THULab/finetune_green_100

Finetune Green 100 (TsFile)

Apache TsFile version of radiuson/finetune_green_100. Modalities: Time-series.

Overview

This LeRobot SO101 dataset contains demonstrations for putting a green cube into a box.

Dataset Scale

• 39 episodes, 9,057 frames, one task, sampled at 30 fps.
• The train split contains 9,057 TsFile rows in data/finetune_green_100_train.tsfile.

Conversion details

Time is round(timestamp * 1000) in milliseconds. The source timestamp column is therefore not stored as a separate FIELD; frame_index is retained and source index is named sample_index. Camera videos are not included in this time-series repository and remain in the original dataset.

Schema (TsFile structure)

• Time (INT64, milliseconds) is the per-episode sample time.
• episode_index and task_index are TAG columns identifying the device series.
• frame_index and sample_index are INT64 FIELD columns.
• action_0 through _5 and observation_state_0 through _5 are FLOAT FIELD columns for the SO101 joints.

For example, select one device with WHERE episode_index='0' AND task_index='0'.

Usage

Read data/finetune_green_100_train.tsfile with the Apache TsFile Java or Python SDK.

Source & license

• Original dataset: https://huggingface.co/datasets/radiuson/finetune_green_100
• Author / publisher: radiuson
• License: Apache-2.0