---
license: apache-2.0
authors:
  - Henry-Ellis
  - lv-1
task_categories:
  - robotics
tags:
  - tsfile
  - timeseries
  - tabular
modality:
  - timeseries
  - tabular
pretty_name: G1 Dex3 Grasp Square Dataset TsFile
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/g1_dex3_graspsquare_dataset_train.tsfile
size_categories:
  - 100K<n<1M
---

# G1 Dex3 Grasp Square Dataset TsFile

This repository is an Apache TsFile conversion of [unitreerobotics/G1_Dex3_GraspSquare_Dataset](https://huggingface.co/datasets/unitreerobotics/G1_Dex3_GraspSquare_Dataset), a LeRobot v3 robot-manipulation dataset. It contains numeric trajectories and metadata; camera videos remain in the original Hugging Face dataset.

## Source dataset

- Original dataset: [unitreerobotics/G1_Dex3_GraspSquare_Dataset](https://huggingface.co/datasets/unitreerobotics/G1_Dex3_GraspSquare_Dataset)
- Original uploader/contributors: [Henry-Ellis](https://huggingface.co/Henry-Ellis) and [lv-1](https://huggingface.co/lv-1); hosted by [Unitree Robotics](https://huggingface.co/unitreerobotics).
- License: Apache-2.0. The source card provides no paper or formal citation.
- Task: stack three 5 cm cubic blocks from bottom to top in red, yellow, blue order on black tape.
- Robot: 7-DOF dual-arm `Unitree_G1` with three-finger dexterous hands.
- Sampling: 30 Hz; 301 episodes, 281,196 frames, one task, one source Parquet shard.
- Source numeric layout: `data/chunk-000/file-000.parquet`.

## Converted files

- TsFile: `data/g1_dex3_graspsquare_dataset_train.tsfile`
- Table: `g1_dex3_graspsquare_dataset_train`
- Rows: 281,196; devices: 301 (one per `episode_index`/`task_index` TAG combination).
- Time precision: milliseconds; `Time = round(timestamp * 1000)` and restarts at the first frame of each episode.
- Size: source Parquet 63,247,041 bytes; flattened/staged Parquet 74,725,871 bytes; TsFile 51,443,386 bytes. The TsFile is smaller than both source and staged representations.

## TsFile schema

| Role | Columns | Source/type |
| --- | --- | --- |
| TIME | `Time` | `int64`, milliseconds from `timestamp` |
| TAG/device | `episode_index`, `task_index` | source `int64`, stored by TsFile device/tag mechanism as string segments |
| FIELD | `frame_index`, `sample_index` | `int64`; `sample_index` is renamed from source `index` |
| FIELD | `observation_state_0` … `observation_state_27` | 28 FLOAT dimensions from `observation.state` |
| FIELD | `action_0` … `action_27` | 28 FLOAT dimensions from `action` |

Vector names retain their source prefix (`.` becomes `_`) and are flattened row-major to scalar fields. `frame_index` is preserved for video/frame alignment. The redundant source `timestamp` column is dropped after deriving `Time`; no numeric rows or vector dimensions are dropped. The four video columns are intentionally omitted from TsFile.

### Encoding and compression

| TsFile type | Encoding | Compression |
| --- | --- | --- |
| FLOAT/DOUBLE | GORILLA | LZ4 |
| INT32/INT64 | TS_2DIFF | LZ4 |
| Time | TS_2DIFF | LZ4 |
| BOOLEAN (if present) | RLE | LZ4 |
| TAG | TsFile device/tag mechanism | TsFile-managed |

## Videos and frame alignment

Videos are not copied or uploaded with this TsFile dataset. The original repository's `videos/` tree is about 9.12 GB and contains four streams:

- [cam_left_high](https://huggingface.co/datasets/unitreerobotics/G1_Dex3_GraspSquare_Dataset/tree/main/videos/observation.images.cam_left_high)
- [cam_right_high](https://huggingface.co/datasets/unitreerobotics/G1_Dex3_GraspSquare_Dataset/tree/main/videos/observation.images.cam_right_high)
- [cam_left_wrist](https://huggingface.co/datasets/unitreerobotics/G1_Dex3_GraspSquare_Dataset/tree/main/videos/observation.images.cam_left_wrist)
- [cam_right_wrist](https://huggingface.co/datasets/unitreerobotics/G1_Dex3_GraspSquare_Dataset/tree/main/videos/observation.images.cam_right_wrist)

The source path template is `videos/{video_key}/chunk-{chunk_index:03d}/file-{file_index:03d}.mp4`. Numeric rows remain frame-aligned through `episode_index` and `frame_index`, with episode metadata under `meta/episodes/`.

## Validation

Local validation read the TsFile with the bundled TsFile Python SDK and compared it with the staged Parquet: 281,196 metadata rows and 281,196 query rows, 301 devices, unique `(episode_index, task_index, Time)` keys, monotonic Time within each episode, and all 28 action plus 28 observation-state dimensions present. Detailed local reports are `validation_report.json` and `VALIDATION.md`.

## Usage

```python
from tsfile import TsFileReader

path = "data/g1_dex3_graspsquare_dataset_train.tsfile"
reader = TsFileReader(path)
table = "g1_dex3_graspsquare_dataset_train"
columns = ["episode_index", "task_index", "frame_index", "sample_index", "action_0", "observation_state_0"]
with reader.query_table(table, columns, batch_size=65536) as result:
    batch = result.read_arrow_batch()
    print(batch.to_pandas().head())
```