Closed evelinehong closed 1 week ago
Hi, thanks for your interest in our work!
We are currently organizing our code and will make it publicly available soon (before September).
Hi, we have released our training codes for action-conditioned video prediction and visual MBRL. Note that we only conduct action-free video pre-training. Thus, we currently do not have a checkpoint for action-conditioned prediction on OXE, but we plan to train such an iVideoGPT in the future.
Please let me know if you have any remaining questions.
Close the issue for now, but please reopen it if you have further questions. Thank you!
Hi,
When will you release your codes for action-conditioned training / inference codes and the checkpoint?