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
Thanks for presenting this work, the method is rather intuitive and effective.
However, it seems that the provided code lacks the registration of the placeholder token and consequently there are no special embeddings set to be trainable, which means that it would be impossible to reproduce the claimed performances in the ACCV paper with the code. It would be really helpful if you could kindly clarify this issue, thanks in advance.
The text was updated successfully, but these errors were encountered:
Thanks for presenting this work, the method is rather intuitive and effective.
However, it seems that the provided code lacks the registration of the placeholder token and consequently there are no special embeddings set to be trainable, which means that it would be impossible to reproduce the claimed performances in the ACCV paper with the code. It would be really helpful if you could kindly clarify this issue, thanks in advance.
The text was updated successfully, but these errors were encountered: