What Matters in Learning from Offline Human Demonstrations for Robot Manipulation / Robomimic

Based off mujoco