Hugging Face's logo

Datasets:
leapshared
/
sciedu_bench_v2_sim_props8_all

Tasks:
Robotics
Modalities:
Tabular
Time-series
Video
Formats:
parquet
Size:
1M - 10M
Tags:
LeRobot
sciedu-bench
simulation
mujoco
openarm
Libraries:
Datasets
Dask
Polars
License:
Dataset card Data Studio Files
xet
 Community
Episodes Preview openarm_sciedu_sim Visualizer
2.37k episodes · 30 fps · 3 cameras · 640×480 av1

This dataset was created using LeRobot.

Dataset Description

leapshared/sciedu_bench_v2_sim_props8_all

SciEdu-Bench 학습 시연 모두를 한 데이터셋으로 합친 것이다(LeRobot 0.6.1 lerobot-train은 데이터셋 하나만 받는다). 장면 판 props-8 + mineral-3(광물 등 따로 규격인 묶음은 제 판, 아래 표), 규격 판 2, 꾸러미 판 0.7.0-draft. 실험마다 출처·생성기·평가 시작과의 분리·한계는 아래 각 데이터셋의 카드에 있다. sim_episodes.jsonl의 dataset이 그 이름이다. 녹화 재생 편은 없다.

실험 데이터셋 장면 판 편 프레임
electroscope_rod leapshared/sciedu_bench_v2_sim_props8_electroscope_rod props-8 300 199137
leaf leapshared/sciedu_bench_v2_sim_props8_leaf props-8 150 114251
redox_steps leapshared/sciedu_bench_v2_sim_props8_redox_steps props-8 450 356423
redox leapshared/sciedu_bench_v2_sim_props8_redox props-8 515 146482
mineral leapshared/sciedu_bench_v2_sim_mineral3_mineral mineral-3 950 621758
과제 묶음 문장 편/목표
electroscope_rod_pick spatial pick up the black charging rod 50/50
electroscope_rod_pick_look object pick up the black charging rod 50/50
electroscope_rod_near goal move the rod toward and away from the electroscope 50/50
electroscope_rod_rub goal rub the rod on the fur 50/50
electroscope_rod_put goal put the rod down on the white holder 50/50
electroscope_rod_script long pick up the black charging rod 50/50
leaf_left goal put the leaf on the left side 50/50
leaf_right goal put the leaf on the right side 50/50
leaf_script long put the leaf on the left side 50/50
redox_steps_lift spatial lift the metal plate out of the left beaker 50/50
redox_steps_move_blue spatial move the plate over the blue solution 50/50
redox_steps_put_left spatial put the plate into the left beaker 50/50
redox_steps_move_blue_look object move the plate over the blue solution 50/50
redox_steps_dip goal lower the plate into the blue solution 50/50
redox_steps_move_left goal move the plate over the left beaker 50/50
redox_steps_script long lift the metal plate out of the left beaker 50/50
redox_steps_raise rest raise the plate out of the solution 50/50
redox_steps_home rest return to the home posture 50/50
redox_lift spatial lift the metal plate out of the left beaker 47/50
redox_move_blue spatial move the plate over the blue solution 50/50
redox_put_left spatial put the plate into the left beaker 50/50
redox_move_blue_look object move the plate over the blue solution 50/50
redox_point goal point at the plate with the left hand 50/50
redox_dip goal lower the plate into the blue solution 50/50
redox_move_left goal move the plate over the left beaker 50/50
redox_script long lift the metal plate out of the left beaker 18/50
redox_raise rest raise the plate out of the solution 50/50
redox_point_again rest point at the dipped plate with the left hand 50/50
redox_home rest return to the home posture 50/50
mineral_pick_right spatial pick up the right mineral with the left hand 50/50
mineral_pick_middle spatial pick up the middle mineral with the left hand 50/50
mineral_pick_left spatial pick up the left mineral with the left hand 50/50
mineral_pick_dropper spatial pick up the dropper by the top with the right hand 50/50
mineral_pick_middle_look object pick up the middle mineral with the left hand 50/50
mineral_rub_look object rub the mineral on the streak plate 50/50
mineral_rub goal rub the mineral on the streak plate 50/50
mineral_filings goal lower the mineral onto the iron filings and hold still 50/50
mineral_put_middle goal put the middle mineral back 50/50
mineral_draw goal draw acid from the petri dish 50/50
mineral_squeeze goal squeeze the dropper onto the mineral 50/50
mineral_dropper_in_beaker goal put the dropper in the beaker 50/50
mineral_round long pick up the right mineral with the left hand 50/50
mineral_round_swap long pick up the left mineral with the left hand 50/50
mineral_rub_below rest rub the mineral on the streak plate below the first streak 50/50
mineral_put_right rest put the right mineral back 50/50
mineral_put_left rest put the mineral in the empty petri dish 50/50
mineral_home rest return to the home posture 50/50
mineral_rub_free rest rub the mineral on the streak plate 50/50

합 2365편, 1438051프레임. 만든 법: 장면에서 짠 IK 코드 1400, 녹화 손 길을 흔든 물체에 실은 IK 코드 965.

라이선스: apache-2.0(허브 메타데이터 apache-2.0, 선생님 2026-11-11, 실물 녹화 leapshared/SciEdu_mineral_steps_20261109와 같게). 커밋을 고정해 부른다(--dataset.revision=<커밋>).

All SciEdu-Bench training demos in one dataset (scenes props-8, mineral-3; each part's scene is in the table), for trainers that take one dataset. Per-experiment sources, generators and the split from the evaluation starts are on each part's card. No recording replay. License: apache-2.0 (metadata apache-2.0). Pin a commit.

  • Homepage: [More Information Needed]
  • Paper: [More Information Needed]
  • License: apache-2.0

Dataset Structure

meta/info.json:

{
    "codebase_version": "v3.0",
    "fps": 30,
    "features": {
        "action": {
            "dtype": "float32",
            "shape": [
                16
            ],
            "names": [
                "left_joint_1.pos",
                "left_joint_2.pos",
                "left_joint_3.pos",
                "left_joint_4.pos",
                "left_joint_5.pos",
                "left_joint_6.pos",
                "left_joint_7.pos",
                "left_gripper.pos",
                "right_joint_1.pos",
                "right_joint_2.pos",
                "right_joint_3.pos",
                "right_joint_4.pos",
                "right_joint_5.pos",
                "right_joint_6.pos",
                "right_joint_7.pos",
                "right_gripper.pos"
            ]
        },
        "observation.state": {
            "dtype": "float32",
            "shape": [
                16
            ],
            "names": [
                "left_joint_1.pos",
                "left_joint_2.pos",
                "left_joint_3.pos",
                "left_joint_4.pos",
                "left_joint_5.pos",
                "left_joint_6.pos",
                "left_joint_7.pos",
                "left_gripper.pos",
                "right_joint_1.pos",
                "right_joint_2.pos",
                "right_joint_3.pos",
                "right_joint_4.pos",
                "right_joint_5.pos",
                "right_joint_6.pos",
                "right_joint_7.pos",
                "right_gripper.pos"
            ]
        },
        "observation.images.follower_d455f": {
            "dtype": "video",
            "shape": [
                480,
                640,
                3
            ],
            "names": [
                "height",
                "width",
                "channels"
            ],
            "info": {
                "video.height": 480,
                "video.width": 640,
                "video.codec": "av1",
                "video.pix_fmt": "yuv420p",
                "video.fps": 30,
                "video.channels": 3,
                "has_audio": false,
                "video.g": 2,
                "video.crf": 30,
                "video.preset": 12,
                "video.fast_decode": 0,
                "video.video_backend": "pyav",
                "video.extra_options": {},
                "is_depth_map": false
            }
        },
        "observation.images.left_wrist": {
            "dtype": "video",
            "shape": [
                480,
                640,
                3
            ],
            "names": [
                "height",
                "width",
                "channels"
            ],
            "info": {
                "video.height": 480,
                "video.width": 640,
                "video.codec": "av1",
                "video.pix_fmt": "yuv420p",
                "video.fps": 30,
                "video.channels": 3,
                "has_audio": false,
                "video.g": 2,
                "video.crf": 30,
                "video.preset": 12,
                "video.fast_decode": 0,
                "video.video_backend": "pyav",
                "video.extra_options": {},
                "is_depth_map": false
            }
        },
        "observation.images.right_wrist": {
            "dtype": "video",
            "shape": [
                480,
                640,
                3
            ],
            "names": [
                "height",
                "width",
                "channels"
            ],
            "info": {
                "video.height": 480,
                "video.width": 640,
                "video.codec": "av1",
                "video.pix_fmt": "yuv420p",
                "video.fps": 30,
                "video.channels": 3,
                "has_audio": false,
                "video.g": 2,
                "video.crf": 30,
                "video.preset": 12,
                "video.fast_decode": 0,
                "video.video_backend": "pyav",
                "video.extra_options": {},
                "is_depth_map": false
            }
        },
        "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
        }
    },
    "total_episodes": 2365,
    "total_frames": 1438051,
    "total_tasks": 30,
    "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": "openarm_sciedu_sim",
    "splits": {
        "train": "0:2365"
    }
}

Citation

BibTeX:

[More Information Needed]
Downloads last month
42
Number of rows:
1,438,051
Total file size:
66.2 GB

Models trained or fine-tuned on leapshared/sciedu_bench_v2_sim_props8_all