Extending to videos

[rahulseetharaman]

Thank you for the library. Is it possible to extend this implementation for video diffusion models ? Could you share an example of it if possible ? Thanks !