Datasets:
|
Download README.md from THULab/so101_test: direct link, hf CLI and curl.
- Browser
- Download file 2.36 kB
-
https://huggingface.co/datasets/THULab/so101_test/resolve/main/README.md
- Command line
-
hf download hf://datasets/THULab/so101_test/README.md
-
curl -L -o README.md https://huggingface.co/datasets/THULab/so101_test/resolve/main/README.md
2.36 kB
| license: apache-2.0 | |
| task_categories: | |
| - robotics | |
| tags: | |
| - LeRobot | |
| - so101 | |
| - tutorial | |
| - tsfile | |
| - format:tsfile | |
| - time-series | |
| - timeseries | |
| modality: | |
| - tabular | |
| - timeseries | |
| configs: | |
| - config_name: default | |
| data_files: | |
| - split: train | |
| path: data/so101_test.tsfile | |
| size_categories: | |
| - 10K<n<100K | |
| # so101_test (TsFile) | |
| Apache TsFile version of [`Jeevesh2009/so101_test`](https://huggingface.co/datasets/Jeevesh2009/so101_test), a LeRobot v2.1 demonstration dataset recorded with an SO-101 arm. | |
| ## Overview | |
| This dataset was created using [LeRobot](https://github.com/huggingface/lerobot). It records one SO-101 manipulation task with per-frame joint state and action vectors, plus two RealSense camera streams (`realsense1`, `realsense2`). | |
| - **Robot:** SO-101 (`so101`) | |
| - **Scale:** 62 episodes, 34,427 frames, 1 task, 30 fps | |
| - **Split:** train | |
| ## Schema (TsFile structure) | |
| The TsFile table is named `so101_test`. | |
| | Role | Columns | | |
| | --- | --- | | |
| | Time | `Time`, INT64 milliseconds | | |
| | TAG | `episode_index`, `task_index` | | |
| | FIELD | `frame_index`, `sample_index` | | |
| | FIELD | `action_0` … `action_5`, FLOAT | | |
| | FIELD | `observation_state_0` … `observation_state_5`, FLOAT | | |
| The original 6-element `action` and `observation.state` vectors are flattened into scalar FLOAT measurements. 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`. | |
| - The RealSense camera streams (`observation.images.realsense1`, `realsense2`) are video and are NOT included in this repository; they remain in the [source videos tree](https://huggingface.co/datasets/Jeevesh2009/so101_test/tree/main/videos). | |
| - `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 | |
| ```python | |
| from tsfile import TsFileReader | |
| with TsFileReader("data/so101_test.tsfile") as reader: | |
| print(reader.get_all_table_schemas()["so101_test"]) | |
| ``` | |
| ## Source & license | |
| - Original dataset: https://huggingface.co/datasets/Jeevesh2009/so101_test | |
| - Author: Jeevesh2009 | |
| - License: apache-2.0 | |