SteveNguyen/sugar_cup_graspproj_chunkrel

This dataset was created using LeRobot.

Dataset Structure

meta/info.json:

{
    "codebase_version": "v3.0",
    "fps": 50,
    "features": {
        "observation.images.cam0": {
            "dtype": "video",
            "shape": [
                3,
                360,
                480
            ],
            "names": [
                "channels",
                "height",
                "width"
            ],
            "info": {
                "video.height": 360,
                "video.width": 480,
                "video.codec": "h264",
                "video.pix_fmt": "yuv420p",
                "video.is_depth_map": false,
                "video.fps": 50,
                "video.channels": 3,
                "has_audio": false
            }
        },
        "action": {
            "dtype": "float32",
            "shape": [
                8
            ],
            "names": [
                "x",
                "y",
                "z",
                "ax",
                "ay",
                "az",
                "strategy",
                "closure"
            ]
        },
        "is_lost": {
            "dtype": "float32",
            "shape": [
                1
            ],
            "names": [
                "right_is_lost"
            ]
        },
        "timestamp": {
            "dtype": "float32",
            "shape": [
                1
            ],
            "names": null
        },
        "frame_index": {
            "dtype": "int64",
            "shape": [
                1
            ],
            "names": null
        },
        "episode_index": {
            "dtype": "int64",
            "shape": [
                1
            ],
            "names": null
        },
        "index": {
            "dtype": "int64",
            "shape": [
                1
            ],
            "names": null
        },
        "task_index": {
            "dtype": "int64",
            "shape": [
                1
            ],
            "names": null
        },
        "observation.state": {
            "dtype": "float32",
            "shape": [
                2
            ],
            "names": [
                "strategy",
                "closure"
            ]
        }
    },
    "total_episodes": 150,
    "total_frames": 23046,
    "total_tasks": 1,
    "chunks_size": 1000,
    "data_files_size_in_mb": 100,
    "video_files_size_in_mb": 200,
    "data_path": "data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet",
    "video_path": "videos/{video_key}/chunk-{chunk_index:03d}/file-{file_index:03d}.mp4",
    "robot_type": "grabette",
    "splits": {
        "train": "0:150"
    }
}

Citation

BibTeX:

[More Information Needed]

Grabette notes

Gripper channels are (strategy, closure), not joint angles: the last two channels of action and observation.state are normalised to the gripper's reachable travel. In action, closure = 1.0 means "drive fully closed" and the object stops the fingers; in observation.state closure stays continuous, so "am I holding something" remains observable. The exact calibration is in meta/grasp_projection.json.

action is 8D absolute pose [x, y, z, ax, ay, az, strategy, closure] — deliberately not per-step deltas. meta/stats.json carries action_relative_meta, so the action stats describe the chunk-relative distribution (chunk_size 50) rather than the absolute one. Training must set GRABETTE_CHUNK_RELATIVE=1; integrations/Pi05/train.py refuses otherwise rather than normalise with the wrong scale.

Prepared from chouziel/grabette-sugar-cup-2008 to be channel-for-channel comparable with chouziel/grabette-sugar-cup-2008_graspproj (the per-step-delta baseline): the projected gripper channels are bit-identical across all 23046 frames, so an A/B between them isolates the action representation.