ロボットネット

  • 説明:

RoboNet には、113 のユニークなカメラ視点から撮影された、ロボットとオブジェクトの相互作用の 1,500 万を超えるビデオ フレームが含まれています。

  • アクションは、グリッパー ジョイント用に予約されたアクション ベクトルの 1 つの追加次元を持つ、ロボット エンドエフェクタに対する位置と回転のデルタです。

  • 状態は、回転が制限されたデカルト エンドエフェクタ制御アクション空間と、グリッパー ジョイントです。

  • 追加のドキュメント:コードを使用したペーパーの探索

  • ホームページ: https://www.robonet.wiki/

  • ソースコード: tfds.datasets.robonet.Builder

  • バージョン:

    • 4.0.1 (デフォルト): リリース ノートはありません。
  • 監視されたキー( as_supervised docを参照): None

  • ( tfds.show_examples ): サポートされていません。

  • 引用

@article{dasari2019robonet,
  title={RoboNet: Large-Scale Multi-Robot Learning},
  author={Dasari, Sudeep and Ebert, Frederik and Tian, Stephen and
  Nair, Suraj and Bucher, Bernadette and Schmeckpeper, Karl
  and Singh, Siddharth and Levine, Sergey and Finn, Chelsea},
  journal={arXiv preprint arXiv:1910.11215},
  year={2019}
}

robonet/robonet_sample_64 (デフォルト設定)

  • 構成の説明: 64x64 RoboNet サンプル。

  • ダウンロードサイズ: 119.80 MiB

  • データセットサイズ: 183.04 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' 700
  • 機能構造:
FeaturesDict({
    'actions': Tensor(shape=(None, 5), dtype=float32),
    'filename': Text(shape=(), dtype=string),
    'states': Tensor(shape=(None, 5), dtype=float32),
    'video': Video(Image(shape=(64, 64, 3), dtype=uint8)),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
行動テンソル(なし、5) float32
ファイル名文章ストリング
状態テンソル(なし、5) float32
ビデオ動画(画像) (なし、64、64、3) uint8

ロボネット/robonet_sample_128

  • 構成の説明: 128x128 RoboNet サンプル。

  • ダウンロードサイズ: 119.80 MiB

  • データセットのサイズ: 638.98 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' 700
  • 機能構造:
FeaturesDict({
    'actions': Tensor(shape=(None, 5), dtype=float32),
    'filename': Text(shape=(), dtype=string),
    'states': Tensor(shape=(None, 5), dtype=float32),
    'video': Video(Image(shape=(128, 128, 3), dtype=uint8)),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
行動テンソル(なし、5) float32
ファイル名文章ストリング
状態テンソル(なし、5) float32
ビデオ動画(画像) (なし、128、128、3) uint8

ロボネット/ロボネット_64

  • 構成の説明: 64x64 RoboNet。

  • ダウンロードサイズ: 36.20 GiB

  • データセットサイズ: 41.37 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' 162,417
  • 機能構造:
FeaturesDict({
    'actions': Tensor(shape=(None, 5), dtype=float32),
    'filename': Text(shape=(), dtype=string),
    'states': Tensor(shape=(None, 5), dtype=float32),
    'video': Video(Image(shape=(64, 64, 3), dtype=uint8)),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
行動テンソル(なし、5) float32
ファイル名文章ストリング
状態テンソル(なし、5) float32
ビデオ動画(画像) (なし、64、64、3) uint8

ロボネット/ロボネット_128

  • 構成の説明: 128x128 RoboNet。

  • ダウンロードサイズ: 36.20 GiB

  • データセットサイズ: 144.90 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' 162,417
  • 機能構造:
FeaturesDict({
    'actions': Tensor(shape=(None, 5), dtype=float32),
    'filename': Text(shape=(), dtype=string),
    'states': Tensor(shape=(None, 5), dtype=float32),
    'video': Video(Image(shape=(128, 128, 3), dtype=uint8)),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
行動テンソル(なし、5) float32
ファイル名文章ストリング
状態テンソル(なし、5) float32
ビデオ動画(画像) (なし、128、128、3) uint8