dexsuite/stack_franka_allegro_mg
This dataset was created using LeRobot. Dataset Description Machine-generated DexSuite data for stack with the Franka arm and the allegro end-effector: 200 successful rollouts of the paper's low-dimensional (state) diffusion policy checkpoint trained on the 50 human demonstrations (checkpoint model_epoch_250_stack_franka_allegro_success_0.06.pth; 202 successes in 5998 attempts, success rate 0.03; env seed = base seed + attempt index). All episodes are successful, recorded… See the full description on the dataset page: https://huggingface.co/datasets/dexsuite/stack_franka_allegro_mg.
This dataset was created using LeRobot.
<a class="flex" href="https://huggingface.co/spaces/lerobot/visualizedataset?path=dexsuite/stackfrankaallegromg"> <img class="block dark:hidden" src="https://huggingface.co/datasets/huggingface/badges/resolve/main/visualize-this-dataset-xl.svg"/> <img class="hidden dark:block" src="https://huggingface.co/datasets/huggingface/badges/resolve/main/visualize-this-dataset-xl-dark.svg"/> </a>
Dataset Description
Machine-generated DexSuite data for stack with the Franka arm and the allegro end-effector: 200 successful rollouts of the paper's low-dimensional (state) diffusion policy checkpoint trained on the 50 human demonstrations (checkpoint model_epoch_250_stack_franka_allegro_success_0.06.pth; 202 successes in 5998 attempts, success rate 0.03; env seed = base seed + attempt index). All episodes are successful, recorded at 20 Hz with a static front camera and a wrist camera (224x224 RGB video), proprioception and ground-truth object state, under DexSuite's shared observation/action interface (actions are OSC pose offsets from the episode-start end-effector pose plus gripper/hand joint targets). As in every DexSuite set, frame t stores the observation before action t; the success flag on the last frame refers to the transition it triggers, so the terminal state itself is not stored. Episodes were collected by several workers with disjoint seed ranges and merged; see meta/mergemanifest.json and meta/mgsource.json.
- Homepage: https://huggingface.co/dexsuite
- Paper: DexSuite (RA-L 2026 submission)
- License: apache-2.0
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": null,
"total_episodes": 200,
"total_frames": 62624,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 20,
"splits": {
"train": "0:200"
},
"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",
"features": {
"obs.state.gripper.qpos": {
"dtype": "float32",
"shape": [
16
]
},
"obs.state.gripper.qtorque": {
"dtype": "float32",
"shape": [
16
]
},
"obs.state.gripper.qvel": {
"dtype": "float32",
"shape": [
16
]
},
"obs.state.gripper.tcp_pos": {
"dtype": "float32",
"shape": [
3
]
},
"obs.state.gripper.tcp_quat": {
"dtype": "float32",
"shape": [
4
]
},
"reward": {
"dtype": "float32",
"shape": [
1
]
},
"truncated": {
"dtype": "bool",
"shape": [
1
]
},
"success": {
"dtype": "bool",
"shape": [
1
]
},
"failure": {
"dtype": "bool",
"shape": [
1
]
},
"action": {
"dtype": "float32",
"shape": [
23
]
},
"obs.state.manipulator.eef_pos": {
"dtype": "float32",
"shape": [
1,
3
]
},
"obs.state.manipulator.eef_quat": {
"dtype": "float32",
"shape": [
1,
4
]
},
"obs.state.manipulator.qpos": {
"dtype": "float32",
"shape": [
7
]
},
"obs.state.manipulator.qtorque": {
"dtype": "float32",
"shape": [
7
]
},
"obs.state.manipulator.qvel": {
"dtype": "float32",
"shape": [
7
]
},
"obs.state.other.action": {
"dtype": "float32",
"shape": [
23
]
},
"obs.state.other.cube_a_pos": {
"dtype": "float32",
"shape": [
3
]
},
"obs.state.other.cube_a_quat": {
"dtype": "float32",
"shape": [
4
]
},
"obs.state.other.cube_b_pos": {
"dtype": "float32",
"shape": [
3
]
},
"obs.state.other.cube_b_quat": {
"dtype": "float32",
"shape": [
4
]
},
"obs.state.other.cube_c_pos": {
"dtype": "float32",
"shape": [
3
]
},
"obs.state.other.cube_c_quat": {
"dtype": "float32",
"shape": [
4
]
},
"obs.state.other.last_action": {
"dtype": "float32",
"shape": [
23
]
},
"obs.cameras.front.depth": {
"dtype": "float32",
"shape": [
224,
224
]
},
"obs.cameras.front.rgb": {
"dtype": "video",
"shape": [
224,
224,
3
],
"info": {
"video.height": 224,
"video.width": 224,
"video.codec": "av1",
"video.pix_fmt": "yuv420p",
"video.is_depth_map": false,
"video.fps": 20,
"video.channels": 3,
"has_audio": false
}
},
"obs.cameras.wrist.depth": {
"dtype": "float32",
"shape": [
224,
224
]
},
"obs.cameras.wrist.rgb": {
"dtype": "video",
"shape": [
224,
224,
3
],
"info": {
"video.height": 224,
"video.width": 224,
"video.codec": "av1",
"video.pix_fmt": "yuv420p",
"video.is_depth_map": false,
"video.fps": 20,
"video.channels": 3,
"has_audio": 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
}
},
"task": "stack",
"options": null
}Citation
BibTeX:
[More Information Needed]