Skip to content

Conversation

@yiyixuxu
Copy link
Collaborator

No description provided.

@yiyixuxu yiyixuxu requested a review from sayakpaul March 28, 2024 19:16
Copy link
Contributor

@tolgacangoz tolgacangoz left a comment

Choose a reason for hiding this comment

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

When I tested this test, I used Colab and it passed. I couldn't predict this OOM 👍. But I wonder if test_sd() gives the same OOM 🤔?

for shape in shapes:
image = torch.zeros(shape, device=torch_device)
pipe.vae.decode(image)
with torch.no_grad():
Copy link
Member

Choose a reason for hiding this comment

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

Was it both? torch.float16 + torch.no_grad()?

Copy link
Member

@sayakpaul sayakpaul left a comment

Choose a reason for hiding this comment

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

Thanks!

@sayakpaul sayakpaul merged commit 34c90db into main Mar 29, 2024
@sayakpaul sayakpaul deleted the fix-tilling-test branch March 29, 2024 02:52
noskill pushed a commit to noskill/diffusers that referenced this pull request Apr 5, 2024
use float16 and add torch.no_grad()
sayakpaul pushed a commit that referenced this pull request Dec 23, 2024
use float16 and add torch.no_grad()
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

Successfully merging this pull request may close these issues.

4 participants