- 説明:
おもちゃのキッチンと対話する WidowX
ソースコード:
tfds.robotics.rtx.Bridge
バージョン:
-
0.1.0
(デフォルト): 初期リリース。
-
ダウンロードサイズ:
Unknown size
データセットのサイズ:
387.49 GiB
自動キャッシュ(ドキュメント): いいえ
分割:
スプリット | 例 |
---|---|
'test' | 3,475 |
'train' | 25,460 |
- 機能の構造:
FeaturesDict({
'steps': Dataset({
'action': FeaturesDict({
'open_gripper': bool,
'rotation_delta': Tensor(shape=(3,), dtype=float32),
'terminate_episode': float32,
'world_vector': Tensor(shape=(3,), dtype=float32),
}),
'is_first': bool,
'is_last': bool,
'is_terminal': bool,
'observation': FeaturesDict({
'image': Image(shape=(480, 640, 3), dtype=uint8),
'natural_language_embedding': Tensor(shape=(512,), dtype=float32),
'natural_language_instruction': string,
'state': Tensor(shape=(7,), dtype=float32),
}),
'reward': Scalar(shape=(), dtype=float32),
}),
})
- 機能ドキュメント:
特徴 | クラス | 形 | Dタイプ | 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 | 特徴辞書 | |||
ステップ/アクション/open_gripper | テンソル | ブール | ||
ステップ/アクション/回転デルタ | テンソル | (3,) | float32 | |
ステップ/アクション/terminate_episode | テンソル | float32 | ||
ステップ/アクション/world_vector | テンソル | (3,) | 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 | 特徴辞書 | |||
手順・観察・イメージ | 画像 | (480、640、3) | uint8 | |
ステップ/観察/natural_lang_embedding | テンソル | (512,) | float32 | |
ステップ/観察/natural_ language_instruction | テンソル | 弦 | ||
ステップ/観察/状態 | テンソル | (7、) | float32 | |
歩数/報酬 | スカラー | float32 |
監視キー(
as_supervised
docを参照):None
図( tfds.show_examples ): サポートされていません。
例( tfds.as_dataframe ):
- 引用:
@inproceedings{walke2023bridgedata,
title={BridgeData V2: A Dataset for Robot Learning at Scale},
author={Walke, Homer and Black, Kevin and Lee, Abraham and Kim, Moo Jin and Du, Max and Zheng, Chongyi and Zhao, Tony and Hansen-Estruch, Philippe and Vuong, Quan and He, Andre and Myers, Vivek and Fang, Kuan and Finn, Chelsea and Levine, Sergey},
booktitle={Conference on Robot Learning (CoRL)},
year={2023}
}