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 @rom1504 Deepspeed may be another option for bigger models. It's also easy and effective to use(PR for this #264). It can be applied with older versions of Pytorch, such as Pytorch <= 1.11.
Hi, I am doing some tests with FSDP, got a basic version running, will try to do a full run and see if we can reproduce some of the results with ViT-B/32
https://pytorch.org/docs/stable/fsdp.html
this should allow us to go to bigger model
would be quite useful to look into
The text was updated successfully, but these errors were encountered: