!!! Exception during processing!!! 'downsampler_block' object has no attribute 'lora_layer'

[patdelphi]

got error msg as title

[florestefano1975]

With which node do you have this error?

[Omarsmsm]

Here is the error 
Error occurred when executing HiDiffusionSDXL:

'downsampler_block' object has no attribute 'lora_layer'

File "C:\Users\\Downloads\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\execution.py", line 151, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\\Downloads\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\execution.py", line 81, in get_output_data
return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\\Downloads\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\execution.py", line 74, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\\Downloads\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HiDiffusion\__init__.py", line 167, in hi_diff_sdxl
image = pipe(prompt=positive_prompt, num_inference_steps=steps, guidance_scale=guidance_scale, height=height, width=width, eta=eta, negative_prompt=negative_prompt).images[0]
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\\Downloads\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\utils\_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\\Downloads\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\diffusers\pipelines\stable_diffusion_xl\pipeline_stable_diffusion_xl.py", line 1241, in __call__
noise_pred = self.unet(
^^^^^^^^^^
File "C:\Users\\Downloads\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1511, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\\Downloads\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1561, in _call_impl
result = forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\\Downloads\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\accelerate\hooks.py", line 165, in new_forward
output = module._old_forward(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\\Downloads\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\diffusers\models\unets\unet_2d_condition.py", line 1121, in forward
sample, res_samples = downsample_block(
^^^^^^^^^^^^^^^^^
File "C:\Users\\Downloads\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1511, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\\Downloads\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1520, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\\Downloads\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HiDiffusion\hidiffusion.py", line 1457, in forward
hidden_states = downsampler(hidden_states)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\\Downloads\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1511, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\\Downloads\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1520, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\\Downloads\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\diffusers\models\downsampling.py", line 152, in forward
hidden_states = self.conv(hidden_states)
^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\\Downloads\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1511, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\\Downloads\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1520, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\\Downloads\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HiDiffusion\hidiffusion.py", line 1672, in forward
if self.lora_layer is None:
^^^^^^^^^^^^^^^
File "C:\Users\\Downloads\ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1688, in __getattr__
raise AttributeError(f"'{type(self).__name__}' object has no attribute '{name}'")

I believe its a torch version issue but I am not sure

[Omarsmsm]

Seems to happen when using Resolution Aware Unet (RAUNET) parameter ON.

[AugmentedRealityCat]

Same problem.

Error occurred when executing HiDiffusionSDXL:

'downsampler_block' object has no attribute 'lora_layer'

  File "D:\ComfyUI_windows_portable\ComfyUI\execution.py", line 151, in recursive_execute
    output_data, output_ui = get_output_data(obj, input_data_all)
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\ComfyUI\execution.py", line 81, in get_output_data
    return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\ComfyUI\execution.py", line 74, in map_node_over_list
    results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HiDiffusion\__init__.py", line 167, in hi_diff_sdxl
    image = pipe(prompt=positive_prompt, num_inference_steps=steps, guidance_scale=guidance_scale, height=height, width=width, eta=eta, negative_prompt=negative_prompt).images[0]
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\utils\_contextlib.py", line 115, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\diffusers\pipelines\stable_diffusion_xl\pipeline_stable_diffusion_xl.py", line 1241, in __call__
    noise_pred = self.unet(
                 ^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1568, in _call_impl
    result = forward_call(*args, **kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\accelerate\hooks.py", line 165, in new_forward
    output = module._old_forward(*args, **kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\diffusers\models\unets\unet_2d_condition.py", line 1121, in forward
    sample, res_samples = downsample_block(
                          ^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HiDiffusion\hidiffusion.py", line 1457, in forward
    hidden_states = downsampler(hidden_states)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\diffusers\models\downsampling.py", line 152, in forward
    hidden_states = self.conv(hidden_states)
                    ^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HiDiffusion\hidiffusion.py", line 1672, in forward
    if self.lora_layer is None:
       ^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1695, in __getattr__
    raise AttributeError(f"'{type(self).__name__}' object has no attribute '{name}'")

[florestefano1975]

Checkpoint used?

[AugmentedRealityCat]

Checkpoint used?

SDXL Base 1.0 + 0.9 VAE (official release)

Seems to happen when using Resolution Aware Unet (RAUNET) parameter ON.

I have observed the same thing. If I turn off RAUNET there is no error message.

[patdelphi]

Using the new installed HiDiffusion SDXL node, and the SDXL base 1.0 checkpoint, then got above error msg.

[AugmentedRealityCat]

I will make a test by removing all the other custom nodes and I'll come back with a report. It might be an incompatibility - I've had quite a few of those recently.

[AugmentedRealityCat]

I made a test with just HiDiffusion as a custom node (I removed all the others) and I still get what I think is the exact same error message. I am copying the one from the console window below just in case there might be a clue hidden in there.

As you can see, it goes through a whole image generation process (50 steps) before stopping just after a second one is called.

got prompt
100%|██████████████████████████████████████████████████████████████████████████████████| 50/50 [00:27<00:00,  1.84it/s]
  0%|                                                                                           | 0/50 [00:01<?, ?it/s]
!!! Exception during processing!!! 'downsampler_block' object has no attribute 'lora_layer'
Traceback (most recent call last):
  File "D:\ComfyUI_windows_portable\ComfyUI\execution.py", line 151, in recursive_execute
    output_data, output_ui = get_output_data(obj, input_data_all)
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\ComfyUI\execution.py", line 81, in get_output_data
    return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\ComfyUI\execution.py", line 74, in map_node_over_list
    results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HiDiffusion\__init__.py", line 167, in hi_diff_sdxl
    image = pipe(prompt=positive_prompt, num_inference_steps=steps, guidance_scale=guidance_scale, height=height, width=width, eta=eta, negative_prompt=negative_prompt).images[0]
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\utils\_contextlib.py", line 115, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\diffusers\pipelines\stable_diffusion_xl\pipeline_stable_diffusion_xl.py", line 1241, in __call__
    noise_pred = self.unet(
                 ^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1568, in _call_impl
    result = forward_call(*args, **kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\accelerate\hooks.py", line 165, in new_forward
    output = module._old_forward(*args, **kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\diffusers\models\unets\unet_2d_condition.py", line 1121, in forward
    sample, res_samples = downsample_block(
                          ^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HiDiffusion\hidiffusion.py", line 1457, in forward
    hidden_states = downsampler(hidden_states)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\diffusers\models\downsampling.py", line 152, in forward
    hidden_states = self.conv(hidden_states)
                    ^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HiDiffusion\hidiffusion.py", line 1672, in forward
    if self.lora_layer is None:
       ^^^^^^^^^^^^^^^
  File "D:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1695, in __getattr__
    raise AttributeError(f"'{type(self).__name__}' object has no attribute '{name}'")
AttributeError: 'downsampler_block' object has no attribute 'lora_layer'

Prompt executed in 50.24 seconds

[AugmentedRealityCat]

My workflow (as simple as it gets !)

[florestefano1975]

My workflow (as simple as it gets !)

Maybe it's due to the checkpoint? Have you tried with SDXL1.0 without vae 0.9?

[AugmentedRealityCat]

I do not actually have SDXL1.0 without vae 0.9, so I'm going to download it and I'll test it.

Is there any model you are using for your own tests ?

[florestefano1975]

I do not actually have SDXL1.0 without vae 0.9, so I'm going to download it and I'll test it.

Is there any model you are using for your own tests ?

I run tests with the base 1.0 version and with other models trained on it. In the original Hidiffusion repository they specify that it works this way.

[AugmentedRealityCat]

I checked with base SDXL 1.0 (sd_xl_base_1.0.safetensors from this page https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0/tree/main ) and I still get the same problem and the same error message.

It does produce an image if I turn RAUNET off, but you can see that it's not the image right image.

But the minute RAUNET is set to ON, you get no image at all - just the error message previously logged by myself and others in this thread.

Anything else I can do to help you with this ?

[florestefano1975]

RAUNET is essential to use HiDiffusion, its exclusion completely disables generative control. For the moment I will insert a notice in the repository, to report possible bugs and malfunctions. In the meantime, thank you very much for the tests and your availability. I will investigate these issues.

[AlUlkesh]

Installing the current diffusers package got rid of this error for me.