search

MrForExample / ComfyUI-AnimateAnyone-Evolved

Improved AnimateAnyone implementation that allows you to use the opse image sequence and reference image to generate stylized video
MIT License
442 stars 41 forks source link

Import Failed error message on ComfyUI startup post install of ComfyUI-AnimateAnyone-Evolved #32

Open bondo01 opened 4 months ago

bondo01 commented 4 months ago

I followed the instructions in the workflow for "Animate Anyone Raw" workflow. Downloaded the files: denoising_unet.pth, motion_module.pth, pose_guider.pth, reference_unet.pth created the folders: ComfyUI\custom_nodes\ComfyUI-AnimateAnyone-Evolved\pretrained_weights and put the 4 PyTorch files into the pretrained_weights sub-folder. Download Pytorch_model.bin renamed it to Clip_Vision_Animate_Anyone.bin and put it in the ComfyUI\models\clip_vision folder. restarted ComfyUI (Nvidia bat file) and get this error message:

0.0 seconds (IMPORT FAILED): D:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-AnimateAnyone-Evolve

is there something I missed?

ffiras commented 4 months ago

Same here

ffiras commented 4 months ago

I was using the portable version . You should install it using comfyui python embeded

cd pathToComfy\ComfyUI_windows_portable\python_embeded

python.exe -m pip install -r pathToComfyI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-AnimateAnyone-Evolved\requirements.txt

If you're using a premade workflow (.json or an image) and the nodes still red just delete them and replace them from node menu.

alexdoroga commented 1 month ago

I was using the portable version . You should install it using comfyui python embeded

cd pathToComfy\ComfyUI_windows_portable\python_embeded

python.exe -m pip install -r pathToComfyI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-AnimateAnyone-Evolved\requirements.txt

If you're using a premade workflow (.json or an image) and the nodes still red just delete them and replace them from node menu.

Woohoo! Awesome! Thank you) You saved me!!