kuka

  • Sự miêu tả :

Nhiệm vụ nhặt và sắp xếp lại thùng

Tách ra Ví dụ
'train' 580.392
  • Cấu trúc tính năng :
FeaturesDict({
    'steps': Dataset({
        'action': FeaturesDict({
            'base_displacement_vector': Tensor(shape=(2,), dtype=float32),
            'base_displacement_vertical_rotation': Tensor(shape=(1,), dtype=float32),
            'gripper_closedness_action': Tensor(shape=(1,), dtype=float32),
            'rotation_delta': Tensor(shape=(3,), dtype=float32),
            'terminate_episode': Tensor(shape=(3,), dtype=int32),
            'world_vector': Tensor(shape=(3,), dtype=float32),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'clip_function_input/base_pose_tool_reached': Tensor(shape=(7,), dtype=float32),
            'clip_function_input/workspace_bounds': Tensor(shape=(3, 3), dtype=float32),
            'gripper_closed': Tensor(shape=(1,), dtype=float32),
            'height_to_bottom': Tensor(shape=(1,), dtype=float32),
            'image': Image(shape=(512, 640, 3), dtype=uint8),
            'natural_language_embedding': Tensor(shape=(512,), dtype=float32),
            'natural_language_instruction': string,
            'task_id': Tensor(shape=(1,), dtype=float32),
        }),
        'reward': Scalar(shape=(), dtype=float32),
    }),
    'success': bool,
})
  • Tài liệu tính năng :
Tính năng Lớp học Hình dạng loại D Sự miêu tả
Tính năngDict
bước Tập dữ liệu
bước/hành động Tính năngDict
bước/hành động/base_displacement_vector Tenxơ (2,) phao32
bước/hành động/base_displacement_vertical_rotation Tenxơ (1,) phao32
bước/hành động/gripper_closedness_action Tenxơ (1,) phao32
bước/hành động/xoay_delta Tenxơ (3,) phao32
bước/hành động/chấm dứt_episode Tenxơ (3,) int32
bước/hành động/world_vector Tenxơ (3,) phao32
bước/is_first Tenxơ bool
bước/is_last Tenxơ bool
bước/is_terminal Tenxơ bool
bước/quan sát Tính năngDict
bước/quan sát/clip_function_input/base_pose_tool_reached Tenxơ (7,) phao32
bước/quan sát/clip_function_input/workspace_bounds Tenxơ (3, 3) phao32
bước/quan sát/gripper_closed Tenxơ (1,) phao32
bước/quan sát/chiều cao_đến_đáy Tenxơ (1,) phao32
bước/quan sát/hình ảnh Hình ảnh (512, 640, 3) uint8
các bước/quan sát/tự nhiên_ngôn ngữ_embedding Tenxơ (512,) phao32
các bước/quan sát/tự nhiên_ngôn ngữ_hướng dẫn Tenxơ sợi dây
bước/quan sát/task_id Tenxơ (1,) phao32
bước/phần thưởng Vô hướng phao32
thành công Tenxơ bool
  • Trích dẫn :
@article{kalashnikov2018qt,
  title={Qt-opt: Scalable deep reinforcement learning for vision-based robotic manipulation},
  author={Kalashnikov, Dmitry and Irpan, Alex and Pastor, Peter and Ibarz, Julian and Herzog, Alexander and Jang, Eric and Quillen, Deirdre and Holly, Ethan and Kalakrishnan, Mrinal and Vanhoucke, Vincent and others},
  journal={arXiv preprint arXiv:1806.10293},
  year={2018}
}