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

tensorflow version #35

Open
Yakuho opened this issue Feb 20, 2021 · 19 comments
Open

tensorflow version #35

Yakuho opened this issue Feb 20, 2021 · 19 comments

Comments

@Yakuho
Copy link

Yakuho commented Feb 20, 2021

Thanks for your excellent work.Do you have tensorflow verision about this code?

@Maclory
Copy link
Owner

Maclory commented Feb 20, 2021

Sorry, we only have the code of pytorch version.

@Yakuho
Copy link
Author

Yakuho commented Feb 20, 2021

ok, thanks. if I want to use LFW and HElEN dataset to train model. What should I do for it.

@Maclory
Copy link
Owner

Maclory commented Feb 20, 2021

You can follow the instructions in README.md, issue#10 and issue#21.

@Yakuho
Copy link
Author

Yakuho commented Feb 20, 2021

thanks!!!

@Yakuho
Copy link
Author

Yakuho commented Feb 20, 2021

Do you have communication groups? such qq?

@Maclory
Copy link
Owner

Maclory commented Feb 20, 2021

It's my pleasure. We don't have communication groups. We can communicate here or by emails:-)

@Yakuho
Copy link
Author

Yakuho commented Feb 20, 2021

what size image should be set(such Face with background or only face, I found that when I used images containing only faces to get landmarks, OpenFace failed to detect the box a lot of the time), when i get the landmark?? :-)

@Maclory
Copy link
Owner

Maclory commented Feb 21, 2021

In my opinion, OpenFace is a powerful tool and you can process images with background.

@Yakuho
Copy link
Author

Yakuho commented Feb 23, 2021

作者你好,我已成功标注好训练集,当我尝试训练的时候它抛出了显存溢出的错误,我是使用的Colab实验室训练的模型(显存有15Gb),batch_size=8等配置都是使用默认的。请问您们在训练的时候,设置的batch_size是多大呢?因为我注意到论文提到您们使用的显卡为2080ti。感谢作者 :-)

@Maclory
Copy link
Owner

Maclory commented Feb 23, 2021

2080ti显卡显存是11GB,用默认设置应该都是可以正常运行的

@Yakuho
Copy link
Author

Yakuho commented Feb 23, 2021

你们使用的是双卡吗,因为我注意到默认的device配置是[0, 1]。

@Maclory
Copy link
Owner

Maclory commented Feb 23, 2021

对,你可以调低batchsize试一下

@Yakuho
Copy link
Author

Yakuho commented Feb 23, 2021

我已经调成4了, 已经成功运行。我就是存在一个疑问,colab的显卡是tesla t4 实际显存15G, 单显卡模式,无法运行你们默认的配置。QAQ(哭泣)

@Maclory
Copy link
Owner

Maclory commented Feb 23, 2021

嗯嗯,可以看看效果怎么样

@Yakuho
Copy link
Author

Yakuho commented Feb 23, 2021

感谢您百忙之中的回复 👍

@Maclory
Copy link
Owner

Maclory commented Feb 23, 2021

不客气:-)

@Yakuho
Copy link
Author

Yakuho commented Feb 24, 2021

早上好,又打扰您了。由于我在colab上训练您的模型,colab的规则是12小时断一次,那么我需要在哪里去配置我的相关东西去继续我的训练呢?

@Maclory
Copy link
Owner

Maclory commented Feb 24, 2021

找一下resume相关的配置就可以了

@Yakuho
Copy link
Author

Yakuho commented Feb 24, 2021

你好,我在使用CelebA的pretrain_HG尝试迁移学习训练DIC_Helen的时候,发现在21K step的时候align loss突然猛降,但在tensorboard中看到landmark的点变得混乱了,不再是依照人脸的轮廓(在这之前可以依照轮廓标点), 这是什么原因呢?或者请问你能给我一些建议吗。谢谢。

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