No module named ‘tqdm.auto’

[Muxelmann]

When installing sd-webui-text2video with the CUDA image I get the following error:

...
File "/app/stable-diffusion-webui/venv/lib/python3.11/site-packages/huggingface_hub/utils/tqdm.py", line 89, in ‹module>
    from tqdm.auto import tqdm as old tqdm
ModuleNotFoundError: No module named 'tqdm.auto'

It appears not to be related to the webui, and the issue is still open for sd-webui-text2video. They do suggest a workaround there (install a certain tqdm version), but I am not sure how to implement it with miniconda. Maybe the issue can be resolved with an updated container?

[siutin]

@Muxelmann You can just install the dependencies by running the container in an interactive shell. e.g:

docker run -it --name sdw --gpus all --network host \
  -v $(pwd)/models:/app/stable-diffusion-webui/models \
  -v $(pwd)/outputs:/app/stable-diffusion-webui/outputs \
  --rm siutin/stable-diffusion-webui-docker:latest-cuda \
  bash

source ./venv/bin/activate 
pip install xxx
pip install yyy

bash webui.sh --share 

if you prefer not to reinstall them each time you start a new container, you can simply make a custom image with your dependencies already installed.

[Muxelmann]

I will try this. However, my plan is to run the image on Unraid, so making a new image will probably be the way forward. I also tried making a new image based on a fork, but maybe it's more straight forward to simply base the new image on yours. If re-installing tqdm works, I'll post it here.