Open ChengYaofeng opened 1 year ago
Hi ChengYaofeng, we actually have another repo for these SPA baselines, see https://github.com/haosulab/ManiSkill2-SPA (will be available later today or tomorrow)
Sorry for not making this repo public earlier.
Hi ChengYaofeng, we actually have another repo for these SPA baselines, see https://github.com/haosulab/ManiSkill2-SPA (will be available later today or tomorrow)
Sorry for not making this repo public earlier.
Open source takes time. I'd appreciate your input on this. And I'm proud that an RL env is created by the Chinese :)
Your work is really impressive. I'm curious about your paper's additional experiment F.1. How did you execute the Contact-GraspNet in your environment? Did you try to use Contact-GraspNet as the RL pretrained model to guide the pick and place task as your paper said that "We directly use the pretrained model provided by the original authors of Contact-GraspNet"?