sock_orange (TsFile)

Apache TsFile version of makermods/sock_orange.

Overview

This dataset is a LeRobot v3.0 demonstration of a sock-manipulation task recorded with an SO-100 follower arm.

• Robot: SO-100 follower (so_follower)
• Scale: 49 episodes, 27,397 frames, 30 fps
• Split: train
• Cameras (not uploaded): front and wrist

Schema (TsFile structure)

The TsFile table is named sock_orange.

Role	Columns
Time	Time, INT64 milliseconds
TAG	episode_index, task_index
FIELD	frame_index, sample_index
FIELD	action_0 … action_5
FIELD	observation_state_0 … observation_state_5

Vector columns are flattened into scalar FLOAT measurements (single precision). The source index column is retained as sample_index.

Conversion notes

• Time = round(timestamp * 1000) with millisecond precision; time restarts inside each episode, while episode_index and task_index identify the TsFile device.
• The original timestamp field is omitted because it is exactly represented by Time / 1000.
• Camera video streams are NOT included in this repository; they remain in the source videos tree.
• The source meta/info.json declares 53 episodes / 29,466 frames, but the uploaded data/ contains 49 episodes / 27,397 frames (episodes 29-32 are absent from the source data files). The TsFile contains the 49 episodes actually present in data/.
• meta/ is mirrored from the source. Aside from the redundant timestamp column and the excluded videos, no source rows or numeric fields are dropped.

Read example

from tsfile import TsFileReader

with TsFileReader("data/sock_orange.tsfile") as reader:
    print(reader.get_all_table_schemas()["sock_orange"])

Source & license

• Original dataset: https://huggingface.co/datasets/makermods/sock_orange
• Author: makermods
• License: not declared by the original dataset; please defer to the original.