what gpu used to finetune the 2D model?

[krishnaadithya]

I treid finetuning the model on 24 gb vram GPU but it runs out of memory and only trains with batch size 2

[anlarro]

Hi. Any update on this? I experience the same issue, training with a Tesla V100 32GB only allows me to use a batch size of 3.

[rocketche]

I treid finetuning the model on 24 gb vram GPU but it runs out of memory and only trains with batch size 2

You can try it on vit_base instead of vit_large or vit_huge. Please let me know if it helps you. @krishnaadithya @anlarro

[anlarro]

I treid finetuning the model on 24 gb vram GPU but it runs out of memory and only trains with batch size 2

You can try it on vit_base instead of vit_large or vit_huge. Please let me know if it helps you. @krishnaadithya @anlarro

Using vit_b with a Tesla V100 32GB only allows me to use a batch size of 3

[WuJunde]

We have added MobileSAM and EffficientSAM as the optional backbone, maybe you can give it a try

[WuJunde]

@anlarro

• 24-01-11. Added a detailed guide on utilizing the Efficient Med-SAM-Adapter, complete with a comparison of performance and speed. You can find this resource in guidance/efficient_sam.ipynb. Credit: @shinning0821
• 24-01-14. We've just launched our first official version, v0.1.0-alpha 🥳. This release includes support for MobileSAM, which can be activated by setting -net mobile_sam. Additionally, you now have the flexibility to use ViT, Tiny ViT, and Efficient ViT as encoders. Check the details here. Credit: @shinning0821
• 24-01-20. Added a guide on utilizing the mobile sam in Med-SAM-Adapter, with a comparison of performance and speed. You can find it in guidance/mobile_sam.ipynb Credit: @shinning0821