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

[PPDiffusers] Not use recompute #374

Closed

Conversation

co63oc
Copy link
Contributor

@co63oc co63oc commented Jan 5, 2024

增加 use_old_recompute

Copy link

paddle-bot bot commented Jan 5, 2024

Thanks for your contribution!

@luotao1 luotao1 added the PaddlePaddle Hackathon 飞桨黑客松活动issue与PR label Jan 15, 2024
@JunnYu JunnYu changed the title Not use recompute [PPDiffusers] Not use recompute Jan 15, 2024
Copy link
Member

@JunnYu JunnYu left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

这些都不用修改吧?

self.training
and self.gradient_checkpointing
and not hidden_states.stop_gradient
and not use_old_recompute()
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

这些应该不用加吧?之前都没有这些模块的。0.19.4

Copy link
Contributor Author

@co63oc co63oc Jan 15, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

https://github.com/PaddlePaddle/PaddleMIX/blob/ppdiffusers0.19.4/ppdiffusers/ppdiffusers/models/unet_3d_blocks.py
DownBlockMotion 是没有这些模块,升级增加的,不是要测试 recompute吗
增加 and not use_old_recompute() 用来测试 recompute

self.training
and self.gradient_checkpointing
and not hidden_states.stop_gradient
and not use_old_recompute()
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

这些应该不用加吧?之前都没有这些模块的。0.19.4

and self.gradient_checkpointing
and not hidden_states.stop_gradient
and not use_old_recompute()
):
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

这些应该不用加吧?之前都没有这些模块的。0.19.4

self.training
and self.gradient_checkpointing
and not hidden_states.stop_gradient
and not use_old_recompute()
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

这些应该不用加吧?之前就没有这

self.training
and self.gradient_checkpointing
and not hidden_states.stop_gradient
and not use_old_recompute()
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

这些应该不用加吧?之前都没有这些模块的。0.19.4

@JunnYu JunnYu self-requested a review January 15, 2024 07:32
@JunnYu
Copy link
Member

JunnYu commented Jan 15, 2024

我看,这两个文件都不应该修改

@JunnYu JunnYu closed this Jan 15, 2024
@co63oc co63oc deleted the not_recompute branch February 18, 2024 07:02
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
contributor PaddlePaddle Hackathon 飞桨黑客松活动issue与PR
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants