SD Web UI can't find xformers

[dochoss]

Hi all, I'm very new to working with AI tools, but have extensive dev experience (not in Python, though...). I got the SD web UI working and added the Dreambooth extension. I followed this guy's (amazing!) walkthrough to get set up for training, and got to the step to "Generate Class Images." When I run that, I get "Exception generating concepts: Refer to https://github.com/facebookresearch/xformers for more information on how to install xformers" in the Output pane of the UI. Full error text/stack trace is below. Any guidance would be greatly appreciated!

I have installed xformers following the instructions and restarted the web UI, but the same issue appears. I've installed all the latest versions of Dreambooth, SD web UI, Python, and xformers as of this writing (8/23/2023).

*** Stack Trace ***

Generating class images 0/608::   0%|                                                          | 0/608 [00:02<?, ?it/s]
Duration: 00:00:02
[2023-08-23 17:14:22,848][DEBUG][dreambooth.dataclasses.db_config] - Saving to C:\Users\me\source\repos\stable-diffusion-webui\models\dreambooth\bo-1 ohwx
Pre-processing images: classification-dataset-dir: : 32it [00:00, 1813.83it/s]                  | 0/16 [00:00<?, ?it/s]
We need a total of 608 class images.n-dataset-dir: : 32it [00:00, 2045.63it/s]
Processing concept 1/1:   6%|███▋                                                       | 1/16 [00:00<00:00, 53.64it/s]
Generating 608 class images for training...
[2023-08-23 17:14:23,315][INFO][dreambooth.utils.gen_utils] - Creating new progress bar
Loading pipeline components...: 100%|████████████████████████████████████████████████████| 7/7 [00:01<00:00,  3.59it/s]
Traceback (most recent call last):1%|█████████████████████████████████████▏              | 5/7 [00:01<00:00,  3.04it/s]
  File "C:\Users\me\source\repos\stable-diffusion-webui\extensions\sd_dreambooth_extension\dreambooth\ui_functions.py", line 841, in ui_classifiers
    count, images = generate_classifiers(config, ui=True)
  File "C:\Users\me\source\repos\stable-diffusion-webui\extensions\sd_dreambooth_extension\dreambooth\utils\gen_utils.py", line 153, in generate_classifiers
    builder = ImageBuilder(
  File "C:\Users\me\source\repos\stable-diffusion-webui\extensions\sd_dreambooth_extension\helpers\image_builder.py", line 110, in __init__
    self.image_pipe.enable_xformers_memory_efficient_attention()
  File "C:\Users\me\source\repos\stable-diffusion-webui\venv\lib\site-packages\diffusers\pipelines\pipeline_utils.py", line 1641, in enable_xformers_memory_efficient_attention
    self.set_use_memory_efficient_attention_xformers(True, attention_op)
  File "C:\Users\me\source\repos\stable-diffusion-webui\venv\lib\site-packages\diffusers\pipelines\pipeline_utils.py", line 1667, in set_use_memory_efficient_attention_xformers
    fn_recursive_set_mem_eff(module)
  File "C:\Users\me\source\repos\stable-diffusion-webui\venv\lib\site-packages\diffusers\pipelines\pipeline_utils.py", line 1657, in fn_recursive_set_mem_eff
    module.set_use_memory_efficient_attention_xformers(valid, attention_op)
  File "C:\Users\me\source\repos\stable-diffusion-webui\venv\lib\site-packages\diffusers\models\modeling_utils.py", line 227, in set_use_memory_efficient_attention_xformers
    fn_recursive_set_mem_eff(module)
  File "C:\Users\me\source\repos\stable-diffusion-webui\venv\lib\site-packages\diffusers\models\modeling_utils.py", line 223, in fn_recursive_set_mem_eff
    fn_recursive_set_mem_eff(child)
  File "C:\Users\me\source\repos\stable-diffusion-webui\venv\lib\site-packages\diffusers\models\modeling_utils.py", line 223, in fn_recursive_set_mem_eff
    fn_recursive_set_mem_eff(child)
  File "C:\Users\me\source\repos\stable-diffusion-webui\venv\lib\site-packages\diffusers\models\modeling_utils.py", line 223, in fn_recursive_set_mem_eff
    fn_recursive_set_mem_eff(child)
  File "C:\Users\me\source\repos\stable-diffusion-webui\venv\lib\site-packages\diffusers\models\modeling_utils.py", line 220, in fn_recursive_set_mem_eff
    module.set_use_memory_efficient_attention_xformers(valid, attention_op)
  File "C:\Users\me\source\repos\stable-diffusion-webui\venv\lib\site-packages\diffusers\models\modeling_utils.py", line 227, in set_use_memory_efficient_attention_xformers
    fn_recursive_set_mem_eff(module)
  File "C:\Users\me\source\repos\stable-diffusion-webui\venv\lib\site-packages\diffusers\models\modeling_utils.py", line 223, in fn_recursive_set_mem_eff
    fn_recursive_set_mem_eff(child)
  File "C:\Users\me\source\repos\stable-diffusion-webui\venv\lib\site-packages\diffusers\models\modeling_utils.py", line 223, in fn_recursive_set_mem_eff
    fn_recursive_set_mem_eff(child)
  File "C:\Users\me\source\repos\stable-diffusion-webui\venv\lib\site-packages\diffusers\models\modeling_utils.py", line 220, in fn_recursive_set_mem_eff
    module.set_use_memory_efficient_attention_xformers(valid, attention_op)
  File "C:\Users\me\source\repos\stable-diffusion-webui\venv\lib\site-packages\diffusers\models\attention_processor.py", line 193, in set_use_memory_efficient_attention_xformers
    raise ModuleNotFoundError(
ModuleNotFoundError: Refer to https://github.com/facebookresearch/xformers for more information on how to install xformers

[OsmondFan]

Are you also using macOS?

[boclifton-MSFT]

Nope, Windows 11.

[biollas]

Can't find xformers in newest downloads either. xformers: 0.0.20 reported on bottom of page, but Memory Attention has nothing other than default.