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
Hi @ailearnerkyh,
With 8 A100 GPUs, you can have a good checkpoint at about 72k iterations within two days (or maybe 48k within one day).
Training longer may lead to better results.
Nice work! How long and how much GPU memory is required to train this model on WebVid-10M?
The text was updated successfully, but these errors were encountered: