when i execute "python vid2vid/main.py ", something wrong happend

[wangbochao1]

Traceback (most recent call last):
File "/root/autodl-tmp/project/StreamDiffusion/examples/vid2vid/main.py", line 106, in
fire.Fire(main)
File "/root/miniconda3/envs/streamdiffusion/lib/python3.10/site-packages/fire/core.py", line 141, in Fire
component_trace = _Fire(component, args, parsed_flag_args, context, name)
File "/root/miniconda3/envs/streamdiffusion/lib/python3.10/site-packages/fire/core.py", line 475, in _Fire
component, remaining_args = _CallAndUpdateTrace(
File "/root/miniconda3/envs/streamdiffusion/lib/python3.10/site-packages/fire/core.py", line 691, in _CallAndUpdateTrace
component = fn(*varargs, **kwargs)
File "/root/autodl-tmp/project/StreamDiffusion/examples/vid2vid/main.py", line 99, in main
video_result[i] = output_image.permute(1, 2, 0)
RuntimeError: The expanded size of the tensor (364) must match the existing size (600) at non-singleton dimension 1. Target sizes: [604, 364, 3]. Tensor sizes: [360, 600, 3]

[earthwormjeff]

replace at the end of main.py :
for _ in range(stream.batch_size):
stream(image=torch.permute(video[0], (2, 1, 0)))

for i in tqdm(range(video.shape[0])):
    output_image = stream(torch.permute(video[i],(2, 1, 0)))
    video_result[i] = torch.permute(output_image, (2, 1, 0))

[wangbochao1]

thanks