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

关于eval结果与mmdet3d官方结果对比问题 #44

Open
hehualin-tut opened this issue Apr 10, 2023 · 2 comments
Open

关于eval结果与mmdet3d官方结果对比问题 #44

hehualin-tut opened this issue Apr 10, 2023 · 2 comments

Comments

@hehualin-tut
Copy link

hehualin-tut commented Apr 10, 2023

作者您这个工程很容易理解也很容易复现,但是我还是有个问题,我看了下最新的pillars的评估结果,发现mmdet3d精度是要略微高一些的,我看了下不同的地方大概在以下几点:
1.学习率设置,官方是0.001,这个工程是0.00025,我相信这是您调试出来的较好的lr。
2.在pillar_encoder里,目前官方代码好像是不再用offset去代替原始绝对坐标了。我不太清楚目前到低使用哪个效果较好。
3.数据增强方面,不管是sample_group还是db_filter您应该是在原始的基础上又进行了调整,不知道我的猜想是否正确。
4.分布式训练所带来的提升,这一点我无法肯定,因为我没有那么多显卡去验证。
最后我再问下,您使用的坐标系最后是激光雷达坐标系吗?

@gopalkumr
Copy link

Hi, @hehualin-tut I need help in setting up the environment and running the pre-trained inferences, It would be grateful if you can help me. If possible I would love to have a little conversation with you, your any contact information like, Email, or social media link would be grateful.

@hehualin-tut
Copy link
Author

Hi, @hehualin-tut I need help in setting up the environment and running the pre-trained inferences, It would be grateful if you can help me. If possible I would love to have a little conversation with you, your any contact information like, Email, or social media link would be grateful.

you could send your issue to [email protected]

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

2 participants