eval_LAVLA_Async_VI_100M_40k_b20_whiteblackcube_v2_2026_05_27_white_latency_bench_v6

该数据集由LeRobot创建，专门用于机器人任务。它包含1个episode，总计586帧，帧率为30fps。数据以parquet文件格式存储，视频则以mp4格式存储。特征涵盖动作（包括6个关节位置：肩部平移、肩部提升、肘部弯曲、腕部弯曲、腕部旋转和夹持器位置）、观察状态（与动作相同的6个关节位置）、三个摄像头图像（每个图像分辨率为480x640，3通道），以及时间戳、帧索引、episode索引、索引和任务索引等元数据。该数据集适用于机器人控制与视觉学习任务，支持相关研究和应用开发。This dataset is created by LeRobot for robotic tasks. It contains 1 episode with a total of 586 frames at a frame rate of 30fps. The data is stored in parquet file format, and videos are stored in mp4 format. Features include actions (6 joint positions: shoulder translation, shoulder elevation, elbow flexion, wrist flexion, wrist rotation, and gripper position), observation states (the same 6 joint positions), three camera images (each with a resolution of 480x640, 3 channels), as well as timestamps, frame indices, episode indices, indices, and task indices. The dataset is suitable for robot control and visual learning tasks.

原始信息汇总

数据集概览

• 数据集名称: eval_LAVLA_Async_VI_100M_40k_b20_whiteblackcube_v2_2026_05_27_white_latency_bench_v6
• 许可证: Apache-2.0
• 任务类别: 机器人学 (Robotics)
• 创建工具: 基于 LeRobot (https://github.com/huggingface/lerobot) 创建
• 可视化链接: 可通过 LeRobot 可视化数据集空间进行查看 (https://huggingface.co/spaces/lerobot/visualize_dataset?path=Alkatt/eval_LAVLA_Async_VI_100M_40k_b20_whiteblackcube_v2_2026_05_27_white_latency_bench_v6)

数据集结构

• 总剧集数 (Total Episodes): 1
• 总帧数 (Total Frames): 586
• 总任务数 (Total Tasks): 1
• 帧率 (FPS): 30
• 数据分块大小: 1000
• 数据文件大小: 100 MB
• 视频文件大小: 200 MB
• 机器人类型: so_follower
• 数据集划分: 仅包含训练集 (train: "0:1")
• 数据文件路径: data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet
• 视频文件路径: videos/{video_key}/chunk-{chunk_index:03d}/file-{file_index:03d}.mp4

特征 (Features)

• 动作 (action): 浮点32位 (float32), 形状 [6], 包含6个关节位置 (shoulder_pan.pos, shoulder_lift.pos, elbow_flex.pos, wrist_flex.pos, wrist_roll.pos, gripper.pos)
• 观察状态 (observation.state): 浮点32位 (float32), 形状 [6], 包含6个关节位置 (shoulder_pan.pos, shoulder_lift.pos, elbow_flex.pos, wrist_flex.pos, wrist_roll.pos, gripper.pos)
• 观察图像 (observation.images): 包含三个摄像头 (camera1, camera2, camera3)
  • 数据类型: 视频 (video)
  • 图像尺寸: 480×640 像素
  • 颜色通道: 3 (RGB)
  • 视频编码: AV1
  • 像素格式: yuv420p
  • 帧率: 30 FPS
  • 非深度图 (is_depth_map: false)
• 时间戳 (timestamp): 浮点32位 (float32), 形状 [1]
• 帧索引 (frame_index): 整型64位 (int64), 形状 [1]
• 剧集索引 (episode_index): 整型64位 (int64), 形状 [1]
• 索引 (index): 整型64位 (int64), 形状 [1]
• 任务索引 (task_index): 整型64位 (int64), 形状 [1]