aligned with the paper to reproduce the results ?
Thanks!.
The text was updated successfully, but these errors were encountered:
Hi,
I noticed in your paper the training epoch is 20 and warmup is 5 epochs, but when I tried to reproduce the results with train.sh, it runs for 100 epochs with warmup_epochs set to 10,
Another disparity is the weight_decay, the paper indicate the end is 0.1 while the code use 0.4.
Do I need to change the code to get it aligned with the paper to reproduce the results ?
Thanks!.
The text was updated successfully, but these errors were encountered: