d4rl_adroit_door

  • คำอธิบาย :

D4RL เป็นเกณฑ์มาตรฐานแบบโอเพ่นซอร์สสำหรับการเรียนรู้แบบออฟไลน์ มีสภาพแวดล้อมและชุดข้อมูลที่เป็นมาตรฐานสำหรับการฝึกอบรมและอัลกอริธึมการเปรียบเทียบ

ชุดข้อมูลเป็นไปตาม รูปแบบ RLDS เพื่อแสดงขั้นตอนและตอนต่างๆ

@misc{fu2020d4rl,
    title={D4RL: Datasets for Deep Data-Driven Reinforcement Learning},
    author={Justin Fu and Aviral Kumar and Ofir Nachum and George Tucker and Sergey Levine},
    year={2020},
    eprint={2004.07219},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

d4rl_adroit_door/v0-human (การกำหนดค่าเริ่มต้น)

  • ขนาดการดาวน์โหลด : 2.97 MiB

  • ขนาดชุดข้อมูล : 3.36 MiB

  • แคชอัตโนมัติ ( เอกสาร ): ใช่

  • แยก :

แยก ตัวอย่าง
'train' 50
  • โครงสร้างคุณลักษณะ :
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(28,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'qpos': Tensor(shape=(30,), dtype=float32),
            'qvel': Tensor(shape=(30,), dtype=float32),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(39,), dtype=float32),
        'reward': float32,
    }),
})
  • เอกสารคุณสมบัติ :
ลักษณะเฉพาะ ระดับ รูปร่าง Dประเภท คำอธิบาย
คุณสมบัติDict
ขั้นตอน ชุดข้อมูล
ขั้นตอน/การดำเนินการ เทนเซอร์ (28,) ลอย32
ขั้นตอน/ส่วนลด เทนเซอร์ ลอย32
ขั้นตอน/ข้อมูล คุณสมบัติDict
ขั้นตอน/ข้อมูล/qpos เทนเซอร์ (30,) ลอย32
ขั้นตอน/ข้อมูล/qvel เทนเซอร์ (30,) ลอย32
ขั้นตอน/is_first เทนเซอร์ บูล
ขั้นตอน/is_last เทนเซอร์ บูล
ขั้นตอน/is_terminal เทนเซอร์ บูล
ขั้นตอน/ข้อสังเกต เทนเซอร์ (39,) ลอย32
ขั้นตอน / รางวัล เทนเซอร์ ลอย32

d4rl_adroit_door/v0-โคลน

  • ขนาดการดาวน์โหลด : 602.42 MiB

  • ขนาดชุดข้อมูล : 497.47 MiB

  • แคชอัตโนมัติ ( เอกสารประกอบ ): ไม่

  • แยก :

แยก ตัวอย่าง
'train' 6,214
  • โครงสร้างคุณลักษณะ :
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(28,), dtype=float32),
        'discount': float64,
        'infos': FeaturesDict({
            'qpos': Tensor(shape=(30,), dtype=float64),
            'qvel': Tensor(shape=(30,), dtype=float64),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(39,), dtype=float64),
        'reward': float64,
    }),
})
  • เอกสารคุณสมบัติ :
ลักษณะเฉพาะ ระดับ รูปร่าง Dประเภท คำอธิบาย
คุณสมบัติDict
ขั้นตอน ชุดข้อมูล
ขั้นตอน/การดำเนินการ เทนเซอร์ (28,) ลอย32
ขั้นตอน/ส่วนลด เทนเซอร์ float64
ขั้นตอน/ข้อมูล คุณสมบัติDict
ขั้นตอน/ข้อมูล/qpos เทนเซอร์ (30,) float64
ขั้นตอน/ข้อมูล/qvel เทนเซอร์ (30,) float64
ขั้นตอน/is_first เทนเซอร์ บูล
ขั้นตอน/is_last เทนเซอร์ บูล
ขั้นตอน/is_terminal เทนเซอร์ บูล
ขั้นตอน/ข้อสังเกต เทนเซอร์ (39,) float64
ขั้นตอน / รางวัล เทนเซอร์ float64

d4rl_adroit_door/v0-ผู้เชี่ยวชาญ

  • ขนาดการดาวน์โหลด : 511.05 MiB

  • ขนาดชุดข้อมูล : 710.30 MiB

  • แคชอัตโนมัติ ( เอกสารประกอบ ): ไม่

  • แยก :

แยก ตัวอย่าง
'train' 5,000
  • โครงสร้างคุณลักษณะ :
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(28,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_logstd': Tensor(shape=(28,), dtype=float32),
            'action_mean': Tensor(shape=(28,), dtype=float32),
            'qpos': Tensor(shape=(30,), dtype=float32),
            'qvel': Tensor(shape=(30,), dtype=float32),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(39,), dtype=float32),
        'reward': float32,
    }),
})
  • เอกสารคุณสมบัติ :
ลักษณะเฉพาะ ระดับ รูปร่าง Dประเภท คำอธิบาย
คุณสมบัติDict
ขั้นตอน ชุดข้อมูล
ขั้นตอน/การดำเนินการ เทนเซอร์ (28,) ลอย32
ขั้นตอน/ส่วนลด เทนเซอร์ ลอย32
ขั้นตอน/ข้อมูล คุณสมบัติDict
ขั้นตอน/infos/action_logstd เทนเซอร์ (28,) ลอย32
ขั้นตอน/ข้อมูล/action_mean เทนเซอร์ (28,) ลอย32
ขั้นตอน/ข้อมูล/qpos เทนเซอร์ (30,) ลอย32
ขั้นตอน/ข้อมูล/qvel เทนเซอร์ (30,) ลอย32
ขั้นตอน/is_first เทนเซอร์ บูล
ขั้นตอน/is_last เทนเซอร์ บูล
ขั้นตอน/is_terminal เทนเซอร์ บูล
ขั้นตอน/ข้อสังเกต เทนเซอร์ (39,) ลอย32
ขั้นตอน / รางวัล เทนเซอร์ ลอย32

d4rl_adroit_door/v1-มนุษย์

  • ขนาดการดาวน์โหลด : 2.98 MiB

  • ขนาดชุดข้อมูล : 3.42 MiB

  • แคชอัตโนมัติ ( เอกสาร ): ใช่

  • แยก :

แยก ตัวอย่าง
'train' 25
  • โครงสร้างคุณลักษณะ :
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(28,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'door_body_pos': Tensor(shape=(3,), dtype=float32),
            'qpos': Tensor(shape=(30,), dtype=float32),
            'qvel': Tensor(shape=(30,), dtype=float32),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(39,), dtype=float32),
        'reward': float32,
    }),
})
  • เอกสารคุณสมบัติ :
ลักษณะเฉพาะ ระดับ รูปร่าง Dประเภท คำอธิบาย
คุณสมบัติDict
ขั้นตอน ชุดข้อมูล
ขั้นตอน/การดำเนินการ เทนเซอร์ (28,) ลอย32
ขั้นตอน/ส่วนลด เทนเซอร์ ลอย32
ขั้นตอน/ข้อมูล คุณสมบัติDict
ขั้นตอน/ข้อมูล/door_body_pos เทนเซอร์ (3,) ลอย32
ขั้นตอน/ข้อมูล/qpos เทนเซอร์ (30,) ลอย32
ขั้นตอน/ข้อมูล/qvel เทนเซอร์ (30,) ลอย32
ขั้นตอน/is_first เทนเซอร์ บูล
ขั้นตอน/is_last เทนเซอร์ บูล
ขั้นตอน/is_terminal เทนเซอร์ บูล
ขั้นตอน/ข้อสังเกต เทนเซอร์ (39,) ลอย32
ขั้นตอน / รางวัล เทนเซอร์ ลอย32

d4rl_adroit_door/v1-โคลน

  • ขนาดการดาวน์โหลด : 280.72 MiB

  • ขนาดชุดข้อมูล : 1.85 GiB

  • แคชอัตโนมัติ ( เอกสารประกอบ ): ไม่

  • แยก :

แยก ตัวอย่าง
'train' 4,358
  • โครงสร้างคุณลักษณะ :
FeaturesDict({
    'algorithm': string,
    'policy': FeaturesDict({
        'fc0': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=float32),
            'weight': Tensor(shape=(39, 256), dtype=float32),
        }),
        'fc1': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=float32),
            'weight': Tensor(shape=(256, 256), dtype=float32),
        }),
        'last_fc': FeaturesDict({
            'bias': Tensor(shape=(28,), dtype=float32),
            'weight': Tensor(shape=(256, 28), dtype=float32),
        }),
        'nonlinearity': string,
        'output_distribution': string,
    }),
    'steps': Dataset({
        'action': Tensor(shape=(28,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'door_body_pos': Tensor(shape=(3,), dtype=float32),
            'qpos': Tensor(shape=(30,), dtype=float32),
            'qvel': Tensor(shape=(30,), dtype=float32),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(39,), dtype=float32),
        'reward': float32,
    }),
})
  • เอกสารคุณสมบัติ :
ลักษณะเฉพาะ ระดับ รูปร่าง Dประเภท คำอธิบาย
คุณสมบัติDict
อัลกอริทึม เทนเซอร์ สตริง
นโยบาย คุณสมบัติDict
นโยบาย/fc0 คุณสมบัติDict
นโยบาย/fc0/อคติ เทนเซอร์ (256,) ลอย32
นโยบาย/fc0/น้ำหนัก เทนเซอร์ (39, 256) ลอย32
นโยบาย/fc1 คุณสมบัติDict
นโยบาย/fc1/อคติ เทนเซอร์ (256,) ลอย32
นโยบาย/fc1/น้ำหนัก เทนเซอร์ (256, 256) ลอย32
นโยบาย/last_fc คุณสมบัติDict
นโยบาย/last_fc/อคติ เทนเซอร์ (28,) ลอย32
นโยบาย/last_fc/น้ำหนัก เทนเซอร์ (256, 28) ลอย32
นโยบาย/ความไม่เป็นเชิงเส้น เทนเซอร์ สตริง
นโยบาย/output_distribution เทนเซอร์ สตริง
ขั้นตอน ชุดข้อมูล
ขั้นตอน/การดำเนินการ เทนเซอร์ (28,) ลอย32
ขั้นตอน/ส่วนลด เทนเซอร์ ลอย32
ขั้นตอน/ข้อมูล คุณสมบัติDict
ขั้นตอน/ข้อมูล/door_body_pos เทนเซอร์ (3,) ลอย32
ขั้นตอน/ข้อมูล/qpos เทนเซอร์ (30,) ลอย32
ขั้นตอน/ข้อมูล/qvel เทนเซอร์ (30,) ลอย32
ขั้นตอน/is_first เทนเซอร์ บูล
ขั้นตอน/is_last เทนเซอร์ บูล
ขั้นตอน/is_terminal เทนเซอร์ บูล
ขั้นตอน/ข้อสังเกต เทนเซอร์ (39,) ลอย32
ขั้นตอน / รางวัล เทนเซอร์ ลอย32

d4rl_adroit_door/v1-ผู้เชี่ยวชาญ

  • ขนาดการดาวน์โหลด : 511.22 MiB

  • ขนาดชุดข้อมูล : 803.48 MiB

  • แคชอัตโนมัติ ( เอกสารประกอบ ): ไม่

  • แยก :

แยก ตัวอย่าง
'train' 5,000
  • โครงสร้างคุณลักษณะ :
FeaturesDict({
    'algorithm': string,
    'policy': FeaturesDict({
        'fc0': FeaturesDict({
            'bias': Tensor(shape=(32,), dtype=float32),
            'weight': Tensor(shape=(32, 39), dtype=float32),
        }),
        'fc1': FeaturesDict({
            'bias': Tensor(shape=(32,), dtype=float32),
            'weight': Tensor(shape=(32, 32), dtype=float32),
        }),
        'last_fc': FeaturesDict({
            'bias': Tensor(shape=(28,), dtype=float32),
            'weight': Tensor(shape=(28, 32), dtype=float32),
        }),
        'last_fc_log_std': FeaturesDict({
            'bias': Tensor(shape=(28,), dtype=float32),
            'weight': Tensor(shape=(28, 32), dtype=float32),
        }),
        'nonlinearity': string,
        'output_distribution': string,
    }),
    'steps': Dataset({
        'action': Tensor(shape=(28,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_log_std': Tensor(shape=(28,), dtype=float32),
            'action_mean': Tensor(shape=(28,), dtype=float32),
            'door_body_pos': Tensor(shape=(3,), dtype=float32),
            'qpos': Tensor(shape=(30,), dtype=float32),
            'qvel': Tensor(shape=(30,), dtype=float32),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(39,), dtype=float32),
        'reward': float32,
    }),
})
  • เอกสารคุณสมบัติ :
ลักษณะเฉพาะ ระดับ รูปร่าง Dประเภท คำอธิบาย
คุณสมบัติDict
อัลกอริทึม เทนเซอร์ สตริง
นโยบาย คุณสมบัติDict
นโยบาย/fc0 คุณสมบัติDict
นโยบาย/fc0/อคติ เทนเซอร์ (32,) ลอย32
นโยบาย/fc0/น้ำหนัก เทนเซอร์ (32, 39) ลอย32
นโยบาย/fc1 คุณสมบัติDict
นโยบาย/fc1/อคติ เทนเซอร์ (32,) ลอย32
นโยบาย/fc1/น้ำหนัก เทนเซอร์ (32, 32) ลอย32
นโยบาย/last_fc คุณสมบัติDict
นโยบาย/last_fc/อคติ เทนเซอร์ (28,) ลอย32
นโยบาย/last_fc/น้ำหนัก เทนเซอร์ (28, 32) ลอย32
นโยบาย/last_fc_log_std คุณสมบัติDict
นโยบาย/last_fc_log_std/อคติ เทนเซอร์ (28,) ลอย32
นโยบาย/last_fc_log_std/น้ำหนัก เทนเซอร์ (28, 32) ลอย32
นโยบาย/ความไม่เป็นเชิงเส้น เทนเซอร์ สตริง
นโยบาย/output_distribution เทนเซอร์ สตริง
ขั้นตอน ชุดข้อมูล
ขั้นตอน/การดำเนินการ เทนเซอร์ (28,) ลอย32
ขั้นตอน/ส่วนลด เทนเซอร์ ลอย32
ขั้นตอน/ข้อมูล คุณสมบัติDict
ขั้นตอน/ข้อมูล/action_log_std เทนเซอร์ (28,) ลอย32
ขั้นตอน/ข้อมูล/action_mean เทนเซอร์ (28,) ลอย32
ขั้นตอน/ข้อมูล/door_body_pos เทนเซอร์ (3,) ลอย32
ขั้นตอน/ข้อมูล/qpos เทนเซอร์ (30,) ลอย32
ขั้นตอน/ข้อมูล/qvel เทนเซอร์ (30,) ลอย32
ขั้นตอน/is_first เทนเซอร์ บูล
ขั้นตอน/is_last เทนเซอร์ บูล
ขั้นตอน/is_terminal เทนเซอร์ บูล
ขั้นตอน/ข้อสังเกต เทนเซอร์ (39,) ลอย32
ขั้นตอน / รางวัล เทนเซอร์ ลอย32