Dataset Viewer
Auto-converted to Parquet Duplicate
codebase_version
string
robot_type
null
total_episodes
int64
total_frames
int64
total_tasks
int64
total_videos
int64
total_chunks
int64
chunks_size
int64
fps
int64
splits
dict
data_path
string
features
dict
video_path_original
string
tsfile_conversion
dict
v2.1
null
103
13,567
1
206
1
1,000
10
{ "train": "0:103" }
data/ur5e_pick_red_cube.tsfile
{ "Time": { "dtype": "int64", "shape": [ 1 ], "tsfile_role": "TIME", "unit": "ms" }, "episode_index": { "dtype": "int64", "shape": [ 1 ], "tsfile_role": "TAG" }, "task_index": { "dtype": "int64", "shape": [ 1 ], "tsfile_role": "TAG" }, ...
videos/chunk-{episode_chunk:03d}/{video_key}/episode_{episode_index:06d}.mp4
{ "source_dataset": "tlpss/ur5e-pick-red-cube", "source_data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet", "converted_data_path": "data/ur5e_pick_red_cube.tsfile", "table_name": "ur5e_pick_red_cube", "granularity": "merged", "time_precision": "ms", "time_mapping": { "sou...

UR5e Pick Red Cube (TsFile)

Apache TsFile version of tlpss/ur5e-pick-red-cube.

Overview

A UR5e robot-arm manipulation dataset in the LeRobot v2.1 format. The single task is to pick up a red cube and place it on a blue square. Each record is one control frame within a demonstration episode, holding the robot's proprioceptive state, the commanded action, and per-step reward / success flags.

  • Scale: 103 episodes, 13,567 frames total (≈ 132 frames per episode on average).
  • Sampling rate: 10 fps.
  • Tasks: 1 — "pick up the red cube and place it on the blue square".
  • Robot: UR5e arm with a gripper.

Schema (TsFile structure)

  • Time (INT64, milliseconds) — round(timestamp * 1000), restarting at 0 for each episode.
  • episode_index (TAG) — source episode id (0–102).
  • task_index (TAG) — source task id.
  • frame_index (FIELD, INT64) — frame position within the episode.
  • sample_index (FIELD, INT64) — the source global index column, renamed.
  • seed (FIELD, INT64) — episode random seed.
  • next_reward (FIELD, FLOAT) — reward at the next step.
  • next_success (FIELD, BOOLEAN) — task-success flag.
  • gripper_state (FIELD, FLOAT) — scalar gripper state.
  • observation_state_0 .. observation_state_6 (FIELD, FLOAT) — the 7-D observation state, flattened.
  • robot_pose_0 .. robot_pose_5 (FIELD, FLOAT) — the 6-D robot pose, flattened.
  • joints_0 .. joints_5 (FIELD, FLOAT) — the 6-D joint vector, flattened.
  • action_0 .. action_9 (FIELD, FLOAT) — the 10-D action, flattened.

All 103 episodes share a single .tsfile; episode_index and task_index are the device (TAG) dimensions. Query one episode with WHERE episode_index=0.

Vector columns are flattened keeping the source name (e.g. observation.stateobservation_state_0..6, actionaction_0..9). The source timestamp column is dropped because it equals Time / 1000 seconds; frame_index is kept.

Usage

Read the .tsfile files with the Apache TsFile Java or Python SDK.

Source & license

  • Original dataset: https://huggingface.co/datasets/tlpss/ur5e-pick-red-cube
  • Author / publisher: tlpss (Hugging Face)
  • Created with LeRobot.
  • Camera videos (observation.images.wrist_image, observation.images.scene_image, 240×320 RGB @ 10 fps) are not included in this repository; see the original dataset's videos/ directory for them.
  • License: apache-2.0.
Downloads last month
61