Update util.py in stable diffusion to avoid Sizes of tensors error#278
Open
CatManJr wants to merge 3 commits intolabmlai:masterfrom
Open
Update util.py in stable diffusion to avoid Sizes of tensors error#278CatManJr wants to merge 3 commits intolabmlai:masterfrom
CatManJr wants to merge 3 commits intolabmlai:masterfrom
Conversation
In my experiments, I found that when I directly load an image, both in_paint.py and image_to_image.py encounter an error:
File "/root/autodl-tmp/SD/model/unet.py", line 178, in forward
x = torch.cat([x, x_input_block.pop()], dim=1)
RuntimeError: Sizes of tensors must match except in dimension 1. Expected size 34 but got size 33 for tensor number 1 in the list.
This error can be avoided by modifying the load_img function in utils.py, specifically by changing the original multiple of 32 to a multiple of 64. This issue might be caused by the sd-v1-4.ckpt checkpoint.
I apologize if I may have mistakenly made two modifications on the master branch. Please disregard those changes, and this Pull-Request is the real debug version.
CatManJr
changed the title
leftyvuelaalto-sudo
approved these changes
Nov 3, 2025
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
In my experiments, I found that when I directly load an image, both in_paint.py and image_to_image.py encounter an error:
File "/root/autodl-tmp/SD/model/unet.py", line 178, in forward
x = torch.cat([x, x_input_block.pop()], dim=1)
RuntimeError: Sizes of tensors must match except in dimension 1. Expected size 34 but got size 33 for tensor number 1 in the list.
This error can be avoided by modifying the load_img function in utils.py, specifically by changing the original multiple of 32 to a multiple of 64. This issue might be caused by the sd-v1-4.ckpt checkpoint.
I apologize if I may have mistakenly made two modifications on the master branch. Please disregard those changes, and this Pull-Request is the real debug version.