robomimic_mg

  • विवरण :

रोबोमिमिक मशीन से उत्पन्न डेटासेट को घने इनाम के साथ प्रशिक्षित सॉफ्ट एक्टर क्रिटिक एजेंट का उपयोग करके एकत्र किया गया था। प्रत्येक डेटासेट में एजेंट का रीप्ले बफ़र होता है।

प्रत्येक कार्य के दो संस्करण होते हैं: एक निम्न आयामी अवलोकनों ( low_dim ) के साथ, और दूसरा छवियों ( image ) के साथ।

डेटासेट चरणों और एपिसोड का प्रतिनिधित्व करने के लिए आरएलडीएस प्रारूप का पालन करते हैं।

@inproceedings{robomimic2021,
  title={What Matters in Learning from Offline Human Demonstrations for Robot Manipulation},
  author={Ajay Mandlekar and Danfei Xu and Josiah Wong and Soroush Nasiriany
          and Chen Wang and Rohun Kulkarni and Li Fei-Fei and Silvio Savarese
          and Yuke Zhu and Roberto Mart\'{i}n-Mart\'{i}n},
  booktitle={Conference on Robot Learning},
  year={2021}
}

robomimic_mg/lift_mg_image (डिफ़ॉल्ट कॉन्फ़िगरेशन)

  • डाउनलोड आकार : 18.04 GiB

  • डेटासेट का आकार : 2.73 GiB

  • ऑटो-कैश्ड ( दस्तावेज़ीकरण ): नहीं

  • विभाजन :

विभाजित करना उदाहरण
'train' 1,500
  • फ़ीचर संरचना :
FeaturesDict({
    'episode_id': string,
    'horizon': int32,
    'steps': Dataset({
        'action': Tensor(shape=(7,), dtype=float64),
        'discount': int32,
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'agentview_image': Image(shape=(84, 84, 3), dtype=uint8),
            'object': Tensor(shape=(10,), dtype=float64),
            'robot0_eef_pos': Tensor(shape=(3,), dtype=float64, description=End-effector position),
            'robot0_eef_quat': Tensor(shape=(4,), dtype=float64, description=End-effector orientation),
            'robot0_eef_vel_ang': Tensor(shape=(3,), dtype=float64, description=End-effector angular velocity),
            'robot0_eef_vel_lin': Tensor(shape=(3,), dtype=float64, description=End-effector cartesian velocity),
            'robot0_eye_in_hand_image': Image(shape=(84, 84, 3), dtype=uint8),
            'robot0_gripper_qpos': Tensor(shape=(2,), dtype=float64, description=Gripper position),
            'robot0_gripper_qvel': Tensor(shape=(2,), dtype=float64, description=Gripper velocity),
            'robot0_joint_pos': Tensor(shape=(7,), dtype=float64, description=7DOF joint positions),
            'robot0_joint_pos_cos': Tensor(shape=(7,), dtype=float64),
            'robot0_joint_pos_sin': Tensor(shape=(7,), dtype=float64),
            'robot0_joint_vel': Tensor(shape=(7,), dtype=float64, description=7DOF joint velocities),
        }),
        'reward': float64,
        'states': Tensor(shape=(32,), dtype=float64),
    }),
})
  • फ़ीचर दस्तावेज़ीकरण :
विशेषता कक्षा आकार डीप्रकार विवरण
फीचर्सडिक्ट
एपिसोड_आईडी टेन्सर डोरी
क्षितिज टेन्सर int32
कदम डेटासेट
कदम/कार्रवाई टेन्सर (7,) फ्लोट64
कदम/छूट टेन्सर int32
चरण/पहला है टेन्सर बूल
चरण/अंतिम है टेन्सर बूल
चरण/is_terminal टेन्सर बूल
चरण/अवलोकन फीचर्सडिक्ट
चरण/अवलोकन/एजेंटव्यू_इमेज छवि (84, 84, 3) uint8
चरण/अवलोकन/वस्तु टेन्सर (10,) फ्लोट64
चरण/अवलोकन/robot0_eef_pos टेन्सर (3,) फ्लोट64 अंत-प्रभावक स्थिति
चरण/अवलोकन/robot0_eef_quat टेन्सर (4,) फ्लोट64 अंत-प्रभावक अभिविन्यास
चरण/अवलोकन/robot0_eef_vel_ang टेन्सर (3,) फ्लोट64 अंत-प्रभावक कोणीय वेग
चरण/अवलोकन/robot0_eef_vel_lin टेन्सर (3,) फ्लोट64 अंत-प्रभावक कार्तीय वेग
चरण/अवलोकन/robot0_eye_in_hand_image छवि (84, 84, 3) uint8
चरण/अवलोकन/robot0_gripper_qpos टेन्सर (2,) फ्लोट64 ग्रिपर की स्थिति
चरण/अवलोकन/robot0_gripper_qvel टेन्सर (2,) फ्लोट64 ग्रिपर वेग
चरण/अवलोकन/robot0_joint_pos टेन्सर (7,) फ्लोट64 7DOF संयुक्त पद
चरण/अवलोकन/robot0_joint_pos_cos टेन्सर (7,) फ्लोट64
चरण/अवलोकन/robot0_joint_pos_sin टेन्सर (7,) फ्लोट64
चरण/अवलोकन/robot0_joint_vel टेन्सर (7,) फ्लोट64 7DOF संयुक्त वेग
कदम/इनाम टेन्सर फ्लोट64
चरण/स्थितियाँ टेन्सर (32,) फ्लोट64

robomimic_mg/lift_mg_low_dim

  • डाउनलोड आकार : 302.25 MiB

  • डेटासेट का आकार : 195.10 MiB

  • ऑटो-कैश्ड ( दस्तावेज़ीकरण ): केवल तभी जब shuffle_files=False (ट्रेन)

  • विभाजन :

विभाजित करना उदाहरण
'train' 1,500
  • फ़ीचर संरचना :
FeaturesDict({
    'episode_id': string,
    'horizon': int32,
    'steps': Dataset({
        'action': Tensor(shape=(7,), dtype=float64),
        'discount': int32,
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'object': Tensor(shape=(10,), dtype=float64),
            'robot0_eef_pos': Tensor(shape=(3,), dtype=float64, description=End-effector position),
            'robot0_eef_quat': Tensor(shape=(4,), dtype=float64, description=End-effector orientation),
            'robot0_eef_vel_ang': Tensor(shape=(3,), dtype=float64, description=End-effector angular velocity),
            'robot0_eef_vel_lin': Tensor(shape=(3,), dtype=float64, description=End-effector cartesian velocity),
            'robot0_gripper_qpos': Tensor(shape=(2,), dtype=float64, description=Gripper position),
            'robot0_gripper_qvel': Tensor(shape=(2,), dtype=float64, description=Gripper velocity),
            'robot0_joint_pos': Tensor(shape=(7,), dtype=float64, description=7DOF joint positions),
            'robot0_joint_pos_cos': Tensor(shape=(7,), dtype=float64),
            'robot0_joint_pos_sin': Tensor(shape=(7,), dtype=float64),
            'robot0_joint_vel': Tensor(shape=(7,), dtype=float64, description=7DOF joint velocities),
        }),
        'reward': float64,
        'states': Tensor(shape=(32,), dtype=float64),
    }),
})
  • फ़ीचर दस्तावेज़ीकरण :
विशेषता कक्षा आकार डीप्रकार विवरण
फीचर्सडिक्ट
एपिसोड_आईडी टेन्सर डोरी
क्षितिज टेन्सर int32
कदम डेटासेट
कदम/कार्रवाई टेन्सर (7,) फ्लोट64
कदम/छूट टेन्सर int32
चरण/पहला है टेन्सर बूल
चरण/अंतिम है टेन्सर बूल
चरण/is_terminal टेन्सर बूल
चरण/अवलोकन फीचर्सडिक्ट
चरण/अवलोकन/वस्तु टेन्सर (10,) फ्लोट64
चरण/अवलोकन/robot0_eef_pos टेन्सर (3,) फ्लोट64 अंत-प्रभावक स्थिति
चरण/अवलोकन/robot0_eef_quat टेन्सर (4,) फ्लोट64 अंत-प्रभावक अभिविन्यास
चरण/अवलोकन/robot0_eef_vel_ang टेन्सर (3,) फ्लोट64 अंत-प्रभावक कोणीय वेग
चरण/अवलोकन/robot0_eef_vel_lin टेन्सर (3,) फ्लोट64 अंत-प्रभावक कार्तीय वेग
चरण/अवलोकन/robot0_gripper_qpos टेन्सर (2,) फ्लोट64 ग्रिपर की स्थिति
चरण/अवलोकन/robot0_gripper_qvel टेन्सर (2,) फ्लोट64 ग्रिपर वेग
चरण/अवलोकन/robot0_joint_pos टेन्सर (7,) फ्लोट64 7DOF संयुक्त पद
चरण/अवलोकन/robot0_joint_pos_cos टेन्सर (7,) फ्लोट64
चरण/अवलोकन/robot0_joint_pos_sin टेन्सर (7,) फ्लोट64
चरण/अवलोकन/robot0_joint_vel टेन्सर (7,) फ्लोट64 7DOF संयुक्त वेग
कदम/इनाम टेन्सर फ्लोट64
चरण/स्थितियाँ टेन्सर (32,) फ्लोट64

robomimic_mg/can_mg_image

  • डाउनलोड आकार : 47.14 GiB

  • डेटासेट का आकार : 11.15 GiB

  • ऑटो-कैश्ड ( दस्तावेज़ीकरण ): नहीं

  • विभाजन :

विभाजित करना उदाहरण
'train' 3,900
  • फ़ीचर संरचना :
FeaturesDict({
    'episode_id': string,
    'horizon': int32,
    'steps': Dataset({
        'action': Tensor(shape=(7,), dtype=float64),
        'discount': int32,
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'agentview_image': Image(shape=(84, 84, 3), dtype=uint8),
            'object': Tensor(shape=(14,), dtype=float64),
            'robot0_eef_pos': Tensor(shape=(3,), dtype=float64, description=End-effector position),
            'robot0_eef_quat': Tensor(shape=(4,), dtype=float64, description=End-effector orientation),
            'robot0_eef_vel_ang': Tensor(shape=(3,), dtype=float64, description=End-effector angular velocity),
            'robot0_eef_vel_lin': Tensor(shape=(3,), dtype=float64, description=End-effector cartesian velocity),
            'robot0_eye_in_hand_image': Image(shape=(84, 84, 3), dtype=uint8),
            'robot0_gripper_qpos': Tensor(shape=(2,), dtype=float64, description=Gripper position),
            'robot0_gripper_qvel': Tensor(shape=(2,), dtype=float64, description=Gripper velocity),
            'robot0_joint_pos': Tensor(shape=(7,), dtype=float64, description=7DOF joint positions),
            'robot0_joint_pos_cos': Tensor(shape=(7,), dtype=float64),
            'robot0_joint_pos_sin': Tensor(shape=(7,), dtype=float64),
            'robot0_joint_vel': Tensor(shape=(7,), dtype=float64, description=7DOF joint velocities),
        }),
        'reward': float64,
        'states': Tensor(shape=(71,), dtype=float64),
    }),
})
  • फ़ीचर दस्तावेज़ीकरण :
विशेषता कक्षा आकार डीप्रकार विवरण
फीचर्सडिक्ट
एपिसोड_आईडी टेन्सर डोरी
क्षितिज टेन्सर int32
कदम डेटासेट
कदम/कार्रवाई टेन्सर (7,) फ्लोट64
कदम/छूट टेन्सर int32
चरण/पहला है टेन्सर बूल
चरण/अंतिम है टेन्सर बूल
चरण/is_terminal टेन्सर बूल
चरण/अवलोकन फीचर्सडिक्ट
चरण/अवलोकन/एजेंटव्यू_इमेज छवि (84, 84, 3) uint8
चरण/अवलोकन/वस्तु टेन्सर (14,) फ्लोट64
चरण/अवलोकन/robot0_eef_pos टेन्सर (3,) फ्लोट64 अंत-प्रभावक स्थिति
चरण/अवलोकन/robot0_eef_quat टेन्सर (4,) फ्लोट64 अंत-प्रभावक अभिविन्यास
चरण/अवलोकन/robot0_eef_vel_ang टेन्सर (3,) फ्लोट64 अंत-प्रभावक कोणीय वेग
चरण/अवलोकन/robot0_eef_vel_lin टेन्सर (3,) फ्लोट64 अंत-प्रभावक कार्तीय वेग
चरण/अवलोकन/robot0_eye_in_hand_image छवि (84, 84, 3) uint8
चरण/अवलोकन/robot0_gripper_qpos टेन्सर (2,) फ्लोट64 ग्रिपर की स्थिति
चरण/अवलोकन/robot0_gripper_qvel टेन्सर (2,) फ्लोट64 ग्रिपर वेग
चरण/अवलोकन/robot0_joint_pos टेन्सर (7,) फ्लोट64 7DOF संयुक्त पद
चरण/अवलोकन/robot0_joint_pos_cos टेन्सर (7,) फ्लोट64
चरण/अवलोकन/robot0_joint_pos_sin टेन्सर (7,) फ्लोट64
चरण/अवलोकन/robot0_joint_vel टेन्सर (7,) फ्लोट64 7DOF संयुक्त वेग
कदम/इनाम टेन्सर फ्लोट64
चरण/स्थितियाँ टेन्सर (71,) फ्लोट64

robomimic_mg/can_mg_low_dim

  • डाउनलोड आकार : 1.01 GiB

  • डेटासेट का आकार : 697.71 MiB

  • ऑटो-कैश्ड ( दस्तावेज़ीकरण ): नहीं

  • विभाजन :

विभाजित करना उदाहरण
'train' 3,900
  • फ़ीचर संरचना :
FeaturesDict({
    'episode_id': string,
    'horizon': int32,
    'steps': Dataset({
        'action': Tensor(shape=(7,), dtype=float64),
        'discount': int32,
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'object': Tensor(shape=(14,), dtype=float64),
            'robot0_eef_pos': Tensor(shape=(3,), dtype=float64, description=End-effector position),
            'robot0_eef_quat': Tensor(shape=(4,), dtype=float64, description=End-effector orientation),
            'robot0_eef_vel_ang': Tensor(shape=(3,), dtype=float64, description=End-effector angular velocity),
            'robot0_eef_vel_lin': Tensor(shape=(3,), dtype=float64, description=End-effector cartesian velocity),
            'robot0_gripper_qpos': Tensor(shape=(2,), dtype=float64, description=Gripper position),
            'robot0_gripper_qvel': Tensor(shape=(2,), dtype=float64, description=Gripper velocity),
            'robot0_joint_pos': Tensor(shape=(7,), dtype=float64, description=7DOF joint positions),
            'robot0_joint_pos_cos': Tensor(shape=(7,), dtype=float64),
            'robot0_joint_pos_sin': Tensor(shape=(7,), dtype=float64),
            'robot0_joint_vel': Tensor(shape=(7,), dtype=float64, description=7DOF joint velocities),
        }),
        'reward': float64,
        'states': Tensor(shape=(71,), dtype=float64),
    }),
})
  • फ़ीचर दस्तावेज़ीकरण :
विशेषता कक्षा आकार डीप्रकार विवरण
फीचर्सडिक्ट
एपिसोड_आईडी टेन्सर डोरी
क्षितिज टेन्सर int32
कदम डेटासेट
कदम/कार्रवाई टेन्सर (7,) फ्लोट64
कदम/छूट टेन्सर int32
चरण/पहला है टेन्सर बूल
चरण/अंतिम है टेन्सर बूल
चरण/is_terminal टेन्सर बूल
चरण/अवलोकन फीचर्सडिक्ट
चरण/अवलोकन/वस्तु टेन्सर (14,) फ्लोट64
चरण/अवलोकन/robot0_eef_pos टेन्सर (3,) फ्लोट64 अंत-प्रभावक स्थिति
चरण/अवलोकन/robot0_eef_quat टेन्सर (4,) फ्लोट64 अंत-प्रभावक अभिविन्यास
चरण/अवलोकन/robot0_eef_vel_ang टेन्सर (3,) फ्लोट64 अंत-प्रभावक कोणीय वेग
चरण/अवलोकन/robot0_eef_vel_lin टेन्सर (3,) फ्लोट64 अंत-प्रभावक कार्तीय वेग
चरण/अवलोकन/robot0_gripper_qpos टेन्सर (2,) फ्लोट64 ग्रिपर की स्थिति
चरण/अवलोकन/robot0_gripper_qvel टेन्सर (2,) फ्लोट64 ग्रिपर वेग
चरण/अवलोकन/robot0_joint_pos टेन्सर (7,) फ्लोट64 7DOF संयुक्त पद
चरण/अवलोकन/robot0_joint_pos_cos टेन्सर (7,) फ्लोट64
चरण/अवलोकन/robot0_joint_pos_sin टेन्सर (7,) फ्लोट64
चरण/अवलोकन/robot0_joint_vel टेन्सर (7,) फ्लोट64 7DOF संयुक्त वेग
कदम/इनाम टेन्सर फ्लोट64
चरण/स्थितियाँ टेन्सर (71,) फ्लोट64