README.md

---
license: apache-2.0
task_categories:
- robotics
tags:
- LeRobot
- tsfile
- timeseries
- format:tsfile
pretty_name: G1_Dex1_PickPlaceRedBlock_Dataset_Sim
modality: timeseries
configs:
- config_name: default
  data_files:
  - split: train
    path: data/g1_dex1_pickplaceredblock_dataset_sim.tsfile
---

# G1_Dex1_PickPlaceRedBlock_Dataset_Sim (TsFile)

Apache TsFile version of [`unitreerobotics/G1_Dex1_PickPlaceRedBlock_Dataset_Sim`](https://huggingface.co/datasets/unitreerobotics/G1_Dex1_PickPlaceRedBlock_Dataset_Sim).

## Overview

A [LeRobot](https://github.com/huggingface/lerobot) robot-manipulation dataset. The source card is auto-generated ("This dataset was created using LeRobot") and does not add a free-text description; the facts below are taken from the source `meta/info.json`.

- **Robot:** Unitree_G1_Gripper
- **Episodes:** 216
- **Frames:** 163330
- **Sampling rate:** 30.0 fps
- **Splits:** a single `train` split

## Schema (TsFile structure)

- **Time** (INT64, milliseconds) — `round(timestamp * 1000)`; the source `timestamp` column (seconds) is dropped because it equals `Time / 1000`.
- **episode_index** (TAG), **task_index** (TAG) — device dimensions; query one episode with `WHERE episode_index = <n>`.
- **frame_index** (FIELD, INT64), **sample_index** (FIELD, INT64, from the source `index`) — per-frame bookkeeping.
- Vector columns (single-precision FLOAT, source name with `.` → `_` and an element index appended):
  - observation.state → `observation_state_0`..`observation_state_15` (FLOAT)
  - action → `action_0`..`action_15` (FLOAT)

Camera video streams are **not** included in this repository; see the original dataset for the videos: https://huggingface.co/datasets/unitreerobotics/G1_Dex1_PickPlaceRedBlock_Dataset_Sim (`videos/` directory).

## Usage

Install the Apache TsFile Python SDK (`pip install tsfile`) and read a converted file:

```python
from pathlib import Path
from tsfile import TsFileReader

path = Path("data/g1_dex1_pickplaceredblock_dataset_sim.tsfile")
with TsFileReader(str(path)) as reader:
    schemas = reader.get_all_table_schemas()
    print("tables:", list(schemas))
    table_name = next(iter(schemas))
    table = schemas[table_name]
    columns = [column.get_column_name() for column in table.get_columns()]
    print("columns:", columns)
    field_names = [
        column.get_column_name()
        for column in table.get_columns()
        if column.get_column_name() not in {"Time", "time"}
    ]
    if field_names:
        with reader.query_table(table_name, field_names[:3], batch_size=1024) as result:
            batch = result.read_arrow_batch()
            if batch is not None:
                print(batch.to_pandas().head())
```

## Source & license

- Original dataset: https://huggingface.co/datasets/unitreerobotics/G1_Dex1_PickPlaceRedBlock_Dataset_Sim
- Author / publisher: unitreerobotics
- License: apache-2.0