Getting black image as output everytime.

[AB00k]

I'm trying to run the workflow and it's generating the complete black image, I even tried to run it on CPU still had this issue. Bellow is the log on cmd when I run the workflow:

E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable>.\python_embeded\python.exe -s ComfyUI\main.py --windows-standalone-build
## ComfyUI-Manager: installing dependencies done.
** ComfyUI startup time: 2024-05-17 15:08:48.897653
** Platform: Windows
** Python version: 3.11.8 (tags/v3.11.8:db85d51, Feb  6 2024, 22:03:32) [MSC v.1937 64 bit (AMD64)]
** Python executable: E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\python.exe
** Log path: E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\comfyui.log

Prestartup times for custom nodes:
   0.3 seconds: E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager

Total VRAM 6144 MB, total RAM 16150 MB
Set vram state to: NORMAL_VRAM
Device: cuda:0 NVIDIA GeForce GTX 1660 SUPER : cudaMallocAsync
VAE dtype: torch.float32
Using pytorch cross attention
### Loading: ComfyUI-Manager (V2.34)
### ComfyUI Revision: 2178 [ece5acb8] | Released on '2024-05-12'
E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\diffusers\utils\outputs.py:63: UserWarning: torch.utils._pytree._register_pytree_node is deprecated. Please use torch.utils._pytree.register_pytree_node instead.
  torch.utils._pytree._register_pytree_node(

Import times for custom nodes:
   0.0 seconds: E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\websocket_image_save.py
   0.2 seconds: E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_MagicClothing
   0.3 seconds: E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager

Starting server

To see the GUI go to: http://127.0.0.1:8188
FETCH DATA from: E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager\extension-node-map.json [DONE]
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/alter-list.json
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/model-list.json
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/extension-node-map.json
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json
got prompt
E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\huggingface_hub\file_download.py:1132: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.
  warnings.warn(
text_encoder\model.safetensors not found
Loading pipeline components...:  57%|█████████████████████████████▋                      | 4/7 [00:01<00:01,  2.27it/s]E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\clip\feature_extraction_clip.py:28: FutureWarning: The class CLIPFeatureExtractor is deprecated and will be removed in version 5 of Transformers. Please use CLIPImageProcessor instead.
  warnings.warn(
Loading pipeline components...: 100%|████████████████████████████████████████████████████| 7/7 [00:06<00:00,  1.12it/s]
----checkpoints loaded from path: E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_MagicClothing\checkpoints\cloth_segm.pth----
E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\functional.py:3809: UserWarning: nn.functional.upsample is deprecated. Use nn.functional.interpolate instead.
  warnings.warn("nn.functional.upsample is deprecated. Use nn.functional.interpolate instead.")
E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\diffusers\models\attention_processor.py:1244: UserWarning: 1Torch was not compiled with flash attention. (Triggered internally at ..\aten\src\ATen\native\transformers\cuda\sdp_utils.cpp:455.)
  hidden_states = F.scaled_dot_product_attention(
100%|██████████████████████████████████████████████████████████████████████████████████| 20/20 [04:03<00:00, 12.18s/it]
E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\conv.py:456: UserWarning: Plan failed with a cudnnException: CUDNN_BACKEND_EXECUTION_PLAN_DESCRIPTOR: cudnnFinalize Descriptor Failed cudnn_status: CUDNN_STATUS_NOT_SUPPORTED (Triggered internally at ..\aten\src\ATen\native\cudnn\Conv_v8.cpp:919.)
  return F.conv2d(input, weight, bias, self.stride,
E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\diffusers\image_processor.py:90: RuntimeWarning: invalid value encountered in cast
  images = (images * 255).round().astype("uint8")
Potential NSFW content was detected in one or more images. A black image will be returned instead. Try again with a different prompt and/or seed.
Prompt executed in 277.47 seconds

Here are the parameters I'm using:

[cardosofelipe]

Basically, in the pipeline there is an NSFW check, that sometimes detect false positives. Maybe it could be useful to add it as parameter of the node to skip or not the check. If this could be an interesting thing I could open a pull request in the next days.

[frankchieng]

you can edit pipelines/OmsDiffusionPipeline.py can comment out this line: image, has_nsfw_concept = self.run_safety_checker(image, device, prompt_embeds.dtype) with has_nsfw_concept = None instead

[AB00k]

@frankchieng I have tried it out but I got another error mentioned bellow.

E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\huggingface_hub\file_download.py:1132: FutureWarning:resume_downloadis deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, useforce_download=True. warnings.warn( safety_checker\model.safetensors not found Loading pipeline components...: 0%| | 0/7 [00:00<?, ?it/s]E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\clip\feature_extraction_clip.py:28: FutureWarning: The class CLIPFeatureExtractor is deprecated and will be removed in version 5 of Transformers. Please use CLIPImageProcessor instead. warnings.warn( Loading pipeline components...: 100%|████████████████████████████████████████████████████| 7/7 [00:08<00:00, 1.17s/it] ----checkpoints loaded from path: E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_MagicClothing\checkpoints\cloth_segm.pth---- E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\functional.py:3809: UserWarning: nn.functional.upsample is deprecated. Use nn.functional.interpolate instead. warnings.warn("nn.functional.upsample is deprecated. Use nn.functional.interpolate instead.") E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\diffusers\models\attention_processor.py:1244: UserWarning: 1Torch was not compiled with flash attention. (Triggered internally at ..\aten\src\ATen\native\transformers\cuda\sdp_utils.cpp:455.) hidden_states = F.scaled_dot_product_attention( 55%|█████████████████████████████████████████████ | 11/20 [02:13<01:49, 12.19s/it]FETCH DATA from: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/model-list.json [DONE] 100%|██████████████████████████████████████████████████████████████████████████████████| 20/20 [04:04<00:00, 12.21s/it] E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\diffusers\models\lora.py:358: UserWarning: Plan failed with an OutOfMemoryError: Allocation on device (Triggered internally at ..\aten\src\ATen\native\cudnn\Conv_v8.cpp:924.) return F.conv2d( E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\diffusers\models\lora.py:358: UserWarning: Plan failed with a cudnnException: CUDNN_BACKEND_EXECUTION_PLAN_DESCRIPTOR: cudnnFinalize Descriptor Failed cudnn_status: CUDNN_STATUS_NOT_SUPPORTED (Triggered internally at ..\aten\src\ATen\native\cudnn\Conv_v8.cpp:919.) return F.conv2d( E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\conv.py:456: UserWarning: Plan failed with a cudnnException: CUDNN_BACKEND_EXECUTION_PLAN_DESCRIPTOR: cudnnFinalize Descriptor Failed cudnn_status: CUDNN_STATUS_NOT_SUPPORTED (Triggered internally at ..\aten\src\ATen\native\cudnn\Conv_v8.cpp:919.) return F.conv2d(input, weight, bias, self.stride, E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\diffusers\image_processor.py:90: RuntimeWarning: invalid value encountered in cast images = (images * 255).round().astype("uint8") Prompt executed in 274.93 seconds

Even if the prompt is executed successfully it still generates black image:

got prompt safety_checker\model.safetensors not found Loading pipeline components...: 100%|████████████████████████████████████████████████████| 7/7 [00:07<00:00, 1.10s/it] ----checkpoints loaded from path: E:\magic_clothing\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_MagicClothing\checkpoints\cloth_segm.pth---- 100%|██████████████████████████████████████████████████████████████████████████████████| 20/20 [04:03<00:00, 12.18s/it] Prompt executed in 273.24 seconds