Issues with cogvideox custom input

[jmahajan117]

Hello,

I'm trying to generate my own example with 16 frames and I'm getting this issue with CogVideoX. At line 71:

latent_model_input = torch.cat([latent_model_input, latent_image_input], dim=2)
RuntimeError: Sizes of tensors must match except in dimension 2. Expected size 4 but got size 13 for tensor number 1 in the list.

Is there a fix for this as this is a custom pipeline?

[RotsteinNoam]

Thanks for reaching out!

The motion signal and mask videos need to match the number of frames in the video you want to generate. Please make sure the number of frames in motion_signal.mp4 and mask.mp4, and the --num-frames value, are all the same.