chaojie / comfyui-moore-animateanyone Goto Github PK

Hi Chaojie,

I like your works a lot and want to continue use them with latest ComfyUI. Would you please upgrade your nodes to support diffusers 0.27.2 at least. So far, your nodes only work if diffusers is around 0.18. Thank you very much. Look forward to hear good news from you.

Error no file named config.json found in directory when using DWPose in UNet 2d and 3D nodes...

Error occurred when executing Moore-AnimateAnyone Reference Unet:

Error no file named config.json found in directory F:\SDXL\ComfyUI_windows_portable\ComfyUI/custom_nodes/ComfyUI-Moore-AnimateAnyone/pretrained_weights\DWPose.

File "F:\SDXL\ComfyUI_windows_portable\ComfyUI\execution.py", line 155, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "F:\SDXL\ComfyUI_windows_portable\ComfyUI\execution.py", line 85, in get_output_data
return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "F:\SDXL\ComfyUI_windows_portable\ComfyUI\execution.py", line 78, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "F:\SDXL\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Moore-AnimateAnyone\nodes.py", line 94, in run_inference
reference_unet = UNet2DConditionModel.from_pretrained(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "F:\SDXL\ComfyUI_windows_portable\python_embeded\Lib\site-packages\diffusers\models\modeling_utils.py", line 712, in from_pretrained
config, unused_kwargs, commit_hash = cls.load_config(
^^^^^^^^^^^^^^^^
File "F:\SDXL\ComfyUI_windows_portable\python_embeded\Lib\site-packages\diffusers\configuration_utils.py", line 365, in load_config
raise EnvironmentError(

I managed to run it in 6GB VRAM! Low res and low framerate of course. And I haven’t gotten anything remotely good so far, but I’ll keep testing. I’m getting short results in 5 minutes, I see this as a win.

Thanks for this, I was thinking of making the custom node myself but couldn’t even make the original project work x).

Several things to note:

Width and height should use steps of 8. Using anything else gives an error saying “expected (X) resolution but received (Y) instead.”

Users should make sure the pretrained_models folder is in the root of the custom node. The way you wrote the code, the download path is relative to the launcher’s location. And that’s a different place between the one-click install and the default install.
So guys, remember: pretrained_models must be next to tools, NOT inside.

What is “length”? I’m pretty sure it’s the number of frames, but it’s unclear by default ^^’.
Finally, in your read me, you write: “cd ComfyUI/custom_nodes/ComfyUI-Moore-AnimateAnyone & python tools/download_weights.py”.
But the folder is “tools”, not “python tools”. And the & is a typo I assume, it should be a slash. A minor issue, but it could get confusing ^^.

Anyway, thanks for your work!

When loading the graph, the following node types were not found:
Moore-AnimateAnyone Image Encoder
Moore-AnimateAnyone Reference Unet
Moore-AnimateAnyone Denoising Unet
Moore-AnimateAnyone Pose Guider
VHS_LoadVideo
Moore-AnimateAnyone Pipeline Loader
VHS_VideoCombine
Moore-AnimateAnyone Simple
Moore-AnimateAnyone VAE
Nodes that have failed to load will show as red on the graph.

Hello,
I am fairly new to the ComfyUI and AnimateAnyone.
I was trying to see if I can repose some drawing of character with another pose image, but I only got noisy output.
I wonder if I am getting error because I am inputing single image for openpose input instead of video.
Is there way to use it for just image instead of video?

attached is my failed workflow,

Traceback (most recent call last):
File "/mnt/data/wangxi/Comfyui/nodes.py", line 1872, in load_custom_node
module_spec.loader.exec_module(module)
File "", line 883, in exec_module
File "", line 241, in _call_with_frames_removed
File "/mnt/data/wangxi/Comfyui/custom_nodes/ComfyUI-Moore-AnimateAnyone/init.py", line 1, in
from .nodes import NODE_CLASS_MAPPINGS
File "/mnt/data/wangxi/Comfyui/custom_nodes/ComfyUI-Moore-AnimateAnyone/nodes.py", line 24, in
from src.pipelines.pipeline_pose2vid_long import Pose2VideoPipeline
File "/mnt/data/wangxi/Comfyui/custom_nodes/ComfyUI-Moore-AnimateAnyone/src/pipelines/pipeline_pose2vid_long.py", line 23, in
from transformers import CLIPImageProcessor
File "", line 1075, in _handle_fromlist
File "/home/wangxi/miniconda3/envs/comfyui/lib/python3.10/site-packages/transformers/utils/import_utils.py", line 1077, in getattr
value = getattr(module, name)
File "/home/wangxi/miniconda3/envs/comfyui/lib/python3.10/site-packages/transformers/utils/import_utils.py", line 1076, in getattr
module = self._get_module(self._class_to_module[name])
File "/home/wangxi/miniconda3/envs/comfyui/lib/python3.10/site-packages/transformers/utils/import_utils.py", line 1088, in _get_module
raise RuntimeError(
RuntimeError: Failed to import transformers.models.clip.image_processing_clip because of the following error (look up to see its traceback):
cannot import name 'builder' from 'google.protobuf.internal' (/home/wangxi/miniconda3/envs/comfyui/lib/python3.10/site-packages/google/protobuf/internal/init.py)

Cannot import /mnt/data/wangxi/Comfyui/custom_nodes/ComfyUI-Moore-AnimateAnyone module for custom nodes: Failed to import transformers.models.clip.image_processing_clip because of the following error (look up to see its traceback):
cannot import name 'builder' from 'google.protobuf.internal' (/home/wangxi/miniconda3/envs/comfyui/lib/python3.10/site-packages/google/protobuf/internal/init.py)

Import times for custom nodes:
0.5 seconds (IMPORT FAILED): /mnt/data/wangxi/Comfyui/custom_nodes/ComfyUI-Moore-AnimateAnyone

Starting server

To see the GUI go to: http://0.0.0.0:7864

I have another set of errors:

Traceback (most recent call last):
File "G:\ComfyUI_windows_portable\ComfyUI\nodes.py", line 1872, in load_custom_node
module_spec.loader.exec_module(module)
File "", line 940, in exec_module
File "", line 241, in call_with_frames_removed
File "G:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Moore-AnimateAnyone_init.py", line 1, in
from .nodes import NODE_CLASS_MAPPINGS
File "G:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Moore-AnimateAnyone\nodes.py", line 22, in
from src.models.unet_2d_condition import UNet2DConditionModel
File "G:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Moore-AnimateAnyone\src\models\unet_2d_condition.py", line 31, in
from diffusers.utils import (
ImportError: cannot import name 'USE_PEFT_BACKEND' from 'diffusers.utils' (G:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\diffusers\utils_init_.py)

Cannot import G:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Moore-AnimateAnyone module for custom nodes: cannot import name 'USE_PEFT_BACKEND' from 'diffusers.utils' (G:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\diffusers\utils_init_.py)

I removed the node and installed via git directly and downloaded all the required models, still have the same issue...

followed your instructions, but

Hi,everyone, I find that the download_weights.py will download many weights from huggingface, could anyone tell me how much disk space must be used?Thanks for all.

Installed PhotoMaker，Animateanyone will not work. Re-installed Animateanyone ,the photomaker will stop work too.
Two interesting plugin, only one can be chosen.

可以不改我的pytorch版本吗?
我安装了你好几个项目,都是改系统的pytorch为2.0.1+cu117

My diffusers version:

Name: diffusers
Version: 0.25.1
Summary: State-of-the-art diffusion in PyTorch and JAX.
Home-page: https://github.com/huggingface/diffusers
Author: The Hugging Face team (past and future) with the help of all our contributors (https://github.com/huggingface/diffusers/graphs/contributors)
Author-email: [email protected]
License: Apache 2.0 License
Location: /root/miniconda3/envs/comfy/lib/python3.10/site-packages
Requires: filelock, huggingface-hub, importlib-metadata, numpy, Pillow, regex, requests, safetensors
Required-by: compel, lama-cleaner, lycoris-lora

After restarting ComfyUI, the node import failed, which used to be fine.
Cannot import E:\DEV\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Moore-AnimateAnyone module for custom nodes: cannot import name 'PositionNet' from 'diffusers.models.embeddings' (E:\DEV\ComfyUI_windows_portable\python_embeded\lib\site-packages\diffusers\models\embeddings.py)

Any solution for that? Update or downgrade diffusers?

I obtain this result using ComfyUI with the example workflow in this repository.

However, when I try to use pose2vid.py in this repository, I can obtain a good result.
It seems very strange to me. I'm trying to modify the bug.

Anyway, has anyone else encountered the same problem?

After successfully git clone && install && run download_weights, I got this error when I start ComfyUI

ERROR:root:Traceback (most recent call last):
File "D:\ComfyUI\execution.py", line 152, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI\execution.py", line 82, in get_output_data
return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI\custom_nodes\ComfyUI-0246\utils.py", line 381, in new_func
res_value = old_func(*final_args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI\execution.py", line 75, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI\custom_nodes\ComfyUI-Moore-AnimateAnyone\nodes.py", line 94, in run_inference
reference_unet = UNet2DConditionModel.from_pretrained(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI.ext\Lib\site-packages\diffusers\models\modeling_utils.py", line 812, in from_pretrained
unexpected_keys = load_model_dict_into_meta(
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI.ext\Lib\site-packages\diffusers\models\modeling_utils.py", line 155, in load_model_dict_into_meta
raise ValueError(
ValueError: Cannot load D:\ComfyUI/custom_nodes/ComfyUI-Moore-AnimateAnyone/pretrained_weights\stable-diffusion-v1-5 because down_blocks.0.attentions.0.transformer_blocks.0.attn2.to_k.weight expected shape tensor(..., device='meta', size=(320, 1280)), but got torch.Size([320, 768]). If you want to instead overwrite randomly initialized weights, please make sure to pass both low_cpu_mem_usage=False and ignore_mismatched_sizes=True. For more information, see also: huggingface/diffusers#1619 (comment) as an example.