Jian Zou

icon company Harbin Institute of Technology

Results 3 issues of Jian Zou

How to train on multi-gpu?

9 comment

Hello, when I run: `torchpack dist-run -np 4 python tools/train.py configs/nuscenes/det/centerhead/lssfpn/camera/256x704/swint/default.yaml --model.encoders.camera.backbone.init_cfg.checkpoint pretrained/swint-nuimages-pretrained.pth` I get an error: ``` -------------------------------------------------------------------------- There are not enough slots available in the system to satisfy...

Did you try pretraining swin-tiny？

For the GPU memory restriction, I could use swin-tiny as my backbone. Did you try pretraining swin-tiny with GreenMIM? I tried it, but it seems get bad performance, even worse...

For an intuitive interpretation, are there h×w pixels in one region s?

And one token serves as one region **_s_**? As a consequence, there are H × W region **_s_** in H × W grids? Thanks for your source code for implementation!