Hugging Face's logo

Datasets:
azithromycin
/
356c_so100_nut_test_11

Tasks:
Robotics
Modalities:
Time-series
Tabular
Size:
10K<n<100K
Tags:
tsfile
timeseries
tabular
License:
Dataset card Files
xet
 Community
356c_so100_nut_test_11 / README.md
azithromycin's picture
azithromycin
Add converted TsFile dataset from 356c/so100_nut_test_11
2402d4c verified
preview code
|
Raw
History Blame Contribute Delete
4.22 kB
metadata 
license: apache-2.0
authors:
  - 356c
task_categories:
  - robotics
tags:
  - tsfile
  - timeseries
  - tabular
modality:
  - timeseries
  - tabular
pretty_name: SO100 Nut Test 11 TsFile
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/356c_so100_nut_test_11.tsfile
size_categories:
  - 10K<n<100K

SO100 Nut Test 11 TsFile

This dataset is an Apache TsFile conversion of 356c/so100_nut_test_11, a LeRobot v2.1 SO100 robot-manipulation dataset for picking up a nut and dropping it in a cup.

Source Dataset and Attribution

  • Original dataset: 356c/so100_nut_test_11
  • Original author, repository owner, uploader, and sole contributor: 356c
  • License: Apache-2.0
  • Task: "Pick up nut and drop in cup."
  • Robot: so100; LeRobot version: v2.1
  • Split: train; sampling rate: 30 fps
  • Scale: 100 episodes, 38,691 frames, 1 task
  • Source shards: 100 Parquet files totaling 2,190,148 bytes
  • Source layout: data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet
  • The source card provides no paper, external homepage, or completed citation.

TsFile Data

  • Path: data/356c_so100_nut_test_11.tsfile
  • Table: 356c_so100_nut_test_11
  • Rows: 38,691; episodes/devices: 100; tasks: 1
  • Time precision: milliseconds
  • TsFile/source-Parquet size ratio: 0.330

Schema and Mapping

Time = round(timestamp * 1000) milliseconds. Time starts at zero and is strictly increasing in every episode. The source timestamp is omitted after conversion because it is recoverable as Time / 1000 seconds.

TsFile column Role Type Source mapping
Time TIME TIMESTAMP round(timestamp * 1000) ms
episode_index TAG STRING Original INT64 episode index
task_index TAG STRING Original INT64 task index
frame_index FIELD INT64 Preserved
sample_index FIELD INT64 Renamed from index
action_0 ... action_5 FIELD FLOAT Flattened from action[6]
observation_state_0 ... observation_state_5 FIELD FLOAT Flattened from observation.state[6]

The six vector dimensions are main_shoulder_pan, main_shoulder_lift, main_elbow_flex, main_wrist_flex, main_wrist_roll, and main_gripper. Dots in source names are replaced by underscores. No numeric row, episode, task, action dimension, or state dimension is dropped.

Encodings and Compression

  • FLOAT/DOUBLE: GORILLA + LZ4
  • INT32/INT64: TS_2DIFF + LZ4
  • Time: TS_2DIFF + LZ4
  • BOOLEAN: RLE + LZ4 (the source has no BOOLEAN field)
  • TAG: TsFile table/device TAG storage

Videos

Videos are not included in this TsFile dataset. The 300 original AV1 MP4 files remain in the source repository, 640x480 at 30 fps with no audio, in three streams with 100 episode files each:

The source layout is videos/chunk-{episode_chunk:03d}/{video_key}/episode_{episode_index:06d}.mp4. Use episode_index and frame_index to align numeric rows with video frames.

Minimal Read Example 

from tsfile import TsFileReader

reader = TsFileReader("data/356c_so100_nut_test_11.tsfile")
with reader.query_table(
    "356c_so100_nut_test_11",
    ["episode_index", "task_index", "frame_index", "sample_index",
     "action_0", "observation_state_0"],
    batch_size=65536,
) as result:
    batch = result.read_arrow_batch()
    print(batch.to_pandas().head())
reader.close()

Citation

The source card supplies no paper or completed citation. Cite the original Hugging Face dataset and its publisher, 356c, when using this conversion.