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.

Sign up for GitHub

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

Jump to bottom

当RT-DETR的backbone为ResNet50或ResNet101时，训练COCO2017数据集，训练时eval的map一直是0 #3

Closed
wmjlincy opened this issue Jul 3, 2023 · 3 comments
Closed

当RT-DETR的backbone为ResNet50或ResNet101时，训练COCO2017数据集，训练时eval的map一直是0 #3

wmjlincy opened this issue Jul 3, 2023 · 3 comments

Comments

@wmjlincy
Copy link

wmjlincy commented Jul 3, 2023

大佬好，当我用RT-DETR的backbone为ResNet50训练COCO2017数据集时，训练过程中eval的mAP一直是0，且loss在不断下降，pretrain_weights加载的是ResNet50_vd_ssld_v2_pretrained.pdparams，其他参数没做改动；除此之外，每次eval保存的bbox.json文件中，每张图片保存了300个检测框，但是每个检测框的类别、box位置和score都是一样的，请问我需要怎么解决呢？非常期待大佬能够回答一下我的问题，感激不尽~

下面是我训练时的环境：
我用的是PaddlePaddle 2.4.2的镜像，CUDA是11.8。
@lyuwenyu
Copy link
Owner

lyuwenyu commented Jul 3, 2023

用这版代码先试一下 我跑了r18验证 好像没啥问题，

@wmjlincy
Copy link
Author

wmjlincy commented Jul 3, 2023

你好，我试了一下，用这版代码跑R50没问题了，感谢大佬~

@wmjlincy wmjlincy closed this as completed Jul 3, 2023
This was referenced Jul 19, 2023
Open
Closed
Open
@lyuwenyu lyuwenyu mentioned this issue Jul 20, 2023
Open
@lkz98
Copy link

lkz98 commented Jan 21, 2024

用这版代码先试一下 我跑了r18验证 好像没啥问题，

你好，请问是哪版代码啊

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@lyuwenyu @wmjlincy @lkz98