search

kijai / ComfyUI-SUPIR

SUPIR upscaling wrapper for ComfyUI
Other
1.2k stars 66 forks source link

RecursionError: maximum recursion depth exceeded while calling a Python object (once more) #112

Open HWiese1980 opened 2 months ago

HWiese1980 commented 2 months ago

Occasionally I'm getting the following error in the SUPIR Denoiser node (and others IIRC):

[Tiled VAE]: the input size is tiny and unnecessary to tile.
[Tiled VAE]: the input size is tiny and unnecessary to tile.
[Tiled VAE]: the input size is tiny and unnecessary to tile.
[Tiled VAE]: the input size is tiny and unnecessary to tile.
[Tiled VAE]: the input size is tiny and unnecessary to tile.
[Tiled VAE]: the input size is tiny and unnecessary to tile.
[Tiled VAE]: the input size is tiny and unnecessary to tile.
[Tiled VAE]: the input size is tiny and unnecessary to tile.
[Tiled VAE]: the input size is tiny and unnecessary to tile.
[Tiled VAE]: the input size is tiny and unnecessary to tile.
[Tiled VAE]: the input size is tiny and unnecessary to tile.
[Tiled VAE]: the input size is tiny and unnecessary to tile.
[Tiled VAE]: the input size is tiny and unnecessary to tile.
[Tiled VAE]: the input size is tiny and unnecessary to tile.
[Tiled VAE]: the input size is tiny and unnecessary to tile.
[Tiled VAE]: the input size is tiny and unnecessary to tile.
[Tiled VAE]: the input size is tiny and unnecessary to tile.
!!! Exception during processing!!! maximum recursion depth exceeded while calling a Python object
Traceback (most recent call last):
  File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-SUPIR\SUPIR\utils\tilevae.py", line 699, in __call__
    self.net.to(device)
  File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1152, in to
    return self._apply(convert)
           ^^^^^^^^^^^^^^^^^^^^
  File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 802, in _apply
    module._apply(fn)
  File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 802, in _apply
    module._apply(fn)
  File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 802, in _apply
    module._apply(fn)
  [Previous line repeated 2 more times]
  File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 801, in _apply
    for module in self.children():
  File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 2283, in children
    for name, module in self.named_children():
  File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 2301, in named_children
    for name, module in self._modules.items():
                        ^^^^^^^^^^^^^^^^^^^^^
RecursionError: maximum recursion depth exceeded while calling a Python object

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "E:\new_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 "E:\new_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 "E:\new_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 "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-SUPIR\nodes_v2.py", line 335, in process
    decoded_images = SUPIR_VAE.decode(sample).float()
                     ^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-SUPIR\sgm\models\autoencoder.py", line 318, in decode
    dec = self.decoder(z, **decoder_kwargs)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\new_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 "E:\new_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 "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-SUPIR\SUPIR\utils\tilevae.py", line 702, in __call__
    return self.net.original_forward(x)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-SUPIR\SUPIR\utils\tilevae.py", line 702, in __call__
    return self.net.original_forward(x)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-SUPIR\SUPIR\utils\tilevae.py", line 702, in __call__
    return self.net.original_forward(x)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  [Previous line repeated 483 more times]
  File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-SUPIR\SUPIR\utils\tilevae.py", line 706, in __call__
    self.net.to(original_device)
  File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1152, in to
    return self._apply(convert)
           ^^^^^^^^^^^^^^^^^^^^
  File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 802, in _apply
    module._apply(fn)
  File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 802, in _apply
    module._apply(fn)
  File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 802, in _apply
    module._apply(fn)
  [Previous line repeated 2 more times]
  File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 801, in _apply
    for module in self.children():
  File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 2283, in children
    for name, module in self.named_children():
  File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 2301, in named_children
    for name, module in self._modules.items():
                        ^^^^^^^^^^^^^^^^^^^^^
RecursionError: maximum recursion depth exceeded while calling a Python object
kijai commented 2 months ago

Some tiled vae thing, I don't really know, but based on the messages it can just be disabled as it's not doing anything anyway when the image size isn't big enough. In general it also shouldn't be used as long as you don't run out of memory on those nodes it's available on.

HWiese1980 commented 1 month ago

Sorry for getting back to you so late.

Disabling tile VAE does not change anything. I'm still occasionally encountering this error.

Here's another error message in this context (the one shown in the ComfyUI web UI): 

Error occurred when executing SUPIR_first_stage:

maximum recursion depth exceeded while calling a Python object

File "E:\new_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 "E:\new_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 "E:\new_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 "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-SUPIR\nodes_v2.py", line 331, in process
h = SUPIR_VAE.denoise_encoder(img.unsqueeze(0))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\new_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 "E:\new_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 "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-SUPIR\SUPIR\utils\tilevae.py", line 702, in __call__
return self.net.original_forward(x)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-SUPIR\SUPIR\utils\tilevae.py", line 702, in __call__
return self.net.original_forward(x)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-SUPIR\SUPIR\utils\tilevae.py", line 702, in __call__
return self.net.original_forward(x)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
[Previous line repeated 483 more times]
File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-SUPIR\SUPIR\utils\tilevae.py", line 706, in __call__
self.net.to(original_device)
File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1152, in to
return self._apply(convert)
^^^^^^^^^^^^^^^^^^^^
File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 802, in _apply
module._apply(fn)
File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 802, in _apply
module._apply(fn)
File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 802, in _apply
module._apply(fn)
[Previous line repeated 2 more times]
File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 824, in _apply
with torch.no_grad():
File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\autograd\grad_mode.py", line 81, in __enter__
torch.set_grad_enabled(False)
File "E:\new_ComfyUI_windows_portable_nvidia_cu121_or_cpu\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\autograd\grad_mode.py", line 184, in __init__
self.prev = torch.is_grad_enabled()
^^^^^^^^^^^^^^^^^^^^^^^
HWiese1980 commented 1 month ago

Reloading the SUPIR-model (by switching back and forth) helps solving this problem momentarily.