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

problem about train.py #38

Closed
jiawen1929 opened this issue Nov 23, 2022 · 8 comments
Closed

problem about train.py #38

jiawen1929 opened this issue Nov 23, 2022 · 8 comments

Comments

@jiawen1929
Copy link

你好，我用单卡3060用segnext-t模型在ade20k数据集上跑实验，按作者您的原config文件跑的话，爆显存了。我就把samples_per_gpu设置成了4，learning rate还是按照0.00006跑，跑完整个实验下来，最终的输出是
image
mIoU只有14.64是啥情况呀，我需要再调整什么吗？感谢协助！
@uyzhang
Copy link
Collaborator

uyzhang commented Nov 23, 2022

需要确认一下是否把预训练权重加载进去了

@jiawen1929
Copy link
Author

你好，是预训练权重加载进去的结构，就是这样吗？
image

@jiawen1929
Copy link
Author

我实验里就是这样设置的

@uyzhang
Copy link
Collaborator

uyzhang commented Nov 23, 2022

这个权重文件是下载下来，放在了pretrained文件夹下了吗？能否把实验日志传到issue这。

@jiawen1929
Copy link
Author

20221122_000256.log
非常麻烦您了！

@jiawen1929
Copy link
Author

为了防止搞错，我在两个路径都放了pretrained文件夹

@uyzhang
Copy link
Collaborator

uyzhang commented Nov 23, 2022

还是预训练权重没有加载的问题，你可以再检查一下。
没有用预训练权重的日志：
截屏2022-11-23 11 11 46
使用预训练权重的日志：
截屏2022-11-23 11 11 50

@jiawen1929
Copy link
Author

重新下载了预训练权重文件，重新放置了一遍，成功输出您截图中的‘使用预训练权重的日志’内容了，应该是问题解决了，等实验跑完再看看结果。非常感谢您的耐心解答，祝您paper无限发！！！

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
2 participants
@uyzhang @jiawen1929