همگی

  • شرح :

فرانکا اسکوپ زدن و ریختن وظایف

  • صفحه اصلی : https://toto-benchmark.org/

  • کد منبع : tfds.robotics.rtx.Toto

  • نسخه ها :

    • 0.1.0 (پیش فرض): انتشار اولیه.
  • اندازه دانلود : Unknown size

  • اندازه مجموعه داده : Unknown size

  • ذخیره خودکار ( اسناد ): ناشناخته

  • تقسیم ها :

شکاف مثال ها
  • ساختار ویژگی :
FeaturesDict({
    'steps': Dataset({
        'action': FeaturesDict({
            'open_gripper': bool,
            'rotation_delta': Tensor(shape=(3,), dtype=float32),
            'terminate_episode': float32,
            'world_vector': Tensor(shape=(3,), dtype=float32),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'image': Image(shape=(480, 640, 3), dtype=uint8),
            'natural_language_embedding': Tensor(shape=(512,), dtype=float32),
            'natural_language_instruction': string,
            'state': Tensor(shape=(7,), dtype=float32),
        }),
        'reward': Scalar(shape=(), dtype=float32),
    }),
})
  • مستندات ویژگی :
ویژگی کلاس شکل نوع D شرح
FeaturesDict
مراحل مجموعه داده
مراحل/عمل FeaturesDict
Steps/Action/open_gripper تانسور بوول
Steps/Action/rotation_delta تانسور (3،) float32
مراحل/عمل/پایان_اپیزود تانسور float32
Steps/Action/world_vector تانسور (3،) float32
Steps/is_first تانسور بوول
Steps/is_last تانسور بوول
Steps/is_terminal تانسور بوول
مراحل / مشاهده FeaturesDict
مراحل / مشاهده / تصویر تصویر (480، 640، 3) uint8
steps/observation/natural_language_embedding تانسور (512،) float32
مراحل/مشاهده/آموزش_زبان_طبیعی تانسور رشته
مراحل / مشاهده / حالت تانسور (7،) float32 آرایه ناتوان از شکل (7،). شامل حالات مفصل ربات (به عنوان زوایای مفصل مطلق) در هر مرحله زمانی است
مراحل/پاداش اسکالر float32
@inproceedings{zhou2023train,
  author={Zhou, Gaoyue and Dean, Victoria and Srirama, Mohan Kumar and Rajeswaran, Aravind and Pari, Jyothish and Hatch, Kyle and Jain, Aryan and Yu, Tianhe and Abbeel, Pieter and Pinto, Lerrel and Finn, Chelsea and Gupta, Abhinav},
  booktitle={2023 IEEE International Conference on Robotics and Automation (ICRA)},
  title={Train Offline, Test Online: A Real Robot Learning Benchmark},
  year={2023},
 }