Closed dedoogong closed 7 months ago
We use A100 80G GPUs for the ViT model training (LVIS). Some techniques, such as FSDP, can help you reduce training memory consumption. Please refer to the latest mmdetection v3 for more details.
Thank you very very much for your kind reply!
Hello ! thanks for your great work! I'm trying to train ViT Co DETR using 4 V100 with LVIS v1 dataset but I met OOM. How much GPU memory do I need? or What GPU did you use for training ViT CoDETR?
Thank you!