d4rl_mujoco_ant

  • Descrizione :

D4RL è un benchmark open source per l'apprendimento per rinforzo offline. Fornisce ambienti e set di dati standardizzati per algoritmi di addestramento e benchmarking.

I set di dati seguono il formato RLDS per rappresentare passaggi ed episodi.

@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_mujoco_ant/v0-expert (configurazione predefinita)

  • Dimensione del download : 131.34 MiB

  • Dimensione del set di dati: 464.94 MiB

  • Cache automatica ( documentazione ): No

  • Divisioni :

Diviso Esempi
'train' 1.288
  • Struttura delle caratteristiche :
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(8,), dtype=float32),
        'discount': float32,
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(111,), dtype=float32),
        'reward': float32,
    }),
})
  • Documentazione delle funzionalità :
Caratteristica Classe Forma Tipo D Descrizione
CaratteristicheDict
passi Set di dati
passi/azione Tensore (8,) galleggiante32
passi/sconto Tensore galleggiante32
passi/è_primo Tensore bool
passi/è_ultimo Tensore bool
passi/è_terminale Tensore bool
passi/osservazione Tensore (111,) galleggiante32
passi/ricompensa Tensore galleggiante32

d4rl_mujoco_ant/v0-medium

  • Dimensione del download : 131.39 MiB

  • Dimensione del set di dati: 464.78 MiB

  • Cache automatica ( documentazione ): No

  • Divisioni :

Diviso Esempi
'train' 1.122
  • Struttura delle caratteristiche :
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(8,), dtype=float32),
        'discount': float32,
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(111,), dtype=float32),
        'reward': float32,
    }),
})
  • Documentazione delle funzionalità :
Caratteristica Classe Forma Tipo D Descrizione
CaratteristicheDict
passi Set di dati
passi/azione Tensore (8,) galleggiante32
passi/sconto Tensore galleggiante32
passi/è_primo Tensore bool
passi/è_ultimo Tensore bool
passi/è_terminale Tensore bool
passi/osservazione Tensore (111,) galleggiante32
passi/ricompensa Tensore galleggiante32

d4rl_mujoco_ant/v0-medium-expert

  • Dimensione del download : 262.73 MiB

  • Dimensione del set di dati: 929.71 MiB

  • Cache automatica ( documentazione ): No

  • Divisioni :

Diviso Esempi
'train' 2.410
  • Struttura delle caratteristiche :
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(8,), dtype=float32),
        'discount': float32,
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(111,), dtype=float32),
        'reward': float32,
    }),
})
  • Documentazione delle funzionalità :
Caratteristica Classe Forma Tipo D Descrizione
CaratteristicheDict
passi Set di dati
passi/azione Tensore (8,) galleggiante32
passi/sconto Tensore galleggiante32
passi/è_primo Tensore bool
passi/è_ultimo Tensore bool
passi/è_terminale Tensore bool
passi/osservazione Tensore (111,) galleggiante32
passi/ricompensa Tensore galleggiante32

d4rl_mujoco_ant/v0-misto

  • Dimensione del download : 104.63 MiB

  • Dimensione del set di dati: 464.93 MiB

  • Cache automatica ( documentazione ): No

  • Divisioni :

Diviso Esempi
'train' 1.320
  • Struttura delle caratteristiche :
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(8,), dtype=float32),
        'discount': float32,
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(111,), dtype=float32),
        'reward': float32,
    }),
})
  • Documentazione delle funzionalità :
Caratteristica Classe Forma Tipo D Descrizione
CaratteristicheDict
passi Set di dati
passi/azione Tensore (8,) galleggiante32
passi/sconto Tensore galleggiante32
passi/è_primo Tensore bool
passi/è_ultimo Tensore bool
passi/è_terminale Tensore bool
passi/osservazione Tensore (111,) galleggiante32
passi/ricompensa Tensore galleggiante32

d4rl_mujoco_ant/v0-random

  • Dimensione del download : 139.50 MiB

  • Dimensione del set di dati: 464.97 MiB

  • Cache automatica ( documentazione ): No

  • Divisioni :

Diviso Esempi
'train' 1.377
  • Struttura delle caratteristiche :
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(8,), dtype=float32),
        'discount': float32,
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(111,), dtype=float32),
        'reward': float32,
    }),
})
  • Documentazione delle funzionalità :
Caratteristica Classe Forma Tipo D Descrizione
CaratteristicheDict
passi Set di dati
passi/azione Tensore (8,) galleggiante32
passi/sconto Tensore galleggiante32
passi/è_primo Tensore bool
passi/è_ultimo Tensore bool
passi/è_terminale Tensore bool
passi/osservazione Tensore (111,) galleggiante32
passi/ricompensa Tensore galleggiante32

d4rl_mujoco_ant/v1-expert

  • Dimensione del download : 220.72 MiB

  • Dimensione del set di dati: 968.63 MiB

  • Cache automatica ( documentazione ): No

  • Divisioni :

Diviso Esempi
'train' 1.033
  • Struttura delle caratteristiche :
FeaturesDict({
    'algorithm': string,
    'iteration': int32,
    'policy': FeaturesDict({
        'fc0': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=float32),
            'weight': Tensor(shape=(256, 111), dtype=float32),
        }),
        'fc1': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=float32),
            'weight': Tensor(shape=(256, 256), dtype=float32),
        }),
        'last_fc': FeaturesDict({
            'bias': Tensor(shape=(8,), dtype=float32),
            'weight': Tensor(shape=(8, 256), dtype=float32),
        }),
        'last_fc_log_std': FeaturesDict({
            'bias': Tensor(shape=(8,), dtype=float32),
            'weight': Tensor(shape=(8, 256), dtype=float32),
        }),
        'nonlinearity': string,
        'output_distribution': string,
    }),
    'steps': Dataset({
        'action': Tensor(shape=(8,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_log_probs': float32,
            'qpos': Tensor(shape=(15,), dtype=float32),
            'qvel': Tensor(shape=(14,), dtype=float32),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(111,), dtype=float32),
        'reward': float32,
    }),
})
  • Documentazione delle funzionalità :
Caratteristica Classe Forma Tipo D Descrizione
CaratteristicheDict
algoritmo Tensore corda
iterazione Tensore int32
politica CaratteristicheDict
politica/fc0 CaratteristicheDict
politica/fc0/bias Tensore (256,) galleggiante32
politica/fc0/peso Tensore (256, 111) galleggiante32
politica/fc1 CaratteristicheDict
politica/fc1/bias Tensore (256,) galleggiante32
politica/fc1/peso Tensore (256, 256) galleggiante32
politica/ultimo_fc CaratteristicheDict
policy/last_fc/bias Tensore (8,) galleggiante32
policy/last_fc/peso Tensore (8, 256) galleggiante32
policy/last_fc_log_std CaratteristicheDict
politica/last_fc_log_std/bias Tensore (8,) galleggiante32
politica/last_fc_log_std/peso Tensore (8, 256) galleggiante32
politica/non linearità Tensore corda
politica/output_distribution Tensore corda
passi Set di dati
passi/azione Tensore (8,) galleggiante32
passi/sconto Tensore galleggiante32
passaggi/informazioni CaratteristicheDict
steps/infos/action_log_probs Tensore galleggiante32
passi/informazioni/qpos Tensore (15,) galleggiante32
passaggi/informazioni/qvel Tensore (14,) galleggiante32
passi/è_primo Tensore bool
passi/è_ultimo Tensore bool
passi/è_terminale Tensore bool
passi/osservazione Tensore (111,) galleggiante32
passi/ricompensa Tensore galleggiante32

d4rl_mujoco_ant/v1-medium

  • Dimensione del download : 222.39 MiB

  • Dimensione del set di dati: 1023.71 MiB

  • Cache automatica ( documentazione ): No

  • Divisioni :

Diviso Esempi
'train' 1.179
  • Struttura delle caratteristiche :
FeaturesDict({
    'algorithm': string,
    'iteration': int32,
    'policy': FeaturesDict({
        'fc0': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=float32),
            'weight': Tensor(shape=(256, 111), dtype=float32),
        }),
        'fc1': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=float32),
            'weight': Tensor(shape=(256, 256), dtype=float32),
        }),
        'last_fc': FeaturesDict({
            'bias': Tensor(shape=(8,), dtype=float32),
            'weight': Tensor(shape=(8, 256), dtype=float32),
        }),
        'last_fc_log_std': FeaturesDict({
            'bias': Tensor(shape=(8,), dtype=float32),
            'weight': Tensor(shape=(8, 256), dtype=float32),
        }),
        'nonlinearity': string,
        'output_distribution': string,
    }),
    'steps': Dataset({
        'action': Tensor(shape=(8,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_log_probs': float32,
            'qpos': Tensor(shape=(15,), dtype=float32),
            'qvel': Tensor(shape=(14,), dtype=float32),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(111,), dtype=float32),
        'reward': float32,
    }),
})
  • Documentazione delle funzionalità :
Caratteristica Classe Forma Tipo D Descrizione
CaratteristicheDict
algoritmo Tensore corda
iterazione Tensore int32
politica CaratteristicheDict
politica/fc0 CaratteristicheDict
politica/fc0/bias Tensore (256,) galleggiante32
politica/fc0/peso Tensore (256, 111) galleggiante32
politica/fc1 CaratteristicheDict
politica/fc1/bias Tensore (256,) galleggiante32
politica/fc1/peso Tensore (256, 256) galleggiante32
politica/ultimo_fc CaratteristicheDict
policy/last_fc/bias Tensore (8,) galleggiante32
policy/last_fc/peso Tensore (8, 256) galleggiante32
policy/last_fc_log_std CaratteristicheDict
politica/last_fc_log_std/bias Tensore (8,) galleggiante32
politica/last_fc_log_std/peso Tensore (8, 256) galleggiante32
politica/non linearità Tensore corda
politica/output_distribution Tensore corda
passi Set di dati
passi/azione Tensore (8,) galleggiante32
passi/sconto Tensore galleggiante32
passaggi/informazioni CaratteristicheDict
steps/infos/action_log_probs Tensore galleggiante32
passi/informazioni/qpos Tensore (15,) galleggiante32
passaggi/informazioni/qvel Tensore (14,) galleggiante32
passi/è_primo Tensore bool
passi/è_ultimo Tensore bool
passi/è_terminale Tensore bool
passi/osservazione Tensore (111,) galleggiante32
passi/ricompensa Tensore galleggiante32

d4rl_mujoco_ant/v1-medium-expert

  • Dimensione del download : 442.25 MiB

  • Dimensione del set di dati : 1.13 GiB

  • Cache automatica ( documentazione ): No

  • Divisioni :

Diviso Esempi
'train' 2.211
  • Struttura delle caratteristiche :
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(8,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_log_probs': float32,
            'qpos': Tensor(shape=(15,), dtype=float32),
            'qvel': Tensor(shape=(14,), dtype=float32),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(111,), dtype=float32),
        'reward': float32,
    }),
})
  • Documentazione delle funzionalità :
Caratteristica Classe Forma Tipo D Descrizione
CaratteristicheDict
passi Set di dati
passi/azione Tensore (8,) galleggiante32
passi/sconto Tensore galleggiante32
passaggi/informazioni CaratteristicheDict
steps/infos/action_log_probs Tensore galleggiante32
passi/informazioni/qpos Tensore (15,) galleggiante32
passaggi/informazioni/qvel Tensore (14,) galleggiante32
passi/è_primo Tensore bool
passi/è_ultimo Tensore bool
passi/è_terminale Tensore bool
passi/osservazione Tensore (111,) galleggiante32
passi/ricompensa Tensore galleggiante32

d4rl_mujoco_ant/v1-medium-replay

  • Dimensione del download : 132.05 MiB

  • Dimensione del set di dati: 175.27 MiB

  • Cache automatica ( documentazione ): solo quando shuffle_files=False (treno)

  • Divisioni :

Diviso Esempi
'train' 485
  • Struttura delle caratteristiche :
FeaturesDict({
    'algorithm': string,
    'iteration': int32,
    'steps': Dataset({
        'action': Tensor(shape=(8,), dtype=float64),
        'discount': float64,
        'infos': FeaturesDict({
            'action_log_probs': float64,
            'qpos': Tensor(shape=(15,), dtype=float64),
            'qvel': Tensor(shape=(14,), dtype=float64),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(111,), dtype=float64),
        'reward': float64,
    }),
})
  • Documentazione delle funzionalità :
Caratteristica Classe Forma Tipo D Descrizione
CaratteristicheDict
algoritmo Tensore corda
iterazione Tensore int32
passi Set di dati
passi/azione Tensore (8,) galleggiante64
passi/sconto Tensore galleggiante64
passaggi/informazioni CaratteristicheDict
steps/infos/action_log_probs Tensore galleggiante64
passi/informazioni/qpos Tensore (15,) galleggiante64
passaggi/informazioni/qvel Tensore (14,) galleggiante64
passi/è_primo Tensore bool
passi/è_ultimo Tensore bool
passi/è_terminale Tensore bool
passi/osservazione Tensore (111,) galleggiante64
passi/ricompensa Tensore galleggiante64

d4rl_mujoco_ant/v1-full-replay

  • Dimensione del download : 437.57 MiB

  • Dimensione del set di dati: 580.09 MiB

  • Cache automatica ( documentazione ): No

  • Divisioni :

Diviso Esempi
'train' 1.319
  • Struttura delle caratteristiche :
FeaturesDict({
    'algorithm': string,
    'iteration': int32,
    'steps': Dataset({
        'action': Tensor(shape=(8,), dtype=float64),
        'discount': float64,
        'infos': FeaturesDict({
            'action_log_probs': float64,
            'qpos': Tensor(shape=(15,), dtype=float64),
            'qvel': Tensor(shape=(14,), dtype=float64),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(111,), dtype=float64),
        'reward': float64,
    }),
})
  • Documentazione delle funzionalità :
Caratteristica Classe Forma Tipo D Descrizione
CaratteristicheDict
algoritmo Tensore corda
iterazione Tensore int32
passi Set di dati
passi/azione Tensore (8,) galleggiante64
passi/sconto Tensore galleggiante64
passaggi/informazioni CaratteristicheDict
steps/infos/action_log_probs Tensore galleggiante64
passi/informazioni/qpos Tensore (15,) galleggiante64
passaggi/informazioni/qvel Tensore (14,) galleggiante64
passi/è_primo Tensore bool
passi/è_ultimo Tensore bool
passi/è_terminale Tensore bool
passi/osservazione Tensore (111,) galleggiante64
passi/ricompensa Tensore galleggiante64

d4rl_mujoco_ant/v1-random

  • Dimensione del download : 225.18 MiB

  • Dimensione del set di dati: 583.83 MiB

  • Cache automatica ( documentazione ): No

  • Divisioni :

Diviso Esempi
'train' 5.741
  • Struttura delle caratteristiche :
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(8,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_log_probs': float32,
            'qpos': Tensor(shape=(15,), dtype=float32),
            'qvel': Tensor(shape=(14,), dtype=float32),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(111,), dtype=float32),
        'reward': float32,
    }),
})
  • Documentazione delle funzionalità :
Caratteristica Classe Forma Tipo D Descrizione
CaratteristicheDict
passi Set di dati
passi/azione Tensore (8,) galleggiante32
passi/sconto Tensore galleggiante32
passaggi/informazioni CaratteristicheDict
steps/infos/action_log_probs Tensore galleggiante32
passi/informazioni/qpos Tensore (15,) galleggiante32
passaggi/informazioni/qvel Tensore (14,) galleggiante32
passi/è_primo Tensore bool
passi/è_ultimo Tensore bool
passi/è_terminale Tensore bool
passi/osservazione Tensore (111,) galleggiante32
passi/ricompensa Tensore galleggiante32

d4rl_mujoco_ant/v2-expert

  • Dimensione del download : 355.94 MiB

  • Dimensione del set di dati: 969.38 MiB

  • Cache automatica ( documentazione ): No

  • Divisioni :

Diviso Esempi
'train' 1.035
  • Struttura delle caratteristiche :
FeaturesDict({
    'algorithm': string,
    'iteration': int32,
    'policy': FeaturesDict({
        'fc0': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=float32),
            'weight': Tensor(shape=(256, 111), dtype=float32),
        }),
        'fc1': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=float32),
            'weight': Tensor(shape=(256, 256), dtype=float32),
        }),
        'last_fc': FeaturesDict({
            'bias': Tensor(shape=(8,), dtype=float32),
            'weight': Tensor(shape=(8, 256), dtype=float32),
        }),
        'last_fc_log_std': FeaturesDict({
            'bias': Tensor(shape=(8,), dtype=float32),
            'weight': Tensor(shape=(8, 256), dtype=float32),
        }),
        'nonlinearity': string,
        'output_distribution': string,
    }),
    'steps': Dataset({
        'action': Tensor(shape=(8,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_log_probs': float64,
            'qpos': Tensor(shape=(15,), dtype=float64),
            'qvel': Tensor(shape=(14,), dtype=float64),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(111,), dtype=float32),
        'reward': float32,
    }),
})
  • Documentazione delle funzionalità :
Caratteristica Classe Forma Tipo D Descrizione
CaratteristicheDict
algoritmo Tensore corda
iterazione Tensore int32
politica CaratteristicheDict
politica/fc0 CaratteristicheDict
politica/fc0/bias Tensore (256,) galleggiante32
politica/fc0/peso Tensore (256, 111) galleggiante32
politica/fc1 CaratteristicheDict
politica/fc1/bias Tensore (256,) galleggiante32
politica/fc1/peso Tensore (256, 256) galleggiante32
politica/ultimo_fc CaratteristicheDict
policy/last_fc/bias Tensore (8,) galleggiante32
policy/last_fc/peso Tensore (8, 256) galleggiante32
policy/last_fc_log_std CaratteristicheDict
politica/last_fc_log_std/bias Tensore (8,) galleggiante32
politica/last_fc_log_std/peso Tensore (8, 256) galleggiante32
politica/non linearità Tensore corda
politica/output_distribution Tensore corda
passi Set di dati
passi/azione Tensore (8,) galleggiante32
passi/sconto Tensore galleggiante32
passaggi/informazioni CaratteristicheDict
steps/infos/action_log_probs Tensore galleggiante64
passi/informazioni/qpos Tensore (15,) galleggiante64
passaggi/informazioni/qvel Tensore (14,) galleggiante64
passi/è_primo Tensore bool
passi/è_ultimo Tensore bool
passi/è_terminale Tensore bool
passi/osservazione Tensore (111,) galleggiante32
passi/ricompensa Tensore galleggiante32

d4rl_mujoco_ant/v2-full-replay

  • Dimensione del download : 428.57 MiB

  • Dimensione del set di dati: 580.09 MiB

  • Cache automatica ( documentazione ): No

  • Divisioni :

Diviso Esempi
'train' 1.319
  • Struttura delle caratteristiche :
FeaturesDict({
    'algorithm': string,
    'iteration': int32,
    'steps': Dataset({
        'action': Tensor(shape=(8,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_log_probs': float64,
            'qpos': Tensor(shape=(15,), dtype=float64),
            'qvel': Tensor(shape=(14,), dtype=float64),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(111,), dtype=float32),
        'reward': float32,
    }),
})
  • Documentazione delle funzionalità :
Caratteristica Classe Forma Tipo D Descrizione
CaratteristicheDict
algoritmo Tensore corda
iterazione Tensore int32
passi Set di dati
passi/azione Tensore (8,) galleggiante32
passi/sconto Tensore galleggiante32
passaggi/informazioni CaratteristicheDict
steps/infos/action_log_probs Tensore galleggiante64
passi/informazioni/qpos Tensore (15,) galleggiante64
passaggi/informazioni/qvel Tensore (14,) galleggiante64
passi/è_primo Tensore bool
passi/è_ultimo Tensore bool
passi/è_terminale Tensore bool
passi/osservazione Tensore (111,) galleggiante32
passi/ricompensa Tensore galleggiante32

d4rl_mujoco_ant/v2-medium

  • Dimensione del download : 358.81 MiB

  • Dimensione del set di dati : 1.01 GiB

  • Cache automatica ( documentazione ): No

  • Divisioni :

Diviso Esempi
'train' 1.203
  • Struttura delle caratteristiche :
FeaturesDict({
    'algorithm': string,
    'iteration': int32,
    'policy': FeaturesDict({
        'fc0': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=float32),
            'weight': Tensor(shape=(256, 111), dtype=float32),
        }),
        'fc1': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=float32),
            'weight': Tensor(shape=(256, 256), dtype=float32),
        }),
        'last_fc': FeaturesDict({
            'bias': Tensor(shape=(8,), dtype=float32),
            'weight': Tensor(shape=(8, 256), dtype=float32),
        }),
        'last_fc_log_std': FeaturesDict({
            'bias': Tensor(shape=(8,), dtype=float32),
            'weight': Tensor(shape=(8, 256), dtype=float32),
        }),
        'nonlinearity': string,
        'output_distribution': string,
    }),
    'steps': Dataset({
        'action': Tensor(shape=(8,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_log_probs': float64,
            'qpos': Tensor(shape=(15,), dtype=float64),
            'qvel': Tensor(shape=(14,), dtype=float64),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(111,), dtype=float32),
        'reward': float32,
    }),
})
  • Documentazione delle funzionalità :
Caratteristica Classe Forma Tipo D Descrizione
CaratteristicheDict
algoritmo Tensore corda
iterazione Tensore int32
politica CaratteristicheDict
politica/fc0 CaratteristicheDict
politica/fc0/bias Tensore (256,) galleggiante32
politica/fc0/peso Tensore (256, 111) galleggiante32
politica/fc1 CaratteristicheDict
politica/fc1/bias Tensore (256,) galleggiante32
politica/fc1/peso Tensore (256, 256) galleggiante32
politica/ultimo_fc CaratteristicheDict
policy/last_fc/bias Tensore (8,) galleggiante32
policy/last_fc/peso Tensore (8, 256) galleggiante32
policy/last_fc_log_std CaratteristicheDict
politica/last_fc_log_std/bias Tensore (8,) galleggiante32
politica/last_fc_log_std/peso Tensore (8, 256) galleggiante32
politica/non linearità Tensore corda
politica/output_distribution Tensore corda
passi Set di dati
passi/azione Tensore (8,) galleggiante32
passi/sconto Tensore galleggiante32
passaggi/informazioni CaratteristicheDict
steps/infos/action_log_probs Tensore galleggiante64
passi/informazioni/qpos Tensore (15,) galleggiante64
passaggi/informazioni/qvel Tensore (14,) galleggiante64
passi/è_primo Tensore bool
passi/è_ultimo Tensore bool
passi/è_terminale Tensore bool
passi/osservazione Tensore (111,) galleggiante32
passi/ricompensa Tensore galleggiante32

d4rl_mujoco_ant/v2-medium-expert

  • Dimensione del download : 713.67 MiB

  • Dimensione del set di dati : 1.13 GiB

  • Cache automatica ( documentazione ): No

  • Divisioni :

Diviso Esempi
'train' 2.237
  • Struttura delle caratteristiche :
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(8,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_log_probs': float64,
            'qpos': Tensor(shape=(15,), dtype=float64),
            'qvel': Tensor(shape=(14,), dtype=float64),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(111,), dtype=float32),
        'reward': float32,
    }),
})
  • Documentazione delle funzionalità :
Caratteristica Classe Forma Tipo D Descrizione
CaratteristicheDict
passi Set di dati
passi/azione Tensore (8,) galleggiante32
passi/sconto Tensore galleggiante32
passaggi/informazioni CaratteristicheDict
steps/infos/action_log_probs Tensore galleggiante64
passi/informazioni/qpos Tensore (15,) galleggiante64
passaggi/informazioni/qvel Tensore (14,) galleggiante64
passi/è_primo Tensore bool
passi/è_ultimo Tensore bool
passi/è_terminale Tensore bool
passi/osservazione Tensore (111,) galleggiante32
passi/ricompensa Tensore galleggiante32

d4rl_mujoco_ant/v2-medium-replay

  • Dimensione del download : 130.16 MiB

  • Dimensione del set di dati: 175.27 MiB

  • Cache automatica ( documentazione ): solo quando shuffle_files=False (treno)

  • Divisioni :

Diviso Esempi
'train' 485
  • Struttura delle caratteristiche :
FeaturesDict({
    'algorithm': string,
    'iteration': int32,
    'steps': Dataset({
        'action': Tensor(shape=(8,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_log_probs': float64,
            'qpos': Tensor(shape=(15,), dtype=float64),
            'qvel': Tensor(shape=(14,), dtype=float64),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(111,), dtype=float32),
        'reward': float32,
    }),
})
  • Documentazione delle funzionalità :
Caratteristica Classe Forma Tipo D Descrizione
CaratteristicheDict
algoritmo Tensore corda
iterazione Tensore int32
passi Set di dati
passi/azione Tensore (8,) galleggiante32
passi/sconto Tensore galleggiante32
passaggi/informazioni CaratteristicheDict
steps/infos/action_log_probs Tensore galleggiante64
passi/informazioni/qpos Tensore (15,) galleggiante64
passaggi/informazioni/qvel Tensore (14,) galleggiante64
passi/è_primo Tensore bool
passi/è_ultimo Tensore bool
passi/è_terminale Tensore bool
passi/osservazione Tensore (111,) galleggiante32
passi/ricompensa Tensore galleggiante32

d4rl_mujoco_ant/v2-random

  • Dimensione del download : 366.66 MiB

  • Dimensione del set di dati: 583.90 MiB

  • Cache automatica ( documentazione ): No

  • Divisioni :

Diviso Esempi
'train' 5.822
  • Struttura delle caratteristiche :
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(8,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_log_probs': float64,
            'qpos': Tensor(shape=(15,), dtype=float64),
            'qvel': Tensor(shape=(14,), dtype=float64),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(111,), dtype=float32),
        'reward': float32,
    }),
})
  • Documentazione delle funzionalità :
Caratteristica Classe Forma Tipo D Descrizione
CaratteristicheDict
passi Set di dati
passi/azione Tensore (8,) galleggiante32
passi/sconto Tensore galleggiante32
passaggi/informazioni CaratteristicheDict
steps/infos/action_log_probs Tensore galleggiante64
passi/informazioni/qpos Tensore (15,) galleggiante64
passaggi/informazioni/qvel Tensore (14,) galleggiante64
passi/è_primo Tensore bool
passi/è_ultimo Tensore bool
passi/è_terminale Tensore bool
passi/osservazione Tensore (111,) galleggiante32
passi/ricompensa Tensore galleggiante32