Kosinkadink / ComfyUI-Advanced-ControlNet

ControlNet scheduling and masking nodes with sliding context support
GNU General Public License v3.0
622 stars 60 forks source link

Error occurred when executing ControlNetLoaderAdvanced #78

Closed yokingma closed 6 months ago

yokingma commented 8 months ago

Hi, I got an error today, but it's fine yesterday, and I don't know where the problem is. today I upgraded compyui, then an error was reported.

Error occurred when executing ControlNetLoaderAdvanced:

T2IAdapter.__init__() missing 2 required positional arguments: 'compression_ratio' and 'upscale_algorithm'

File "H:\AI\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\execution.py", line 152, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "H:\AI\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\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 "H:\AI\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\execution.py", line 75, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "H:\AI\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Advanced-ControlNet\adv_control\nodes.py", line 90, in load_controlnet
controlnet = load_controlnet(controlnet_path, timestep_keyframe)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "H:\AI\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Advanced-ControlNet\adv_control\control.py", line 555, in load_controlnet
return convert_to_advanced(control, timestep_keyframe=timestep_keyframe)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "H:\AI\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Advanced-ControlNet\adv_control\control.py", line 570, in convert_to_advanced
return T2IAdapterAdvanced.from_vanilla(v=control, timestep_keyframe=timestep_keyframe)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "H:\AI\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Advanced-ControlNet\adv_control\control.py", line 141, in from_vanilla
return T2IAdapterAdvanced(t2i_model=v.t2i_model, timestep_keyframes=timestep_keyframe, channels_in=v.channels_in, device=v.device)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "H:\AI\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Advanced-ControlNet\adv_control\control.py", line 95, in __init__
super().__init__(t2i_model=t2i_model, channels_in=channels_in, device=device)

screenshot

NopenahnaddaGH commented 8 months ago

Same error after fat fingering an update all yesterday. I'm trying to use coadapter depth. Was working flawlessly for a couple months previously.

Kosinkadink commented 8 months ago

Looks like there was an update for ComfyUI I'll need to account for to make T2IAdapter work (sd-cascade CN support was added to comfy). I should have that ready in an hour or so.

Kosinkadink commented 8 months ago

I've pushed an update to match ComfyUI changes. It is not backwards compatible - most recent ComfyUI versions are required. Let me know if it works for you now, from my local testing it works.

NopenahnaddaGH commented 8 months ago

It'll be a good few hours before I'm back at my machine. Hopefully someone can confirm before then. Thanks for your excellent work regardless.

yokingma commented 8 months ago

@Kosinkadink it's working very well now 👍 , and thank you so so so much :).

TharindaMarasingha commented 4 days ago

@Kosinkadink it's working very well now! Thank you mate