Hi! This is the support thread for the SDXL Customization Example Project on GitHub. Any major updates we push to the project will be announced here. Further, feel free to discuss, raise issues, and ask for assistance in this thread.
Please keep discussion in this thread project-related. Any issues with the Workbench application should be raised as a standalone thread. Thanks!
An update has been pushed to correct an issue with the PEFT package dependency. Please update your repos accordingly, and let us know of any additional issues. Thanks!
Found this error in the Notebook when trying to run the King Charles section:
The lora-weights file is missing in the Model/tuned-charles-3 folder. I copied the one from tuned-toy-jensen into it.
One more issue - when the current FineTuning-SDXL code is updated, it exceeds the GitHub file size limit of 5MB. I only have the choice to move or delete the file once that happens, as Workbench throws an exception that does not allow me to commit files back to GitHub.
Hey Brian, thanks for reaching out. Thatâs strange. Usually the model/tuned-charles-3 folder is populated with the weights of the fine-tuned model after running the fine-tuning cell. Perhaps you forgot to run a cell or populate the charles-3 folder with images for the model to fine-tune on? Not exactly sure. Try restarting the kernel and running the cells again, and let me know if the issue persists.
I believe the file size comes from the images that get generated/displayed in-notebook. I believe if you go to kernel â clear outputs, the images get removed from the notebook and that reduces the file size enough to commit the notebook to Github. The fine-tuned model weights will still be in the models folder for future useâyou can open the notebook and run the right cells again to generate images at a later time. Hope this helps!
Thanks - I ran the script as written, but it may be that the code was missing a line to generate this. Iâve pointed this out to our NVIDIA resource and he is re-writing the code.
Restarting from scratch. NVWB is installed and working properly. I cloned the SDXL project from GitHub. I am logged into both GitHub and Huggingface. My internet connection is good. When I cloned the SDXL project from GitHub, the setup begins and end then Crashes with Error connecting to local. I have attached a copy of my log files. How do I fix this?
nvwblogs_02152024.txt (8.4 KB)
Hi Mark
Were you using the CLI for this?
I note the log: {"level":"error","isWrapped":true,"isInteractive":true,"engine":"markdown","output":"**you must provide a context name when calling 'nvwb activate <context_name>'**","time":"2024-01-18T18:53:28-05:00","message":"you must provide a context name when calling 'nvwb activate <context_name>'"}
This suggests you ran nvwb activate without a location argument, i.e. nvwb activate local.
Hi, No CLI, I am doing this on the NVWB installed in Windows WSLX env. Just leveraging the GUI to do everything.
(6/4/2024) Project spec has been updated for AI Workbench version 0.50.17.
(9/3/2024)
Added a basic frontend UI for image generation. The user can now generate and visualize images from the stable diffusion XL base model, as well as across all finetuned models the user has created in this project. Generated images will be saved to the project.
In the top post of the support threadsâŠ
Can you add links to the NVIDIA site announcing the project and to the GIT project itself?
Thanks for the catch. Github link is added now. This project was published back in March for the general availability of AI Workbench. (blog)
Packages have been updated. Fixes an issue in the inference application that prevents images from rendering in the UI. Hopefully the inference app is functional now, but feel free to ping this thread with issues.
When I try the "SDXL Customizationâ example I met below error, any one can help me? thank you.
test2-error-log.txt (22.3 KB)
Hi @Alex.Y thanks for sending the error logs. The team is taking a look into this and will get back to you asap! Thanks for your patience,
Sophie
Hi Alex,
Thanks for reaching out. From the text in the log file you sent, it looks like the connection just dropped midway through the download.
LocalEntryNotFoundError: An error happened while trying to locate the file on the Hub and we cannot find the requested files in the local cache. Please check your connection and try again or make sure your Internet connection is on.
Can you make sure you have a stronger connection and try again? Thanks!
Hi edwli,
I am using the VPN to access the internet, and i checked to access the hugging face and github web site with web browser and found successfully, and i retried many times to run the âFineTuning-SDXL.ipynbâ, each time had the same error(stop at 5% process).
Do you have any suggestion which website I need more check?
Or is there any method that i can download the files from the web browser directly and put the filed into the project?
Thank you.
I tried reproducing your issue on my end and ended up running into another error:
ImportError: cannot import name âDDUFEntryâ from âhuggingface_hubâ (/home/workbench/.local/lib/python3.10/site-packages/huggingface_hub/init.py)
Looks like this stems from a huggingface_hub package being outdated. I upgraded huggingface_hub[cli,torch]==0.27.1 and downgraded numpy==1.23.1 in requirements.txt. This seemed to do the trick for me.
I would push these changes upstream but it appears I have lost edit access to the repo temporarily. Working on it.
I will note that once I upgraded the above package versions, I did not face any issue connecting to the Hugging Face Hub for model download. Perhaps try that?
