THULab/robomimic_sim_transport

Robomimic Sim Transport TsFile

This dataset is a TsFile conversion of the Hugging Face dataset iantc104/robomimic_sim_transport, which was created using LeRobot.

Modalities: Time-series. The original dataset also contains three simulated camera video streams. Those videos are not included in this repository and remain available in the source dataset.

Source Dataset

• Original dataset: iantc104/robomimic_sim_transport
• License: Apache-2.0
• Task category: robotics
• Task: Transport the object
• LeRobot codebase version: v2.1
• Split: single train split (0:164)
• Scale: 164 episodes, 76,007 frames, 1 task, 164 source frame Parquet files
• Sampling rate: 20 fps
• Original video features: observation.images.agentview_image, observation.images.robot0_eye_in_hand_image, and observation.images.robot1_eye_in_hand_image (256 x 256 RGB, AV1, 20 fps)

The source metadata does not specify a robot type, and the source dataset card does not provide a paper, homepage, or detailed task protocol beyond the LeRobot provenance and task metadata above.

Converted Files

• TsFile: data/robomimic_sim_transport.tsfile
• Rows: 76,007
• Episodes: 164
• Tasks: 1 (task_index = 0)
• Table name: robomimic_sim_transport
• Time precision: milliseconds
• Metadata: meta/ is mirrored from the source dataset, with meta/info.json rewritten to describe the TsFile artifact and conversion mapping.

Schema

Column	Role	Type	Notes
Time	TIME	INT64	round(timestamp * 1000), in milliseconds; restarts per episode
episode_index	TAG	INT64	Source episode identifier
task_index	TAG	INT64	Source task identifier
frame_index	FIELD	INT64	Source frame index, preserved
sample_index	FIELD	INT64	Renamed from source index
action_0 ... action_13	FIELD	FLOAT	Flattened from action[14]
action_delta_0 ... action_delta_13	FIELD	FLOAT	Flattened from action.delta[14]
action_absolute_0 ... action_absolute_13	FIELD	FLOAT	Flattened from action.absolute[14]
observation_state_0 ... observation_state_85	FIELD	FLOAT	Flattened from observation.state[86]
observation_environment_state_0 ... observation_environment_state_40	FIELD	FLOAT	Flattened from observation.environment_state[41]

The source metadata does not provide element names for these vectors. Their original order is preserved by the numeric suffixes.

episode_index and task_index are TAG columns, so an individual episode can be selected by filtering on episode_index.

Conversion Notes

• The 164 LeRobot frame Parquet files under data/chunk-000/ were merged into one TsFile for the train split.
• Vector columns were flattened by preserving the source column name, replacing . with _, and appending the element index.
• The source timestamp column is dropped because it is redundant with Time / 1000 seconds.
• The source index column is renamed to sample_index.
• The three camera video features are omitted from the converted schema and are not uploaded here. Use the original dataset videos: https://huggingface.co/datasets/iantc104/robomimic_sim_transport/tree/main/videos

Read Example

from tsfile import TsFileReader

path = "data/robomimic_sim_transport.tsfile"
table = "robomimic_sim_transport"

with TsFileReader(path) as reader:
    columns = [
        "episode_index",
        "task_index",
        "frame_index",
        "sample_index",
        "action_0",
        "observation_state_0",
        "observation_environment_state_0",
    ]
    with reader.query_table(table, columns, batch_size=4096) as result:
        batch = result.read_arrow_batch()
        print(batch)

Citation

The original dataset card does not provide a BibTeX citation.