| --- |
| license: apache-2.0 |
| task_categories: |
| - robotics |
| tags: |
| - LeRobot |
| - tutorial |
| - tsfile |
| - timeseries |
| - robotics |
| - so100 |
| - format:tsfile |
| modality: |
| - tabular |
| - timeseries |
| pretty_name: Test NNNN TsFile |
| configs: |
| - config_name: default |
| data_files: |
| - split: train |
| path: data/test_nnnn.tsfile |
| size_categories: |
| - 1K<n<10K |
| --- |
| |
| # Test NNNN TsFile |
|
|
| This dataset is converted from [`FeiYjf/Test_NNNN`](https://huggingface.co/datasets/FeiYjf/Test_NNNN) to Apache TsFile format. |
|
|
| Modalities: Time-series. The converted repository contains numeric robot state/action trajectories and LeRobot metadata. Camera videos are not included in this TsFile conversion; they remain in the original Hugging Face dataset under its `videos/` tree. |
|
|
| ## Source Dataset |
|
|
| - Original dataset: [`FeiYjf/Test_NNNN`](https://huggingface.co/datasets/FeiYjf/Test_NNNN) |
| - License: `apache-2.0` |
| - Robot type from source metadata: `so100` |
| - Source codebase version from metadata: `v2.0` |
| - Source split: `train`, episodes `0:40` |
| - Sampling rate: 30 fps |
| - Source scale: 40 episodes, 8,226 frames, 1 task, and 80 source videos. |
| - Source data layout: `data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet` |
| - Source video layout: `videos/chunk-{episode_chunk:03d}/{video_key}/episode_{episode_index:06d}.mp4` |
| - Source card note: created using LeRobot. |
|
|
| ## Converted Data |
|
|
| - TsFile path: `data/test_nnnn.tsfile` |
| - Table name: `test_nnnn` |
| - Converted rows: 8,226 |
| - Episodes: 40 |
| - Tasks: 1 |
| - Source frame Parquet files converted: 40 |
| - Output file size: 117,979 bytes |
| - Time precision: milliseconds |
|
|
| ## Schema |
|
|
| - `Time`: integer timestamp in milliseconds, computed as `round(timestamp * 1000)`. |
| - TAG columns: `episode_index`, `task_index`. |
| - Integer FIELD columns: `frame_index`, `sample_index`. |
| - FLOAT FIELD columns: `action_0` through `action_5`. |
| - FLOAT FIELD columns: `observation_state_0` through `observation_state_5`. |
|
|
| The six action and observation-state dimensions follow the SO100 joint names from the source metadata: |
|
|
| - `main_shoulder_pan` |
| - `main_shoulder_lift` |
| - `main_elbow_flex` |
| - `main_wrist_flex` |
| - `main_wrist_roll` |
| - `main_gripper` |
|
|
| ## Conversion Notes |
|
|
| - Vector columns are flattened while preserving source names: `action` becomes `action_0..action_5`, and `observation.state` becomes `observation_state_0..observation_state_5`. |
| - The source `timestamp` column is dropped because it is represented by `Time / 1000` seconds after conversion. |
| - The source `index` column is renamed to `sample_index`. |
| - Video features `observation.images.left` and `observation.images.right` are omitted from this repository. Use the original dataset videos at [`FeiYjf/Test_NNNN/videos`](https://huggingface.co/datasets/FeiYjf/Test_NNNN/tree/main/videos). |
| - Source `meta/` files are mirrored, with `meta/info.json` updated to describe the converted TsFile path and schema. |
|
|
| ## Read Example |
|
|
| ```python |
| from pathlib import Path |
| |
| tsfile_path = Path("data/test_nnnn.tsfile") |
| print(tsfile_path) |
| |
| # Read this file with the Apache TsFile Java/Python SDK or TsFile tools. |
| ``` |
|
|
