ssxxx1a

thanks for your reply! Due to memory issues, I tested model parallelism, which means that the encoder and decoder are placed on different GPUs. Then I found that there have...

confused also....

You should copy the package in this github repository to your own installation directory, it can sovle this bug.

is possible to generate longer videos(>8 frames)? i had tried to train the mode with “n_sample_frames=16” and inference with “video length=16”, but the inference video is full black.