Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

GPU usage #34

Open
SzhangS opened this issue Nov 4, 2024 · 7 comments
Open

GPU usage #34

SzhangS opened this issue Nov 4, 2024 · 7 comments

Comments

@SzhangS
Copy link

SzhangS commented Nov 4, 2024

您好, 我在预处理阶段使用SMPLer-X l32, gpu占用峰值能达到55G,请问这合理吗?或者您有什么建议?

@theFoxofSky
Copy link
Collaborator

请问预处理阶段是指Inference还是Training还是SMPL提取?

@SzhangS
Copy link
Author

SzhangS commented Nov 4, 2024

Inference阶段

@SzhangS
Copy link
Author

SzhangS commented Nov 4, 2024

训练和推理阶段的数据预处理 ,应该是一致的吧? 准确描述的话,我是在构建推理数据时候,按照prepare_pose描述的那样数据被smpler-x处理的时候,GPU峰值达到55G

@theFoxofSky
Copy link
Collaborator

训练和推理阶段的数据预处理 ,应该是一致的吧? 准确描述的话,我是在构建推理数据时候,按照prepare_pose描述

Prepare pose这个是预先在集群上跑好的,具体要多少显存我也不清楚,@Wangbenzhi 你来看看

@Wangbenzhi
Copy link
Contributor

不是很正常,V100就可以完成数据推理, 请问你的分辨率是多少?

@SzhangS
Copy link
Author

SzhangS commented Nov 5, 2024

输入mmdet的帧分辨率时(772, 1200)

@Wangbenzhi
Copy link
Contributor

我重新测试了显存使用,mmdet默认分辨率是img_scale=(1333, 800), 这个分辨率下显存开销~30G。 具体参照prepare_pose/smplerX/common/utils/human_model_files/mmdet_faster_rcnn_r50_fpn_coco.py。 如有需要,可以降低img_scale, 不过55G肯定是不正常的,可以debug看下哪一步显存开销较大。

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants