Skip to content

fixed bug in defining embed dim for UNet1D #12111

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.

Sign up for GitHub

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

Merged
merged 5 commits into from
Sep 16, 2025
+11 −2
Merged

fixed bug in defining embed dim for UNet1D #12111

Changes from 2 commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
f1b3cca
fixed bug in defining embed dim
SammyAgrawal Aug 10, 2025
ff2df6e
Merge branch 'main' into unet-1d-time-embed-debugging
SammyAgrawal Aug 11, 2025
9335c36
matched 1d temb process to 2d
SammyAgrawal Aug 26, 2025
7881392
Merge branch 'unet-1d-time-embed-debugging' of https://github.com/Sam…
SammyAgrawal Aug 26, 2025
4488713
Update src/diffusers/models/unets/unet_1d.py
SammyAgrawal Sep 16, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion src/diffusers/models/unets/unet_1d.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -101,7 +101,7 @@ def __init__(
# time
if time_embedding_type == "fourier":
self.time_proj = GaussianFourierProjection(
embedding_size=8, set_W_to_weight=False, log=False, flip_sin_to_cos=flip_sin_to_cos
embedding_size=block_out_channels[0], set_W_to_weight=False, log=False, flip_sin_to_cos=flip_sin_to_cos
Copy link
Collaborator

@DN6 DN6 Aug 12, 2025 •
edited
Loading

Choose a reason for hiding this comment

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

I think it might be better to introduce an optional time_embedding_dim argument to the class init and configure this way

diffusers/src/diffusers/models/unets/unet_2d_condition.py

Line 549 in f8ba5cd

time_embed_dim = time_embedding_dim or block_out_channels[0] * 2

)
timestep_input_dim = 2 * block_out_channels[0]
elif time_embedding_type == "positional":
Expand Down