-
Notifications
You must be signed in to change notification settings - Fork 5.4k
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
[PNDM] Stable diffusion #186
Conversation
The documentation is not available anymore as the PR was closed or merged. |
@@ -56,6 +56,7 @@ def __init__( | |||
beta_end=0.02, | |||
beta_schedule="linear", | |||
tensor_format="pt", | |||
skip_prk_steps=False, |
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.
This should default to True
for all stable diffusion models
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.
Great refactor, thank you Patrick!
elif len(self.ets) == 3: | ||
model_output = (23 * self.ets[-1] - 16 * self.ets[-2] + 5 * self.ets[-3]) / 12 | ||
else: | ||
model_output = (1 / 24) * (55 * self.ets[-1] - 59 * self.ets[-2] + 37 * self.ets[-3] - 9 * self.ets[-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.
Now it looks much more familiar :D
Tests are passing - merging now! |
Make PyTorch model tests importer-independent and add benchmarks.
* [PNDM] Stable diffusino * finish
This PR allows to use the PNDM scheduler without the prk steps so that it matches 1-to-1 @crowsonkb implementation as shown here.
As soon as a new diffusers version is released, we have to add a flag "skip_prk_ steps" to all stable diffusion configs.