LeWAM real-robot evaluations
Collection
Human-scored 30-trial policy rollouts on a Rebot B601 arm: Pick and Place T, Push-T, Tape Stack. • 12 items • Updated
Datasets:
30 episodes · 30 fps · 3 cameras · 672×376 h264
This dataset was created using LeRobot.
Dataset Description
Real-robot evaluation rollouts recorded on a Rebot B601 7-DoF arm, from the LeWAM project. Every episode here is a policy rollout on the physical robot — not a teleoperated demonstration — scored by a human operator immediately after it ran.
- Task: tape stack (
b601_tape_stack): the arm stacks the green roll of tape on the white roll of tape. Trained onLeWAM/b601_tape_stack. - Policy: LeWAM self-supervised WAM (
lewam_frozen), the same joint action + future-latent objective, but with the vision encoder frozen - Success rate: 5/30 = 16.7%
- Episodes: 30 · frames: 29335 · fps: 30 · robot time: 16.3 min
- Wall-clock: 23.9 min (median 40s, max 147s per episode) — longer than robot time because inference on this hardware runs slower than real time
- wandb run:
8hthdbbj(eval-lewam_frozen_b601_tape_stack) - Episode cap: 5400 steps — an episode that hits it is scored a failure.
Observations
Three synchronised camera streams (observation.images.zed_left,
observation.images.zed_right from a ZED Mini, observation.images.wrist_cam from a
wrist-mounted USB camera) plus observation.state. action is the 7-DoF joint command
the policy emitted.
Note the policies were served torque-only state (observation.state sliced to
*.torq); the observation.state recorded here is the robot's full state vector, so
reproducing what the policy actually saw means applying the same slice.
Per-episode scoring
eval_session.json holds the operator's label for each episode:
| episode | success | steps | seconds |
|---|---|---|---|
| 0 | ❌ | 1532 | 73.3 |
| 1 | ❌ | 1777 | 87.8 |
| 2 | ✅ | 498 | 24.5 |
| 3 | ✅ | 158 | 8.4 |
| 4 | ❌ | 3124 | 146.6 |
| 5 | ❌ | 422 | 20.9 |
| 6 | ❌ | 663 | 32.7 |
| 7 | ❌ | 956 | 46.6 |
| 8 | ❌ | 938 | 45.2 |
| 9 | ❌ | 970 | 46.4 |
| 10 | ❌ | 534 | 26.5 |
| 11 | ❌ | 764 | 37.7 |
| 12 | ❌ | 668 | 32.4 |
| 13 | ❌ | 1568 | 75.2 |
| 14 | ❌ | 717 | 38.1 |
| 15 | ❌ | 1154 | 57.1 |
| 16 | ❌ | 857 | 42.1 |
| 17 | ✅ | 474 | 24.7 |
| 18 | ❌ | 2154 | 105.5 |
| 19 | ✅ | 1127 | 54.4 |
| 20 | ❌ | 2036 | 97.2 |
| 21 | ❌ | 420 | 21.4 |
| 22 | ❌ | 850 | 41.6 |
| 23 | ❌ | 882 | 43.7 |
| 24 | ❌ | 510 | 25.9 |
| 25 | ❌ | 1226 | 59.6 |
| 26 | ❌ | 740 | 36.6 |
| 27 | ❌ | 737 | 36.5 |
| 28 | ❌ | 632 | 31.0 |
| 29 | ✅ | 277 | 17.4 |
- Homepage: https://github.com/erykhalicki/LeWAM
- Paper: [More Information Needed]
- License: apache-2.0
Dataset Structure
{
"codebase_version": "v3.0",
"fps": 30,
"features": {
"observation.state": {
"dtype": "float32",
"shape": [
21
],
"names": [
"shoulder_pan.pos",
"shoulder_lift.pos",
"elbow_flex.pos",
"wrist_flex.pos",
"wrist_yaw.pos",
"wrist_roll.pos",
"gripper.pos",
"shoulder_pan.torq",
"shoulder_lift.torq",
"elbow_flex.torq",
"wrist_flex.torq",
"wrist_yaw.torq",
"wrist_roll.torq",
"gripper.torq",
"shoulder_pan.vel",
"shoulder_lift.vel",
"elbow_flex.vel",
"wrist_flex.vel",
"wrist_yaw.vel",
"wrist_roll.vel",
"gripper.vel"
]
},
"observation.images.zed_left": {
"dtype": "video",
"shape": [
376,
672,
3
],
"names": [
"height",
"width",
"channels"
],
"info": {
"is_depth_map": false,
"video.height": 376,
"video.width": 672,
"video.codec": "h264",
"video.pix_fmt": "yuv420p",
"video.fps": 30,
"video.channels": 3,
"has_audio": false,
"video.g": 2,
"video.crf": 30,
"video.preset": null,
"video.fast_decode": 0,
"video.video_backend": "pyav",
"video.extra_options": {}
}
},
"observation.images.zed_right": {
"dtype": "video",
"shape": [
376,
672,
3
],
"names": [
"height",
"width",
"channels"
],
"info": {
"is_depth_map": false,
"video.height": 376,
"video.width": 672,
"video.codec": "h264",
"video.pix_fmt": "yuv420p",
"video.fps": 30,
"video.channels": 3,
"has_audio": false,
"video.g": 2,
"video.crf": 30,
"video.preset": null,
"video.fast_decode": 0,
"video.video_backend": "pyav",
"video.extra_options": {}
}
},
"observation.images.wrist_cam": {
"dtype": "video",
"shape": [
360,
640,
3
],
"names": [
"height",
"width",
"channels"
],
"info": {
"is_depth_map": false,
"video.height": 360,
"video.width": 640,
"video.codec": "h264",
"video.pix_fmt": "yuv420p",
"video.fps": 30,
"video.channels": 3,
"has_audio": false,
"video.g": 2,
"video.crf": 30,
"video.preset": null,
"video.fast_decode": 0,
"video.video_backend": "pyav",
"video.extra_options": {}
}
},
"action": {
"dtype": "float32",
"shape": [
7
],
"names": [
"shoulder_pan.pos",
"shoulder_lift.pos",
"elbow_flex.pos",
"wrist_flex.pos",
"wrist_yaw.pos",
"wrist_roll.pos",
"gripper.pos"
]
},
"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": 30,
"total_frames": 29335,
"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": null,
"splits": {
"train": "0:30"
}
}
Citation
BibTeX:
[More Information Needed]
- Downloads last month
- 32