This figure summarizes the data distribution of the ywxia/fold_box_dual_arm_60_ana dataset, auto-generated after each conversion via analysis/postprocess_with_overview.py. It shows episode-length distribution, the 3-D EEF workspace, per-dimension state histograms, per-arm action magnitudes, and a sample of frames from each camera.
Task: fold the box on the desk
Episodes: 59 | Frames:… See the full description on the dataset page: https://huggingface.co/datasets/ywxia/fold_box_dual_arm_60_ana.