Is there any provision for training models with gradient accumulation?

IDEA-Research / MaskDINO

[CVPR 2023] Official implementation of the paper "Mask DINO: Towards A Unified Transformer-based Framework for Object Detection and Segmentation"

Apache License 2.0

1.15k stars 102 forks source link

Open aman0044 opened 3 months ago

aman0044 commented 3 months ago

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

FengLi-ust commented 2 months ago

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