-
Notifications
You must be signed in to change notification settings - Fork 186
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
Conversation
Thanks for your contribution! |
There was a problem hiding this 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() |
There was a problem hiding this comment.
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
There was a problem hiding this comment.
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() |
There was a problem hiding this comment.
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() | ||
): |
There was a problem hiding this comment.
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() |
There was a problem hiding this comment.
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() |
There was a problem hiding this comment.
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
我看,这两个文件都不应该修改 |
增加 use_old_recompute