Jiuniu Wang

icon indicating a website link https://wangjiuniu.github.io/ icon indicating an email [email protected]

Results 8 comments of Jiuniu Wang

TypeError: __init__() got an unexpected keyword argument 'pretrained'

Maybe the version of torchvision should be updated.

_linear() got an unexpected keyword argument 'bias_start'

I meet the same problem when I try to use tensorflow==1.3.0. The problem is saved when I change it to the version guided. Good luck!

great product , can i make a longer video ?

We currently only support generating a 16 frames video. Generating longer videos could be another task, and it will be supported soon.

RuntimeError: indices should be either on cpu or on the same device as the indexed tensor (cpu)

This demo requires about 16GB CPU RAM and 16GB GPU RAM. @mic2112

RuntimeError: indices should be either on cpu or on the same device as the indexed tensor (cpu)

See [this page](https://github.com/modelscope/modelscope/blob/master/modelscope/models/multi_modal/video_synthesis/diffusion.py) We have fixed this bug by adding "tensor = tensor.to(x.device)", so the "return tensor[t].view(shape).to(x)" is Line 15 now. @ThatCoffeeGuy

RuntimeError: indices should be either on cpu or on the same device as the indexed tensor (cpu)

You can try to use Colab like [this page](https://colab.research.google.com/drive/1uW1ZqswkQ9Z9bp5Nbo5z59cAn7I0hE6R?usp=sharing). @mic2112

RuntimeError: indices should be either on cpu or on the same device as the indexed tensor (cpu)

The UNet module needs about 16GB GPU memory. So it will be quite complex to run on 12GB GPU. Maybe model parallel would work.

TypeError: unsupported operand type(s) for //: 'int' and 'list'

Hello, I met the same problem because I tried to use stable diffusion v2.1 checkpoint. I solved this problem by using stable diffusion v1.5 as guided in the code. Good...