usc_cloth_sim_converted_externally_to_rlds

  • বর্ণনা :

ফ্রাঙ্কা কাপড় মিথস্ক্রিয়া কাজ

বিভক্ত উদাহরণ
'train' 800
'val' 200
  • বৈশিষ্ট্য গঠন :
FeaturesDict({
    'episode_metadata': FeaturesDict({
        'file_path': Text(shape=(), dtype=string),
    }),
    'steps': Dataset({
        'action': Tensor(shape=(4,), dtype=float32),
        'discount': Scalar(shape=(), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'language_embedding': Tensor(shape=(512,), dtype=float32),
        'language_instruction': Text(shape=(), dtype=string),
        'observation': FeaturesDict({
            'image': Image(shape=(32, 32, 3), dtype=uint8),
        }),
        'reward': Scalar(shape=(), dtype=float32),
    }),
})
  • বৈশিষ্ট্য ডকুমেন্টেশন :
বৈশিষ্ট্য ক্লাস আকৃতি ডিটাইপ বর্ণনা
ফিচারসডিক্ট
episode_metadata ফিচারসডিক্ট
episode_metadata/file_path পাঠ্য স্ট্রিং মূল ডেটা ফাইলের পথ।
পদক্ষেপ ডেটাসেট
পদক্ষেপ/ক্রিয়া টেনসর (৪,) float32 রোবট অ্যাকশন, x,y,z গোল এবং পিকার কমান্ডপিকার <0.5 = খোলা, পিকার>0.5 = বন্ধ নিয়ে গঠিত।
পদক্ষেপ/ছাড় স্কেলার float32 ডিসকাউন্ট দেওয়া হলে, ডিফল্ট 1.
steps/is_first টেনসর bool
ধাপ/শেষ_শেষ টেনসর bool
steps/is_terminal টেনসর bool
পদক্ষেপ/ভাষা_এম্বেডিং টেনসর (512,) float32 কোন ভাষা এম্বেডিং. https://tfhub.dev/google/universal-sentence-encoder-large/5 দেখুন
পদক্ষেপ/ভাষা_নির্দেশ পাঠ্য স্ট্রিং ভাষার নির্দেশনা।
পদক্ষেপ/পর্যবেক্ষণ ফিচারসডিক্ট
পদক্ষেপ/পর্যবেক্ষণ/চিত্র ছবি (৩২, ৩২, ৩) uint8 কাপড়ের ছবি পর্যবেক্ষণ।
পদক্ষেপ/পুরস্কার স্কেলার float32 [0, 1].0 এ একটি স্বাভাবিক কর্মক্ষমতা মেট্রিক হিসাবে পুরষ্কার = প্রাথমিক অবস্থা থেকে কোন পরিবর্তন নেই। 1 = নিখুঁত ভাঁজ।-ve পারফরম্যান্স মানে কাপড়টি প্রাথমিক অবস্থার চেয়ে খারাপ।
  • তত্ত্বাবধান করা কী (দেখুন as_supervised doc ): None

  • চিত্র ( tfds.show_examples ): সমর্থিত নয়।

  • উদাহরণ ( tfds.as_dataframe ): অনুপস্থিত।

  • উদ্ধৃতি :

@article{salhotra2022dmfd,
    author={Salhotra, Gautam and Liu, I-Chun Arthur and Dominguez-Kuhne, Marcus and Sukhatme, Gaurav S.},
    journal={IEEE Robotics and Automation Letters},
    title={Learning Deformable Object Manipulation From Expert Demonstrations},
    year={2022},
    volume={7},
    number={4},
    pages={8775-8782},
    doi={10.1109/LRA.2022.3187843}
}