遇见数据集

eval_insert_gear_in_gripper_13h42m_05-oct-2026

收藏
Hugging Face2026-10-05 更新2026-10-05 收录
官方服务:

资源简介:

该数据集使用LeRobot框架创建,包含双机械臂任务演示数据。机器人类型为bimanual_yam,涉及1个任务(从文件名推断为将齿轮插入夹爪),共5个演示片段,总计4155帧。每个样本包含14维动作(左右机械臂关节位置)、14维状态、三个摄像头图像(左腕、右腕和顶部,分辨率720x1280,帧率20fps)以及时间戳、帧索引等元数据。数据适用于机器人模仿学习等场景。

This dataset is created using the LeRobot framework and contains demonstration data for a dual-arm robot task. The robot type is bimanual_yam, involving one task (inferred from the filename as inserting a gear into the gripper), with 5 demonstration episodes totaling 4155 frames. Each sample includes 14-dimensional actions (joint positions of left and right arms), 14-dimensional states, three camera images (left wrist, right wrist, and top, with resolution 720x1280 and frame rate 20fps), as well as metadata such as timestamps and frame indices. The data is suitable for robot imitation learning and similar scenarios.

提供机构:
fanqi-robo
创建时间:
2026-10-05
原始信息汇总

数据集概述

基本信息

  • 数据集地址:https://huggingface.co/datasets/fanqi-robo/eval_insert_gear_in_gripper_13h42m_05-oct-2026
  • 许可证:apache-2.0
  • 任务类别:robotics(机器人)
  • 标签:LeRobot
  • 创建工具:LeRobot(https://github.com/huggingface/lerobot)
  • 可视化入口:https://huggingface.co/spaces/lerobot/visualize_dataset?path=fanqi-robo/eval_insert_gear_in_gripper_13h42m_05-oct-2026

数据集描述

  • Homepage:[More Information Needed]
  • Paper:[More Information Needed]
  • License:apache-2.0

数据集结构

配置信息

  • 配置名称:default
  • 数据文件:data//.parquet

元信息(meta/info.json)

项目 值
codebase_version v3.0
robot_type lerobot/bimanual_yam
total_episodes 5
total_frames 4155
total_tasks 1
chunks_size 1000
data_files_size_in_mb 100
video_files_size_in_mb 200
fps 20
splits train: 0:5
data_path data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet
video_path videos/{video_key}/chunk-{chunk_index:03d}/file-{file_index:03d}.mp4

特征(Features)

action(动作)

  • dtype:float32
  • shape:[14]
  • names:
    • left_shoulder_pan.pos
    • left_shoulder_lift.pos
    • left_elbow_flex.pos
    • left_wrist_flex.pos
    • left_wrist_roll.pos
    • left_wrist_yaw.pos
    • left_gripper.pos
    • right_shoulder_pan.pos
    • right_shoulder_lift.pos
    • right_elbow_flex.pos
    • right_wrist_flex.pos
    • right_wrist_roll.pos
    • right_wrist_yaw.pos
    • right_gripper.pos

observation.state(观测状态)

  • dtype:float32
  • shape:[14]
  • names:与 action 相同(左右肩、肘、腕、夹爪各关节位置)

observation.images.left_wrist(左腕图像)

  • dtype:video
  • shape:[720, 1280, 3]
  • names:height, width, channels
  • info:
    • video.height:720
    • video.width:1280
    • video.codec:av1
    • video.pix_fmt:yuv420p
    • video.is_depth_map:false
    • video.fps:20
    • video.channels:3
    • has_audio:false

observation.images.right_wrist(右腕图像)

  • dtype:video
  • shape:[720, 1280, 3]
  • names:height, width, channels
  • info:
    • video.height:720
    • video.width:1280
    • video.codec:av1
    • video.pix_fmt:yuv420p
    • video.is_depth_map:false
    • video.fps:20
    • video.channels:3
    • has_audio:false

observation.images.top(顶部图像)

  • dtype:video
  • shape:[720, 1280, 3]
  • names:height, width, channels
  • info:
    • video.height:720
    • video.width:1280
    • video.codec:av1
    • video.pix_fmt:yuv420p
    • video.is_depth_map:false
    • video.fps:20
    • video.channels:3
    • has_audio:false

timestamp(时间戳)

  • dtype:float32
  • shape:[1]
  • names:null

frame_index(帧索引)

  • dtype:int64
  • shape:[1]
  • names:null

episode_index(回合索引)

  • dtype:int64
  • shape:[1]
  • names:null

index(索引)

  • dtype:int64
  • shape:[1]
  • names:null

task_index(任务索引)

  • dtype:int64
  • shape:[1]
  • names:null

引用

BibTeX:

bibtex [More Information Needed]

二维码
社区交流群
二维码
科研交流群
商业服务