d4rl_mujoco_hopper

  • 説明:

D4RL は、オフライン強化学習のオープンソース ベンチマークです。トレーニングおよびベンチマーク アルゴリズム用の標準化された環境とデータセットを提供します。

データセットはRLDS 形式に従って、ステップとエピソードを表します。

@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_hopper/v0-expert (デフォルト設定)

  • ダウンロードサイズ: 51.56 MiB

  • データセットサイズ: 64.10 MiB

  • 自動キャッシュ(ドキュメント): はい

  • スプリット:

スプリット
'train' 1,029人
  • 機能構造:
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=float32),
        'discount': float32,
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(11,), dtype=float32),
        'reward': float32,
    }),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
ステップデータセット
ステップ/アクションテンソル(3) float32
ステップ/割引テンソルfloat32
ステップ/is_firstテンソルブール
ステップ/is_lastテンソルブール
ステップ/is_terminalテンソルブール
ステップ/観察テンソル(11) float32
ステップ/報酬テンソルfloat32

d4rl_mujoco_hopper/v0-中

  • ダウンロードサイズ: 51.74 MiB

  • データセットのサイズ: 64.68 MiB

  • 自動キャッシュ(ドキュメント): はい

  • スプリット:

スプリット
'train' 3,064
  • 機能構造:
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=float32),
        'discount': float32,
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(11,), dtype=float32),
        'reward': float32,
    }),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
ステップデータセット
ステップ/アクションテンソル(3) float32
ステップ/割引テンソルfloat32
ステップ/is_firstテンソルブール
ステップ/is_lastテンソルブール
ステップ/is_terminalテンソルブール
ステップ/観察テンソル(11) float32
ステップ/報酬テンソルfloat32

d4rl_mujoco_hopper/v0-medium-expert

  • ダウンロードサイズ: 62.01 MiB

  • データセットのサイズ: 77.25 MiB

  • 自動キャッシュ(ドキュメント): はい

  • スプリット:

スプリット
'train' 2,277
  • 機能構造:
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=float32),
        'discount': float32,
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(11,), dtype=float32),
        'reward': float32,
    }),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
ステップデータセット
ステップ/アクションテンソル(3) float32
ステップ/割引テンソルfloat32
ステップ/is_firstテンソルブール
ステップ/is_lastテンソルブール
ステップ/is_terminalテンソルブール
ステップ/観察テンソル(11) float32
ステップ/報酬テンソルfloat32

d4rl_mujoco_hopper/v0-mixed

  • ダウンロードサイズ: 10.48 MiB

  • データセットのサイズ: 13.15 MiB

  • 自動キャッシュ(ドキュメント): はい

  • スプリット:

スプリット
'train' 1,250
  • 機能構造:
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=float32),
        'discount': float32,
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(11,), dtype=float32),
        'reward': float32,
    }),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
ステップデータセット
ステップ/アクションテンソル(3) float32
ステップ/割引テンソルfloat32
ステップ/is_firstテンソルブール
ステップ/is_lastテンソルブール
ステップ/is_terminalテンソルブール
ステップ/観察テンソル(11) float32
ステップ/報酬テンソルfloat32

d4rl_mujoco_hopper/v0-ランダム

  • ダウンロードサイズ: 51.83 MiB

  • データセットサイズ: 66.06 MiB

  • 自動キャッシュ(ドキュメント): はい

  • スプリット:

スプリット
'train' 8,793
  • 機能構造:
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=float32),
        'discount': float32,
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(11,), dtype=float32),
        'reward': float32,
    }),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
ステップデータセット
ステップ/アクションテンソル(3) float32
ステップ/割引テンソルfloat32
ステップ/is_firstテンソルブール
ステップ/is_lastテンソルブール
ステップ/is_terminalテンソルブール
ステップ/観察テンソル(11) float32
ステップ/報酬テンソルfloat32

d4rl_mujoco_hopper/v1-expert

  • ダウンロードサイズ: 93.19 MiB

  • データセットサイズ: 608.03 MiB

  • 自動キャッシュ(ドキュメント): いいえ

  • スプリット:

スプリット
'train' 1,836
  • 機能構造:
FeaturesDict({
    'algorithm': string,
    'iteration': int32,
    'policy': FeaturesDict({
        'fc0': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=float32),
            'weight': Tensor(shape=(256, 11), dtype=float32),
        }),
        'fc1': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=float32),
            'weight': Tensor(shape=(256, 256), dtype=float32),
        }),
        'last_fc': FeaturesDict({
            'bias': Tensor(shape=(3,), dtype=float32),
            'weight': Tensor(shape=(3, 256), dtype=float32),
        }),
        'last_fc_log_std': FeaturesDict({
            'bias': Tensor(shape=(3,), dtype=float32),
            'weight': Tensor(shape=(3, 256), dtype=float32),
        }),
        'nonlinearity': string,
        'output_distribution': string,
    }),
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_log_probs': float32,
            'qpos': Tensor(shape=(6,), dtype=float32),
            'qvel': Tensor(shape=(6,), dtype=float32),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(11,), dtype=float32),
        'reward': float32,
    }),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
アルゴリズムテンソルストリング
反復テンソルint32
ポリシー特徴辞書
ポリシー/fc0特徴辞書
ポリシー/fc0/バイアステンソル(256,) float32
ポリシー/fc0/重みテンソル(256,11) float32
ポリシー/fc1特徴辞書
ポリシー/fc1/バイアステンソル(256,) float32
ポリシー/fc1/重量テンソル(256, 256) float32
ポリシー/last_fc特徴辞書
ポリシー/last_fc/バイアステンソル(3) float32
ポリシー/last_fc/重量テンソル(3,256) float32
ポリシー/last_fc_log_std特徴辞書
ポリシー/last_fc_log_std/バイアステンソル(3) float32
ポリシー/last_fc_log_std/重量テンソル(3,256) float32
ポリシー/非線形性テンソルストリング
ポリシー/出力配布テンソルストリング
ステップデータセット
ステップ/アクションテンソル(3) float32
ステップ/割引テンソルfloat32
手順/情報特徴辞書
ステップ/情報/action_log_probsテンソルfloat32
ステップ/情報/qposテンソル(6,) float32
ステップ/情報/qvelテンソル(6,) float32
ステップ/is_firstテンソルブール
ステップ/is_lastテンソルブール
ステップ/is_terminalテンソルブール
ステップ/観察テンソル(11) float32
ステップ/報酬テンソルfloat32

d4rl_mujoco_hopper/v1-中

  • ダウンロードサイズ: 92.03 MiB

  • データセットサイズ: 1.78 GiB

  • 自動キャッシュ(ドキュメント): いいえ

  • スプリット:

スプリット
'train' 6,328
  • 機能構造:
FeaturesDict({
    'algorithm': string,
    'iteration': int32,
    'policy': FeaturesDict({
        'fc0': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=float32),
            'weight': Tensor(shape=(256, 11), dtype=float32),
        }),
        'fc1': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=float32),
            'weight': Tensor(shape=(256, 256), dtype=float32),
        }),
        'last_fc': FeaturesDict({
            'bias': Tensor(shape=(3,), dtype=float32),
            'weight': Tensor(shape=(3, 256), dtype=float32),
        }),
        'last_fc_log_std': FeaturesDict({
            'bias': Tensor(shape=(3,), dtype=float32),
            'weight': Tensor(shape=(3, 256), dtype=float32),
        }),
        'nonlinearity': string,
        'output_distribution': string,
    }),
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_log_probs': float32,
            'qpos': Tensor(shape=(6,), dtype=float32),
            'qvel': Tensor(shape=(6,), dtype=float32),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(11,), dtype=float32),
        'reward': float32,
    }),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
アルゴリズムテンソルストリング
反復テンソルint32
ポリシー特徴辞書
ポリシー/fc0特徴辞書
ポリシー/fc0/バイアステンソル(256,) float32
ポリシー/fc0/重みテンソル(256,11) float32
ポリシー/fc1特徴辞書
ポリシー/fc1/バイアステンソル(256,) float32
ポリシー/fc1/重量テンソル(256, 256) float32
ポリシー/last_fc特徴辞書
ポリシー/last_fc/バイアステンソル(3) float32
ポリシー/last_fc/重量テンソル(3,256) float32
ポリシー/last_fc_log_std特徴辞書
ポリシー/last_fc_log_std/バイアステンソル(3) float32
ポリシー/last_fc_log_std/重量テンソル(3,256) float32
ポリシー/非線形性テンソルストリング
ポリシー/出力配布テンソルストリング
ステップデータセット
ステップ/アクションテンソル(3) float32
ステップ/割引テンソルfloat32
手順/情報特徴辞書
ステップ/情報/action_log_probsテンソルfloat32
ステップ/情報/qposテンソル(6,) float32
ステップ/情報/qvelテンソル(6,) float32
ステップ/is_firstテンソルブール
ステップ/is_lastテンソルブール
ステップ/is_terminalテンソルブール
ステップ/観察テンソル(11) float32
ステップ/報酬テンソルfloat32

d4rl_mujoco_hopper/v1-medium-expert

  • ダウンロードサイズ: 184.59 MiB

  • データセットサイズ: 230.24 MiB

  • 自動キャッシュ(ドキュメント): shuffle_files=Falseの場合のみ (トレーニング)

  • スプリット:

スプリット
'train' 8,163
  • 機能構造:
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_log_probs': float32,
            'qpos': Tensor(shape=(6,), dtype=float32),
            'qvel': Tensor(shape=(6,), dtype=float32),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(11,), dtype=float32),
        'reward': float32,
    }),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
ステップデータセット
ステップ/アクションテンソル(3) float32
ステップ/割引テンソルfloat32
手順/情報特徴辞書
ステップ/情報/action_log_probsテンソルfloat32
ステップ/情報/qposテンソル(6,) float32
ステップ/情報/qvelテンソル(6,) float32
ステップ/is_firstテンソルブール
ステップ/is_lastテンソルブール
ステップ/is_terminalテンソルブール
ステップ/観察テンソル(11) float32
ステップ/報酬テンソルfloat32

d4rl_mujoco_hopper/v1-medium-replay

  • ダウンロードサイズ: 55.65 MiB

  • データセットのサイズ: 34.78 MiB

  • 自動キャッシュ(ドキュメント): はい

  • スプリット:

スプリット
'train' 1,151
  • 機能構造:
FeaturesDict({
    'algorithm': string,
    'iteration': int32,
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=float64),
        'discount': float64,
        'infos': FeaturesDict({
            'action_log_probs': float64,
            'qpos': Tensor(shape=(6,), dtype=float64),
            'qvel': Tensor(shape=(6,), dtype=float64),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(11,), dtype=float64),
        'reward': float64,
    }),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
アルゴリズムテンソルストリング
反復テンソルint32
ステップデータセット
ステップ/アクションテンソル(3) float64
ステップ/割引テンソルfloat64
手順/情報特徴辞書
ステップ/情報/action_log_probsテンソルfloat64
ステップ/情報/qposテンソル(6,) float64
ステップ/情報/qvelテンソル(6,) float64
ステップ/is_firstテンソルブール
ステップ/is_lastテンソルブール
ステップ/is_terminalテンソルブール
ステップ/観察テンソル(11) float64
ステップ/報酬テンソルfloat64

d4rl_mujoco_hopper/v1-full-replay

  • ダウンロードサイズ: 183.32 MiB

  • データセットのサイズ: 114.78 MiB

  • 自動キャッシュ(ドキュメント): はい

  • スプリット:

スプリット
'train' 2,907
  • 機能構造:
FeaturesDict({
    'algorithm': string,
    'iteration': int32,
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=float64),
        'discount': float64,
        'infos': FeaturesDict({
            'action_log_probs': float64,
            'qpos': Tensor(shape=(6,), dtype=float64),
            'qvel': Tensor(shape=(6,), dtype=float64),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(11,), dtype=float64),
        'reward': float64,
    }),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
アルゴリズムテンソルストリング
反復テンソルint32
ステップデータセット
ステップ/アクションテンソル(3) float64
ステップ/割引テンソルfloat64
手順/情報特徴辞書
ステップ/情報/action_log_probsテンソルfloat64
ステップ/情報/qposテンソル(6,) float64
ステップ/情報/qvelテンソル(6,) float64
ステップ/is_firstテンソルブール
ステップ/is_lastテンソルブール
ステップ/is_terminalテンソルブール
ステップ/観察テンソル(11) float64
ステップ/報酬テンソルfloat64

d4rl_mujoco_hopper/v1-ランダ​​ム

  • ダウンロードサイズ: 91.11 MiB

  • データセットのサイズ: 130.73 MiB

  • 自動キャッシュ(ドキュメント): shuffle_files=Falseの場合のみ (トレーニング)

  • スプリット:

スプリット
'train' 45,265
  • 機能構造:
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_log_probs': float32,
            'qpos': Tensor(shape=(6,), dtype=float32),
            'qvel': Tensor(shape=(6,), dtype=float32),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(11,), dtype=float32),
        'reward': float32,
    }),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
ステップデータセット
ステップ/アクションテンソル(3) float32
ステップ/割引テンソルfloat32
手順/情報特徴辞書
ステップ/情報/action_log_probsテンソルfloat32
ステップ/情報/qposテンソル(6,) float32
ステップ/情報/qvelテンソル(6,) float32
ステップ/is_firstテンソルブール
ステップ/is_lastテンソルブール
ステップ/is_terminalテンソルブール
ステップ/観察テンソル(11) float32
ステップ/報酬テンソルfloat32

d4rl_mujoco_hopper/v2-expert

  • ダウンロードサイズ: 145.37 MiB

  • データセットのサイズ: 390.40 MiB

  • 自動キャッシュ(ドキュメント): いいえ

  • スプリット:

スプリット
'train' 1,028人
  • 機能構造:
FeaturesDict({
    'algorithm': string,
    'iteration': int32,
    'policy': FeaturesDict({
        'fc0': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=float32),
            'weight': Tensor(shape=(256, 11), dtype=float32),
        }),
        'fc1': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=float32),
            'weight': Tensor(shape=(256, 256), dtype=float32),
        }),
        'last_fc': FeaturesDict({
            'bias': Tensor(shape=(3,), dtype=float32),
            'weight': Tensor(shape=(3, 256), dtype=float32),
        }),
        'last_fc_log_std': FeaturesDict({
            'bias': Tensor(shape=(3,), dtype=float32),
            'weight': Tensor(shape=(3, 256), dtype=float32),
        }),
        'nonlinearity': string,
        'output_distribution': string,
    }),
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_log_probs': float64,
            'qpos': Tensor(shape=(6,), dtype=float64),
            'qvel': Tensor(shape=(6,), dtype=float64),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(11,), dtype=float32),
        'reward': float32,
    }),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
アルゴリズムテンソルストリング
反復テンソルint32
ポリシー特徴辞書
ポリシー/fc0特徴辞書
ポリシー/fc0/バイアステンソル(256,) float32
ポリシー/fc0/重みテンソル(256,11) float32
ポリシー/fc1特徴辞書
ポリシー/fc1/バイアステンソル(256,) float32
ポリシー/fc1/重量テンソル(256, 256) float32
ポリシー/last_fc特徴辞書
ポリシー/last_fc/バイアステンソル(3) float32
ポリシー/last_fc/重量テンソル(3,256) float32
ポリシー/last_fc_log_std特徴辞書
ポリシー/last_fc_log_std/バイアステンソル(3) float32
ポリシー/last_fc_log_std/重量テンソル(3,256) float32
ポリシー/非線形性テンソルストリング
ポリシー/出力配布テンソルストリング
ステップデータセット
ステップ/アクションテンソル(3) float32
ステップ/割引テンソルfloat32
手順/情報特徴辞書
ステップ/情報/action_log_probsテンソルfloat64
ステップ/情報/qposテンソル(6,) float64
ステップ/情報/qvelテンソル(6,) float64
ステップ/is_firstテンソルブール
ステップ/is_lastテンソルブール
ステップ/is_terminalテンソルブール
ステップ/観察テンソル(11) float32
ステップ/報酬テンソルfloat32

d4rl_mujoco_hopper/v2-full-replay

  • ダウンロードサイズ: 179.29 MiB

  • データセットのサイズ: 115.04 MiB

  • 自動キャッシュ(ドキュメント): はい

  • スプリット:

スプリット
'train' 3,515
  • 機能構造:
FeaturesDict({
    'algorithm': string,
    'iteration': int32,
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_log_probs': float64,
            'qpos': Tensor(shape=(6,), dtype=float64),
            'qvel': Tensor(shape=(6,), dtype=float64),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(11,), dtype=float32),
        'reward': float32,
    }),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
アルゴリズムテンソルストリング
反復テンソルint32
ステップデータセット
ステップ/アクションテンソル(3) float32
ステップ/割引テンソルfloat32
手順/情報特徴辞書
ステップ/情報/action_log_probsテンソルfloat64
ステップ/情報/qposテンソル(6,) float64
ステップ/情報/qvelテンソル(6,) float64
ステップ/is_firstテンソルブール
ステップ/is_lastテンソルブール
ステップ/is_terminalテンソルブール
ステップ/観察テンソル(11) float32
ステップ/報酬テンソルfloat32

d4rl_mujoco_hopper/v2-中

  • ダウンロードサイズ: 145.68 MiB

  • データセットのサイズ: 702.57 MiB

  • 自動キャッシュ(ドキュメント): いいえ

  • スプリット:

スプリット
'train' 2,187
  • 機能構造:
FeaturesDict({
    'algorithm': string,
    'iteration': int32,
    'policy': FeaturesDict({
        'fc0': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=float32),
            'weight': Tensor(shape=(256, 11), dtype=float32),
        }),
        'fc1': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=float32),
            'weight': Tensor(shape=(256, 256), dtype=float32),
        }),
        'last_fc': FeaturesDict({
            'bias': Tensor(shape=(3,), dtype=float32),
            'weight': Tensor(shape=(3, 256), dtype=float32),
        }),
        'last_fc_log_std': FeaturesDict({
            'bias': Tensor(shape=(3,), dtype=float32),
            'weight': Tensor(shape=(3, 256), dtype=float32),
        }),
        'nonlinearity': string,
        'output_distribution': string,
    }),
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_log_probs': float64,
            'qpos': Tensor(shape=(6,), dtype=float64),
            'qvel': Tensor(shape=(6,), dtype=float64),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(11,), dtype=float32),
        'reward': float32,
    }),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
アルゴリズムテンソルストリング
反復テンソルint32
ポリシー特徴辞書
ポリシー/fc0特徴辞書
ポリシー/fc0/バイアステンソル(256,) float32
ポリシー/fc0/重みテンソル(256,11) float32
ポリシー/fc1特徴辞書
ポリシー/fc1/バイアステンソル(256,) float32
ポリシー/fc1/重量テンソル(256, 256) float32
ポリシー/last_fc特徴辞書
ポリシー/last_fc/バイアステンソル(3) float32
ポリシー/last_fc/重量テンソル(3,256) float32
ポリシー/last_fc_log_std特徴辞書
ポリシー/last_fc_log_std/バイアステンソル(3) float32
ポリシー/last_fc_log_std/重量テンソル(3,256) float32
ポリシー/非線形性テンソルストリング
ポリシー/出力配布テンソルストリング
ステップデータセット
ステップ/アクションテンソル(3) float32
ステップ/割引テンソルfloat32
手順/情報特徴辞書
ステップ/情報/action_log_probsテンソルfloat64
ステップ/情報/qposテンソル(6,) float64
ステップ/情報/qvelテンソル(6,) float64
ステップ/is_firstテンソルブール
ステップ/is_lastテンソルブール
ステップ/is_terminalテンソルブール
ステップ/観察テンソル(11) float32
ステップ/報酬テンソルfloat32

d4rl_mujoco_hopper/v2-medium-expert

  • ダウンロードサイズ: 290.43 MiB

  • データセットサイズ: 228.28 MiB

  • 自動キャッシュ(ドキュメント): shuffle_files=Falseの場合のみ (トレーニング)

  • スプリット:

スプリット
'train' 3,214
  • 機能構造:
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_log_probs': float64,
            'qpos': Tensor(shape=(6,), dtype=float64),
            'qvel': Tensor(shape=(6,), dtype=float64),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(11,), dtype=float32),
        'reward': float32,
    }),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
ステップデータセット
ステップ/アクションテンソル(3) float32
ステップ/割引テンソルfloat32
手順/情報特徴辞書
ステップ/情報/action_log_probsテンソルfloat64
ステップ/情報/qposテンソル(6,) float64
ステップ/情報/qvelテンソル(6,) float64
ステップ/is_firstテンソルブール
ステップ/is_lastテンソルブール
ステップ/is_terminalテンソルブール
ステップ/観察テンソル(11) float32
ステップ/報酬テンソルfloat32

d4rl_mujoco_hopper/v2-medium-replay

  • ダウンロードサイズ: 72.34 MiB

  • データセットサイズ: 46.51 MiB

  • 自動キャッシュ(ドキュメント): はい

  • スプリット:

スプリット
'train' 2,041
  • 機能構造:
FeaturesDict({
    'algorithm': string,
    'iteration': int32,
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_log_probs': float64,
            'qpos': Tensor(shape=(6,), dtype=float64),
            'qvel': Tensor(shape=(6,), dtype=float64),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(11,), dtype=float32),
        'reward': float32,
    }),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
アルゴリズムテンソルストリング
反復テンソルint32
ステップデータセット
ステップ/アクションテンソル(3) float32
ステップ/割引テンソルfloat32
手順/情報特徴辞書
ステップ/情報/action_log_probsテンソルfloat64
ステップ/情報/qposテンソル(6,) float64
ステップ/情報/qvelテンソル(6,) float64
ステップ/is_firstテンソルブール
ステップ/is_lastテンソルブール
ステップ/is_terminalテンソルブール
ステップ/観察テンソル(11) float32
ステップ/報酬テンソルfloat32

d4rl_mujoco_hopper/v2-ランダム

  • ダウンロードサイズ: 145.46 MiB

  • データセットのサイズ: 130.72 MiB

  • 自動キャッシュ(ドキュメント): shuffle_files=Falseの場合のみ (トレーニング)

  • スプリット:

スプリット
'train' 45,240
  • 機能構造:
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=float32),
        'discount': float32,
        'infos': FeaturesDict({
            'action_log_probs': float64,
            'qpos': Tensor(shape=(6,), dtype=float64),
            'qvel': Tensor(shape=(6,), dtype=float64),
        }),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': Tensor(shape=(11,), dtype=float32),
        'reward': float32,
    }),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
ステップデータセット
ステップ/アクションテンソル(3) float32
ステップ/割引テンソルfloat32
手順/情報特徴辞書
ステップ/情報/action_log_probsテンソルfloat64
ステップ/情報/qposテンソル(6,) float64
ステップ/情報/qvelテンソル(6,) float64
ステップ/is_firstテンソルブール
ステップ/is_lastテンソルブール
ステップ/is_terminalテンソルブール
ステップ/観察テンソル(11) float32
ステップ/報酬テンソルfloat32