ffxvs / sd-webui-complete-setup

Jupyter notebook for Stable Diffusion Web UI and Stable Diffusion Web UI Forge.
https://ffxvs.github.io/sd-webui-complete-setup/
MIT License
33 stars 3 forks source link

About Regional Prompt #23

Open AtemSchiffer opened 3 weeks ago

AtemSchiffer commented 3 weeks ago

Hi, this new version of stable diffusion seems to have problems with the installation of the regional prompt, here is the line of the problem. I would like to ask, is there a way to get the previous versions installed? I notice that there are also problems with the sound notification at the end and you can notice a bit of lag in the interface. Maybe it is the new version and not the paper. Thank you very much for everything!

⏳ Preparing... It will take a little longer... Launching WebUI... Total VRAM 16109 MB, total RAM 45140 MB pytorch version: 2.3.1+cu121 xformers version: 0.0.27 Set vram state to: NORMAL_VRAM Device: cuda:0 NVIDIA RTX A4000 : native Hint: your device supports --cuda-malloc for potential speed improvements. VAE dtype preferences: [torch.bfloat16, torch.float32] -> torch.bfloat16 CUDA Using Stream: False Using xformers cross attention Using xformers attention for VAE ControlNet preprocessor location: /notebooks/stable-diffusion-webui-forge/models/ControlNetPreprocessor Tag Autocomplete: Could not locate model-keyword extension, Lora trigger word completion will be limited to those added through the extra networks menu. CivitAI Browser+: Aria2 RPC started *** Error loading script: attention.py Traceback (most recent call last): File "/notebooks/stable-diffusion-webui-forge/modules/scripts.py", line 525, in load_scripts script_module = script_loading.load_module(scriptfile.path) File "/notebooks/stable-diffusion-webui-forge/modules/script_loading.py", line 13, in load_module module_spec.loader.exec_module(module) File "", line 883, in exec_module File "", line 241, in _call_with_frames_removed File "/notebooks/stable-diffusion-webui-forge/extensions/sd-webui-regional-prompter/scripts/attention.py", line 3, in import ldm.modules.attention as atm ModuleNotFoundError: No module named 'ldm'


*** Error loading script: latent.py Traceback (most recent call last): File "/notebooks/stable-diffusion-webui-forge/modules/scripts.py", line 525, in load_scripts script_module = script_loading.load_module(scriptfile.path) File "/notebooks/stable-diffusion-webui-forge/modules/script_loading.py", line 13, in load_module module_spec.loader.exec_module(module) File "", line 883, in exec_module File "", line 241, in _call_with_frames_removed File "/notebooks/stable-diffusion-webui-forge/extensions/sd-webui-regional-prompter/scripts/latent.py", line 11, in import scripts.attention as att File "/notebooks/stable-diffusion-webui-forge/extensions/sd-webui-regional-prompter/scripts/attention.py", line 3, in import ldm.modules.attention as atm ModuleNotFoundError: No module named 'ldm'


*** Error loading script: rp.py Traceback (most recent call last): File "/notebooks/stable-diffusion-webui-forge/modules/scripts.py", line 525, in load_scripts script_module = script_loading.load_module(scriptfile.path) File "/notebooks/stable-diffusion-webui-forge/modules/script_loading.py", line 13, in load_module module_spec.loader.exec_module(module) File "", line 883, in exec_module File "", line 241, in _call_with_frames_removed File "/notebooks/stable-diffusion-webui-forge/extensions/sd-webui-regional-prompter/scripts/rp.py", line 15, in import scripts.attention File "/notebooks/stable-diffusion-webui-forge/extensions/sd-webui-regional-prompter/scripts/attention.py", line 3, in import ldm.modules.attention as atm ModuleNotFoundError: No module named 'ldm'

ffxvs commented 3 weeks ago

Hi, since the new Forge, Regional prompter extension doesn't work anymore.
Although, someone has made an fork out of it, not all features work. You can remove the old extension and try this fork.

https://github.com/jessearodriguez/sd-forge-regional-prompter

This is an attempt to port over the regional prompting webui extension over to forge. At the moment, only attention mode is usable. In addition, I have only tested sdxl models, please be aware that there is likely to be issues with other formats. Latent mode seems to generate without any console errors, but results in fused concepts. It will probably remain this way until someone more knowledgable than me fixes/improves/rewrites that code chunk in the future. Assume any posted issues on the main extension page apply to this one.

There is also an alternative to Regional Prompter called Attention Couple which has similar functionality. You can install this extension from Extension list in the notebook.

I don't plan on supporting previous version of Forge as there would be a lot of compatibility issues to take into account. If you really need that one extension, I recommend sticking with the original Auto1111's WebUI.


I notice that there are also problems with the sound notification at the end and you can notice a bit of lag in the interface.

I didn't experience any of this. Maybe this is due to a broken Regional prompter extension.
Please let me know if the two extensions above can solve your problem! 👍