Dataset Viewer
Auto-converted to Parquet Duplicate
mission_id
stringclasses
10 values
series
int64
1
1
pref_type
stringclasses
6 values
pref_canonical
stringclasses
8 values
task
stringclasses
9 values
hdf5_file
stringclasses
10 values
num_episodes
int64
100
100
size_gb
float64
3.43
13.6
1-1_004
1
speed_range
Speed up
put alphabet soup and cream cheese in basket
demos/Mission_1-1_004.hdf5
100
7.72
1-1_006
1
speed_range
Speed up
put both the cream cheese box and the butter in the basket
demos/Mission_1-1_006.hdf5
100
9.31
1-2_113
1
speed_range
Slow down
pick up the butter and place it in the basket
demos/Mission_1-2_113.hdf5
100
13.6
1-3_107
1
max_acc
Avoid sudden acceleration
put the cream cheese in the bowl
demos/Mission_1-3_107.hdf5
100
3.43
1-3_109
1
max_acc
Avoid sudden acceleration
put the wine bottle on top of the cabinet
demos/Mission_1-3_109.hdf5
100
3.75
1-4_034
1
avoid_object
Keep away from wine bottle
put the black bowl in the bottom drawer of the cabinet
demos/Mission_1-4_034.hdf5
100
5.73
1-5_114
1
height_offset
Close to ground when moving
pick up the chocolate pudding and place it in the basket
demos/Mission_1-5_114.hdf5
100
7.44
1-6_105
1
straight_path
Move straight
put the bowl on the stove
demos/Mission_1-6_105.hdf5
100
3.52
1-7_104
1
height_offset
Move higher
put the bowl on the plate
demos/Mission_1-7_104.hdf5
100
4.63
1-8_114
1
rotation
Rotate when moving
pick up the chocolate pudding and place it in the basket
demos/Mission_1-8_114.hdf5
100
6.85

Anonymous Submission — Series 1 (Motion Preference) demos

10 LIBERO missions × 100 successful scripted-policy rollouts each, stored as HDF5 under demos/Mission_<id>.hdf5. The viewer below lists per-mission metadata (preference axis, task instruction, file size, episode count); the actual trajectory data is in the HDF5 files.

Quick-inspect samples (< 4 GB each)

For reviewers who want a fast look without downloading the full 62 GB:

Mission Preference Size
Mission_1-3_107.hdf5 Avoid sudden acceleration 3.43 GB
Mission_1-6_105.hdf5 Move straight 3.52 GB
Mission_1-3_109.hdf5 Avoid sudden acceleration 3.75 GB

Pull just one with:

hf download AnonymousSubmissionAccount/AnonymousSubmission \
    --repo-type dataset --include "demos/Mission_1-3_107.hdf5" --local-dir ./sample

Loading

import h5py
with h5py.File("demos/Mission_1-1_004.hdf5", "r") as f:
    print(list(f["data"].keys()))               # 100 episode keys
    ep0 = f["data/demo_0"]
    print(ep0["actions"].shape, ep0["obs/agentview_rgb"].shape)

Code & full benchmark

https://github.com/AnonymousSubmissionAccount328/PlacehHolder328

Downloads last month
94