action, end_effector_pos and image_aux1/encoded explanation

[anestisdotpy]

Hello @febert and thank you for sharing the code. I am using the BAIR robot dataset for my research and I could not understand something.

To begin with, what do the values represent in the action and end_effector_pos features? I assume in the end_effector_pos it is the [x,y,z] coordinates of the end effector but what about the action ? Also, what is the difference between image_aux1/encoded and image_main/encoded?

I have been looking around for hours trying to find the answer to these but I could not find anything. Any help would be much appreciated.

Thank you very much for your time!