Episodes Preview ALOHA Visualizer
103 episodes · 50 fps

Two Pipes TsFile

This is an Apache TsFile conversion of pyrrosk/two_pipes, a LeRobot v3.0 ALOHA dataset containing bilateral robot state/action time series.

Source and attribution

  • Original creator/uploader: Pyrros Koussios (pyrrosk)
  • Pinned revision: a7554820e1bba23648f2e067d812fbbece26612e
  • License: Apache-2.0
  • Train split: 103 episodes, 248,065 rows, 2 tasks, 50 fps, 1 Parquet shard
  • Task 0: insert the red pipe into the fitting (85,561 rows)
  • Task 1: insert the blue pipe into the fitting (162,504 rows)
  • The source card provides no paper or completed BibTeX citation.

Source videos remain at pyrrosk/two_pipes/videos: observation.images.top (53 MP4s), observation.images.left_wrist (45), and observation.images.right_wrist (39), for 137 files / 23.89 GB total. Videos are not included here. Align them with episode_index, frame_index, and the per-stream from_timestamp/to_timestamp offsets in meta/episodes.

Converted artifact

  • Files: data/pyrrosk_two_pipes_part_00.tsfile through data/pyrrosk_two_pipes_part_15.tsfile
  • Sharding: 16 episode-boundary TsFiles; no episode crosses a file boundary
  • Combined TsFile size: 14.40 MiB
  • Table: pyrrosk_two_pipes
  • Rows: 248,065; episode IDs: 103; TAG devices: 206
  • Time: round(timestamp * 1000) milliseconds, restarting per episode
  • Size ratio versus source Parquet: 0.502

Schema

Role Columns
TIME Time
TAG/device episode_index, task_index
FIELD frame_index, sample_index (renamed from index)
FIELD observation_state_0 through observation_state_13
FIELD action_0 through action_13

The source timestamp column is dropped after Time synthesis because it equals Time / 1000 seconds. The 14-element FLOAT32 vectors are flattened row-major, preserving their full prefixes with dots changed to underscores. No row or numeric vector element is dropped.

Physical encoding

  • FLOAT/DOUBLE: GORILLA + ZSTD
  • INT32/INT64: TS_2DIFF + ZSTD
  • Time: TS_2DIFF + LZ4
  • BOOLEAN, if present in a later revision: RLE + LZ4
  • TAGs: TsFile table device/TAG mechanism

meta/ is preserved and meta/info.json documents the conversion. Video features and openpi_assets remain at the source. Conversion scripts and validation reports are local only and are not part of this upload-ready directory.

Read example

from tsfile import TsFileReader

reader = TsFileReader("data/pyrrosk_two_pipes_part_00.tsfile")
schema = reader.get_all_table_schemas()["pyrrosk_two_pipes"]
columns = [c.get_column_name() for c in schema.get_columns()
           if c.get_column_name() != "Time"]
with reader.query_table("pyrrosk_two_pipes", columns, batch_size=65536) as result:
    batch = result.read_arrow_batch()

Citation

Cite the original dataset and Pyrros Koussios (pyrrosk). The source card does not include a paper or BibTeX entry.

Downloads last month
158