MaskDINO icon indicating copy to clipboard operation
MaskDINO copied to clipboard

Is there any provision for training models with gradient accumulation?

Open aman0044 opened this issue 1 year ago • 1 comments

Support for Gradient accumulation for lower batch size to accommodate large size images in single 16gb GPU?

aman0044 avatar May 19 '24 11:05 aman0044

Sorry, we did implement that. You are welcome to open a PR if you have any ideas.

FengLi-ust avatar Jul 02 '24 05:07 FengLi-ust