THULab/ruanwz_so101_test

SO101 Test TsFile

Apache TsFile edition of ruanwz/so101_test, a LeRobot v2.1 SO101 dataset for placing a coffee capsule into a cup. The numeric trajectories are stored in one table-model TsFile.

Source and attribution

• Original author and repository owner: David Ruan (ruanwz). All source repository commits are attributed to ruanwz.
• License: Apache-2.0.
• The source card does not provide a homepage, paper, or completed BibTeX citation.
• Split: train; 50 episodes; 20,203 frame rows; one task; 30 FPS; 50 source Parquet shards.
• Task 0: Place the coffee capsule into the cup.

Data layout

The table is ruanwz_so101_test and contains 20,203 rows across 50 TAG devices. The source Parquet shards total 1,005,865 bytes; the TsFile is 307,516 bytes (30.6% of the source Parquet size).

Column	TsFile role	Type	Meaning
Time	TIME	INT64 milliseconds	round(timestamp * 1000), restarting at zero per episode
episode_index	TAG	STRING from source INT64	Source episode identity
task_index	TAG	STRING from source INT64	Source task identity
frame_index	FIELD	INT64	Frame position within the episode
sample_index	FIELD	INT64	Source index, renamed for clarity
action_0 ... action_5	FIELD	FLOAT	Flattened action[6]
observation_state_0 ... observation_state_5	FIELD	FLOAT	Flattened observation.state[6]

timestamp is not retained as a separate FIELD because it is represented by Time / 1000 seconds. The vector column names preserve their source prefixes, with dots changed to underscores. No trajectory row, episode, task, action dimension, or state dimension is removed.

Videos and alignment

The 100 source AV1 videos are not included here. They remain under videos/chunk-000 in the original repository:

• observation.images.laptop: 50 files, 237,672,232 bytes
• observation.images.phone: 50 files, 73,213,148 bytes

The source template is videos/chunk-{episode_chunk:03d}/{video_key}/episode_{episode_index:06d}.mp4. Use episode_index and frame_index to align each numeric row with both 30 FPS video streams.

Read example

from tsfile import TsFileReader

reader = TsFileReader("data/ruanwz_so101_test.tsfile")
with reader.query_table(
    "ruanwz_so101_test",
    ["episode_index", "task_index", "frame_index", "sample_index", "action_0", "observation_state_0"],
    batch_size=1024,
) as result:
    batch = result.read_arrow_batch()
    print(batch.to_pandas().head())
reader.close()