diffusers import error

[marson666]

hi, thanks for you great work! i follow the README.md to install the deps, but when i run the stable_diffusion_xl_test.py, i find the import error below:

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/home/test/Free-Lunch-for-Text-guided-Style-Transfer-using-Diffusion-Models/diffusers/src/diffusers/__init__.py", line 1, in <module>
    from ..utils import (
ImportError: attempted relative import beyond top-level package

look forward to your reply

[JACKYLUO1991]

@marson666 I'm experiencing the same problem. Also, could the author list the file structure of the weights so that we can easily verify the results? @FreeStyleFreeLunch

[Loafei123]

The model and its weight file directory structure are shown in the figure. Consistent with the given link structure.

[FreeStyleFreeLunch]

The model and its weight file directory structure are shown in the figure. Consistent with the given link structure. @JACKYLUO1991 @marson666

[JACKYLUO1991]

@FreeStyleFreeLunch What is the difference between the official diffusers==0.21.0 and the ones in the code base?

[FreeStyleFreeLunch]

Unlike the official version, the diffusers code provided in the file has been modified in the UNet section.

[JACKYLUO1991]

@marson666 With so many bugs, did you run it successfully?

[haofanwang]

It is necessary to specify what modification has been made in UNet. If only several files get modified, can you make it compatible with official diffusers?

[FreeStyleFreeLunch]

We have made some modifications to the file structure within the diffusers project. Please download the code directly from the repository and install it following the instructions in the README.md file. We will endeavor to resolve any compatibility issues to the best of our ability in our subsequent work.

[marson666]

@marson666 With so many bugs, did you run it successfully?

@JACKYLUO1991 not yet

[loboere]

Has anyone managed to run it in colab? I tried but I only get import errors

[FreeStyleFreeLunch]

The bug has been fixed and it should be running normally now. @loboere @marson666 @JACKYLUO1991 @haofanwang @marson666 @Loafei123

[geyanqi]

It is necessary to specify what modification has been made in UNet. If only several files get modified, can you make it compatible with official diffusers?

If I understand correctly, this part should be in:

upper branch:

1. https://github.com/FreeStyleFreeLunch/FreeStyle/blob/main/diffusers_test/pipeline_stable_diffusion_xl.py#L1007 where skip_stack == None

2. https://github.com/FreeStyleFreeLunch/FreeStyle/blob/main/diffusers/src/diffusers/models/unet_2d_condition.py#L976 return the unet encoder features to skip_stack,

lower branch:

1. https://github.com/FreeStyleFreeLunch/FreeStyle/blob/main/diffusers_test/pipeline_stable_diffusion_xl.py#L1018 where skip_stach is the upper branch's return

2. https://github.com/FreeStyleFreeLunch/FreeStyle/blob/main/diffusers/src/diffusers/models/unet_2d_condition.py#L97 use skip_stach

3. https://github.com/FreeStyleFreeLunch/FreeStyle/blob/main/diffusers/src/diffusers/models/unet_2d_blocks.py#L2182 core code.

[loboere]

I want to download sdxl in folders mode with !wget https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0 but download everything including fp16 mode and all that is 80gb. How do I download SDXL in folder mode without downloading repeated files?

[ucasligang]

How do I download SDXL in folder mode without downloading repeated files?

Have you solved it now？

[BlueDyee]

It is necessary to specify what modification has been made in UNet. If only several files get modified, can you make it compatible with official diffusers?

If I understand correctly, this part should be in:

upper branch:

1. https://github.com/FreeStyleFreeLunch/FreeStyle/blob/main/diffusers_test/pipeline_stable_diffusion_xl.py#L1007 where skip_stack == None
2. https://github.com/FreeStyleFreeLunch/FreeStyle/blob/main/diffusers/src/diffusers/models/unet_2d_condition.py#L976 return the unet encoder features to skip_stack,

lower branch:

1. https://github.com/FreeStyleFreeLunch/FreeStyle/blob/main/diffusers_test/pipeline_stable_diffusion_xl.py#L1018 where skip_stach is the upper branch's return
2. https://github.com/FreeStyleFreeLunch/FreeStyle/blob/main/diffusers/src/diffusers/models/unet_2d_condition.py#L97 use skip_stach
3. https://github.com/FreeStyleFreeLunch/FreeStyle/blob/main/diffusers/src/diffusers/models/unet_2d_blocks.py#L2182 core code.

Thank you for sharing your observation😍 this does save me a lot of time. But, I think there might be a typo (or maybe I misunderstood you) in this line: "> 2. https://github.com/FreeStyleFreeLunch/FreeStyle/blob/main/diffusers/src/diffusers/models/unet_2d_condition.py#L97

use skip_stach"

I guess this line is more correct "https://github.com/FreeStyleFreeLunch/FreeStyle/blob/c55481d4ab32ef0f96d72838253721707fede5c2/diffusers/src/diffusers/models/unet_2d_condition.py#L973"