when running comfyui with the --use-pytorch-cross-attention flag, received the following error:
Prompt executed in 3727.55 seconds
got prompt
0%| | 0/4 [00:00<?, ?it/s]
ERROR:root:!!! Exception during processing !!!
ERROR:root:Traceback (most recent call last):
File "/Users/dmarx/proj/ComfyUI/execution.py", line 153, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
File "/Users/dmarx/proj/ComfyUI/execution.py", line 83, in get_output_data
return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
File "/Users/dmarx/proj/ComfyUI/execution.py", line 76, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
File "/Users/dmarx/proj/ComfyUI/custom_nodes/ComfyUI-LCM/nodes.py", line 63, in sample
result = self.pipe(
File "/Users/dmarx/proj/ComfyUI/_venv/lib/python3.9/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "/Users/dmarx/proj/ComfyUI/custom_nodes/ComfyUI-LCM/lcm/lcm_pipeline.py", line 232, in __call__
model_pred = self.unet(
File "/Users/dmarx/proj/ComfyUI/_venv/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1519, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/Users/dmarx/proj/ComfyUI/_venv/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1528, in _call_impl
return forward_call(*args, **kwargs)
File "/Users/dmarx/proj/ComfyUI/_venv/lib/python3.9/site-packages/diffusers/models/unet_2d_condition.py", line 1031, in forward
sample = upsample_block(
File "/Users/dmarx/proj/ComfyUI/_venv/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1519, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/Users/dmarx/proj/ComfyUI/_venv/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1528, in _call_impl
return forward_call(*args, **kwargs)
File "/Users/dmarx/proj/ComfyUI/_venv/lib/python3.9/site-packages/diffusers/models/unet_2d_blocks.py", line 2320, in forward
hidden_states = upsampler(hidden_states, upsample_size, scale=scale)
File "/Users/dmarx/proj/ComfyUI/_venv/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1519, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/Users/dmarx/proj/ComfyUI/_venv/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1528, in _call_impl
return forward_call(*args, **kwargs)
File "/Users/dmarx/proj/ComfyUI/_venv/lib/python3.9/site-packages/diffusers/models/resnet.py", line 158, in forward
hidden_states = F.interpolate(hidden_states, scale_factor=2.0, mode="nearest")
File "/Users/dmarx/proj/ComfyUI/_venv/lib/python3.9/site-packages/torch/nn/functional.py", line 3990, in interpolate
return torch._C._nn.upsample_nearest2d(input, output_size, scale_factors)
RuntimeError: "upsample_nearest2d_channels_last" not implemented for 'Half'
disabling the "use_fp16" option didn't resolve the issue, but removing the commandline flag did
when running comfyui with the
--use-pytorch-cross-attention
flag, received the following error:disabling the "use_fp16" option didn't resolve the issue, but removing the commandline flag did