Thanks for making this project. I'm trying to replicate the evaluations. And I saw that you mentioned in a previous issue that you didn't have the bandwidth to release the code you used for the simulation experiments. So I had some questions.
For the LIBERO dataset that contained the RGB-D images did you generate a new dataset? Since the publicly available LIBERO dataset from what I saw only had RGB image.
If you can't release the refactored code can you send me it personally? It would be of great help.