Error occurred when executing StyleModelApply: Sizes of tensors must match except in dimension 1
MMoneer opened this issue · 5 comments
Hello,
I got this error when trying to use the style model workflow.
Error occurred when executing StyleModelApply:
Sizes of tensors must match except in dimension 1. Expected size 1280 but got size 1024 for tensor number 1 in the list.
File "D:\ComfyUI_windows_portable\ComfyUI\execution.py", line 154, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
File "D:\ComfyUI_windows_portable\ComfyUI\execution.py", line 84, 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 77, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
File "D:\ComfyUI_windows_portable\ComfyUI\nodes.py", line 923, in apply_stylemodel
cond = style_model.get_cond(clip_vision_output).flatten(start_dim=0, end_dim=1).unsqueeze(dim=0)
File "D:\ComfyUI_windows_portable\ComfyUI\comfy\sd.py", line 290, in get_cond
return self.model(input.last_hidden_state)
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\comfy\t2i_adapter\adapter.py", line 214, in forward
x = torch.cat([x, style_embedding], dim=1)
I have the same error :(
I have the same error :(
Add --force-fp16 to run_nvidia_gpu.bat
.\python_embeded\python.exe -s ComfyUI\main.py --windows-standalone-build --force-fp16
pause
Still same problem unfortunately. Would you mind sharing a working workflow so I can test and see if I have missed something?
Never mind the images. Those are just a trial and error. Have tried changing resolution of images, scalers and everything.
I'm so sorry, I mixed up the topics. This is not a solution to this problem, This didn't fix it for me too. The --force-fp16 is for Dtype mismatch issue, not the tensor mismatch.
To Fix it you can download this clip vision
https://huggingface.co/openai/clip-vit-large-patch14/resolve/main/pytorch_model.bin?download=true
It's work for me
You can Save it as clip-vit-large-patch14.bin instead of pytorch_model.bin
Also, switch between t2iadapter_style_sd14v1.pth and coadapter-style-sd15v1.pth to figure out which one you like