d4rl_adroit_pen

  • תיאור :

D4RL הוא אמת מידה בקוד פתוח ללמידת חיזוק לא מקוון. הוא מספק סביבות סטנדרטיות ומערכי נתונים עבור אלגוריתמי הדרכה ומידוד.

מערכי הנתונים פועלים לפי פורמט RLDS כדי לייצג שלבים ופרקים.

@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_pen/v0-human (תצורת ברירת המחדל)

לְפַצֵל דוגמאות
'train' 50
  • מבנה תכונה :
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(24,), 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=(45,), dtype=float32),
        'reward': float32,
    }),
})
  • תיעוד תכונה :
תכונה מעמד צוּרָה Dtype תיאור
FeaturesDict
צעדים מערך נתונים
צעדים/פעולה מוֹתֵחַ (24,) לצוף32
צעדים/הנחה מוֹתֵחַ לצוף32
שלבים/מידע FeaturesDict
צעדים/מידע/qpos מוֹתֵחַ (30,) לצוף32
צעדים/מידע/qvel מוֹתֵחַ (30,) לצוף32
צעדים/הוא_ראשון מוֹתֵחַ bool
צעדים/הוא_אחרון מוֹתֵחַ bool
steps/is_terminal מוֹתֵחַ bool
צעדים/תצפית מוֹתֵחַ (45,) לצוף32
צעדים/פרס מוֹתֵחַ לצוף32

d4rl_adroit_pen/v0-cloned

לְפַצֵל דוגמאות
'train' 5,023
  • מבנה תכונה :
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(24,), 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=(45,), dtype=float64),
        'reward': float64,
    }),
})
  • תיעוד תכונה :
תכונה מעמד צוּרָה Dtype תיאור
FeaturesDict
צעדים מערך נתונים
צעדים/פעולה מוֹתֵחַ (24,) לצוף32
צעדים/הנחה מוֹתֵחַ לצוף64
שלבים/מידע FeaturesDict
צעדים/מידע/qpos מוֹתֵחַ (30,) לצוף64
צעדים/מידע/qvel מוֹתֵחַ (30,) לצוף64
צעדים/הוא_ראשון מוֹתֵחַ bool
צעדים/הוא_אחרון מוֹתֵחַ bool
steps/is_terminal מוֹתֵחַ bool
צעדים/תצפית מוֹתֵחַ (45,) לצוף64
צעדים/פרס מוֹתֵחַ לצוף64

d4rl_adroit_pen/v0-expert

לְפַצֵל דוגמאות
'train' 5,000
  • מבנה תכונה :
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(24,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_logstd': Tensor(shape=(24,), dtype=float32),
            'action_mean': Tensor(shape=(24,), 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=(45,), dtype=float32),
        'reward': float32,
    }),
})
  • תיעוד תכונה :
תכונה מעמד צוּרָה Dtype תיאור
FeaturesDict
צעדים מערך נתונים
צעדים/פעולה מוֹתֵחַ (24,) לצוף32
צעדים/הנחה מוֹתֵחַ לצוף32
שלבים/מידע FeaturesDict
steps/infos/action_logstd מוֹתֵחַ (24,) לצוף32
steps/infos/action_mean מוֹתֵחַ (24,) לצוף32
צעדים/מידע/qpos מוֹתֵחַ (30,) לצוף32
צעדים/מידע/qvel מוֹתֵחַ (30,) לצוף32
צעדים/הוא_ראשון מוֹתֵחַ bool
צעדים/הוא_אחרון מוֹתֵחַ bool
steps/is_terminal מוֹתֵחַ bool
צעדים/תצפית מוֹתֵחַ (45,) לצוף32
צעדים/פרס מוֹתֵחַ לצוף32

d4rl_adroit_pen/v1-human

לְפַצֵל דוגמאות
'train' 25
  • מבנה תכונה :
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(24,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'desired_orien': Tensor(shape=(4,), 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=(45,), dtype=float32),
        'reward': float32,
    }),
})
  • תיעוד תכונה :
תכונה מעמד צוּרָה Dtype תיאור
FeaturesDict
צעדים מערך נתונים
צעדים/פעולה מוֹתֵחַ (24,) לצוף32
צעדים/הנחה מוֹתֵחַ לצוף32
שלבים/מידע FeaturesDict
steps/infos/desired_orien מוֹתֵחַ (4,) לצוף32
צעדים/מידע/qpos מוֹתֵחַ (30,) לצוף32
צעדים/מידע/qvel מוֹתֵחַ (30,) לצוף32
צעדים/הוא_ראשון מוֹתֵחַ bool
צעדים/הוא_אחרון מוֹתֵחַ bool
steps/is_terminal מוֹתֵחַ bool
צעדים/תצפית מוֹתֵחַ (45,) לצוף32
צעדים/פרס מוֹתֵחַ לצוף32

d4rl_adroit_pen/v1-cloned

לְפַצֵל דוגמאות
'train' 3,755
  • מבנה תכונה :
FeaturesDict({
    'algorithm': string,
    'policy': FeaturesDict({
        'fc0': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=float32),
            'weight': Tensor(shape=(45, 256), dtype=float32),
        }),
        'fc1': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=float32),
            'weight': Tensor(shape=(256, 256), dtype=float32),
        }),
        'last_fc': FeaturesDict({
            'bias': Tensor(shape=(24,), dtype=float32),
            'weight': Tensor(shape=(256, 24), dtype=float32),
        }),
        'nonlinearity': string,
        'output_distribution': string,
    }),
    'steps': Dataset({
        'action': Tensor(shape=(24,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'desired_orien': Tensor(shape=(4,), 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=(45,), dtype=float32),
        'reward': float32,
    }),
})
  • תיעוד תכונה :
תכונה מעמד צוּרָה Dtype תיאור
FeaturesDict
אַלגוֹרִיתְם מוֹתֵחַ חוּט
מְדִינִיוּת FeaturesDict
מדיניות/fc0 FeaturesDict
policy/fc0/bias מוֹתֵחַ (256,) לצוף32
policy/fc0/weight מוֹתֵחַ (45, 256) לצוף32
מדיניות/fc1 FeaturesDict
policy/fc1/bias מוֹתֵחַ (256,) לצוף32
policy/fc1/weight מוֹתֵחַ (256, 256) לצוף32
policy/last_fc FeaturesDict
policy/last_fc/bias מוֹתֵחַ (24,) לצוף32
policy/last_fc/weight מוֹתֵחַ (256, 24) לצוף32
מדיניות/אי-לינאריות מוֹתֵחַ חוּט
מדיניות/הפצה_תפוקה מוֹתֵחַ חוּט
צעדים מערך נתונים
צעדים/פעולה מוֹתֵחַ (24,) לצוף32
צעדים/הנחה מוֹתֵחַ לצוף32
שלבים/מידע FeaturesDict
steps/infos/desired_orien מוֹתֵחַ (4,) לצוף32
צעדים/מידע/qpos מוֹתֵחַ (30,) לצוף32
צעדים/מידע/qvel מוֹתֵחַ (30,) לצוף32
צעדים/הוא_ראשון מוֹתֵחַ bool
צעדים/הוא_אחרון מוֹתֵחַ bool
steps/is_terminal מוֹתֵחַ bool
צעדים/תצפית מוֹתֵחַ (45,) לצוף32
צעדים/פרס מוֹתֵחַ לצוף32

d4rl_adroit_pen/v1-expert

  • גודל הורדה : 249.90 MiB

  • גודל מערך נתונים : 548.47 MiB

  • שמירה אוטומטית במטמון ( תיעוד ): לא

  • פיצולים :

לְפַצֵל דוגמאות
'train' 5,000
  • מבנה תכונה :
FeaturesDict({
    'algorithm': string,
    'policy': FeaturesDict({
        'fc0': FeaturesDict({
            'bias': Tensor(shape=(64,), dtype=float32),
            'weight': Tensor(shape=(64, 45), dtype=float32),
        }),
        'fc1': FeaturesDict({
            'bias': Tensor(shape=(64,), dtype=float32),
            'weight': Tensor(shape=(64, 64), dtype=float32),
        }),
        'last_fc': FeaturesDict({
            'bias': Tensor(shape=(24,), dtype=float32),
            'weight': Tensor(shape=(24, 64), dtype=float32),
        }),
        'last_fc_log_std': FeaturesDict({
            'bias': Tensor(shape=(24,), dtype=float32),
            'weight': Tensor(shape=(24, 64), dtype=float32),
        }),
        'nonlinearity': string,
        'output_distribution': string,
    }),
    'steps': Dataset({
        'action': Tensor(shape=(24,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_log_std': Tensor(shape=(24,), dtype=float32),
            'action_mean': Tensor(shape=(24,), dtype=float32),
            'desired_orien': Tensor(shape=(4,), 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=(45,), dtype=float32),
        'reward': float32,
    }),
})
  • תיעוד תכונה :
תכונה מעמד צוּרָה Dtype תיאור
FeaturesDict
אַלגוֹרִיתְם מוֹתֵחַ חוּט
מְדִינִיוּת FeaturesDict
מדיניות/fc0 FeaturesDict
policy/fc0/bias מוֹתֵחַ (64,) לצוף32
policy/fc0/weight מוֹתֵחַ (64, 45) לצוף32
מדיניות/fc1 FeaturesDict
policy/fc1/bias מוֹתֵחַ (64,) לצוף32
policy/fc1/weight מוֹתֵחַ (64, 64) לצוף32
policy/last_fc FeaturesDict
policy/last_fc/bias מוֹתֵחַ (24,) לצוף32
policy/last_fc/weight מוֹתֵחַ (24, 64) לצוף32
policy/last_fc_log_std FeaturesDict
policy/last_fc_log_std/bias מוֹתֵחַ (24,) לצוף32
policy/last_fc_log_std/weight מוֹתֵחַ (24, 64) לצוף32
מדיניות/אי-לינאריות מוֹתֵחַ חוּט
מדיניות/הפצה_תפוקה מוֹתֵחַ חוּט
צעדים מערך נתונים
צעדים/פעולה מוֹתֵחַ (24,) לצוף32
צעדים/הנחה מוֹתֵחַ לצוף32
שלבים/מידע FeaturesDict
steps/infos/action_log_std מוֹתֵחַ (24,) לצוף32
steps/infos/action_mean מוֹתֵחַ (24,) לצוף32
steps/infos/desired_orien מוֹתֵחַ (4,) לצוף32
צעדים/מידע/qpos מוֹתֵחַ (30,) לצוף32
צעדים/מידע/qvel מוֹתֵחַ (30,) לצוף32
צעדים/הוא_ראשון מוֹתֵחַ bool
צעדים/הוא_אחרון מוֹתֵחַ bool
steps/is_terminal מוֹתֵחַ bool
צעדים/תצפית מוֹתֵחַ (45,) לצוף32
צעדים/פרס מוֹתֵחַ לצוף32