rlu_rwrl

• Description:

RL Unplugged is suite of benchmarks for offline reinforcement learning. The RL Unplugged is designed around the following considerations: to facilitate ease of use, we provide the datasets with a unified API which makes it easy for the practitioner to work with all data in the suite once a general pipeline has been established.

The datasets follow the RLDS format to represent steps and episodes.

Examples in the dataset represent SAR transitions stored when running a partially online trained agent as described in https://arxiv.org/abs/1904.12901 We follow the RLDS dataset format, as specified in https://github.com/google-research/rlds#dataset-format

We release 40 datasets on 8 tasks in total -- with no combined challenge and easy combined challenge on the cartpole, walker, quadruped, and humanoid tasks. Each task contains 5 different sizes of datasets, 1%, 5%, 20%, 40%, and 100%. Note that the smaller dataset is not guaranteed to be a subset of the larger ones. For details on how the dataset was generated, please refer to the paper.

• Homepage: https://github.com/deepmind/deepmind-research/tree/master/rl_unplugged

• Source code: tfds.rl_unplugged.rlu_rwrl.RluRwrl

• Versions:

  • 1.0.0: Initial release.
  • 1.0.1 (default): Fixes a bug in RLU RWRL dataset where there are duplicated episode ids in one of the humanoid datasets.

• Download size: Unknown size

• Supervised keys (See as_supervised doc): None

• Figure (tfds.show_examples): Not supported.

• Citation:

@misc{gulcehre2020rl,
    title={RL Unplugged: Benchmarks for Offline Reinforcement Learning},
    author={Caglar Gulcehre and Ziyu Wang and Alexander Novikov and Tom Le Paine
        and  Sergio Gómez Colmenarejo and Konrad Zolna and Rishabh Agarwal and
        Josh Merel and Daniel Mankowitz and Cosmin Paduraru and Gabriel
        Dulac-Arnold and Jerry Li and Mohammad Norouzi and Matt Hoffman and
        Ofir Nachum and George Tucker and Nicolas Heess and Nando deFreitas},
    year={2020},
    eprint={2006.13888},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

rlu_rwrl/cartpole_swingup_combined_challenge_none_1_percent (default config)

• Dataset size: 172.43 KiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	5

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(1,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'position': Tensor(shape=(3,), dtype=float32),
            'velocity': Tensor(shape=(2,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(1,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/position	Tensor	(3,)	float32
steps/observation/velocity	Tensor	(2,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/cartpole_swingup_combined_challenge_none_5_percent

• Dataset size: 862.13 KiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	25

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(1,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'position': Tensor(shape=(3,), dtype=float32),
            'velocity': Tensor(shape=(2,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(1,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/position	Tensor	(3,)	float32
steps/observation/velocity	Tensor	(2,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/cartpole_swingup_combined_challenge_none_20_percent

• Dataset size: 3.37 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	100

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(1,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'position': Tensor(shape=(3,), dtype=float32),
            'velocity': Tensor(shape=(2,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(1,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/position	Tensor	(3,)	float32
steps/observation/velocity	Tensor	(2,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/cartpole_swingup_combined_challenge_none_40_percent

• Dataset size: 6.74 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	200

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(1,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'position': Tensor(shape=(3,), dtype=float32),
            'velocity': Tensor(shape=(2,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(1,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/position	Tensor	(3,)	float32
steps/observation/velocity	Tensor	(2,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/cartpole_swingup_combined_challenge_none_100_percent

• Dataset size: 16.84 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	500

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(1,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'position': Tensor(shape=(3,), dtype=float32),
            'velocity': Tensor(shape=(2,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(1,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/position	Tensor	(3,)	float32
steps/observation/velocity	Tensor	(2,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/quadruped_walk_combined_challenge_none_1_percent

• Dataset size: 1.77 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	5

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(12,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'egocentric_state': Tensor(shape=(44,), dtype=float32),
            'force_torque': Tensor(shape=(24,), dtype=float32),
            'imu': Tensor(shape=(6,), dtype=float32),
            'torso_upright': Tensor(shape=(1,), dtype=float32),
            'torso_velocity': Tensor(shape=(3,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(12,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/egocentric_state	Tensor	(44,)	float32
steps/observation/force_torque	Tensor	(24,)	float32
steps/observation/imu	Tensor	(6,)	float32
steps/observation/torso_upright	Tensor	(1,)	float32
steps/observation/torso_velocity	Tensor	(3,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/quadruped_walk_combined_challenge_none_5_percent

• Dataset size: 8.86 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	25

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(12,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'egocentric_state': Tensor(shape=(44,), dtype=float32),
            'force_torque': Tensor(shape=(24,), dtype=float32),
            'imu': Tensor(shape=(6,), dtype=float32),
            'torso_upright': Tensor(shape=(1,), dtype=float32),
            'torso_velocity': Tensor(shape=(3,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(12,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/egocentric_state	Tensor	(44,)	float32
steps/observation/force_torque	Tensor	(24,)	float32
steps/observation/imu	Tensor	(6,)	float32
steps/observation/torso_upright	Tensor	(1,)	float32
steps/observation/torso_velocity	Tensor	(3,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/quadruped_walk_combined_challenge_none_20_percent

• Dataset size: 35.46 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	100

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(12,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'egocentric_state': Tensor(shape=(44,), dtype=float32),
            'force_torque': Tensor(shape=(24,), dtype=float32),
            'imu': Tensor(shape=(6,), dtype=float32),
            'torso_upright': Tensor(shape=(1,), dtype=float32),
            'torso_velocity': Tensor(shape=(3,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(12,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/egocentric_state	Tensor	(44,)	float32
steps/observation/force_torque	Tensor	(24,)	float32
steps/observation/imu	Tensor	(6,)	float32
steps/observation/torso_upright	Tensor	(1,)	float32
steps/observation/torso_velocity	Tensor	(3,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/quadruped_walk_combined_challenge_none_40_percent

• Dataset size: 70.92 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	200

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(12,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'egocentric_state': Tensor(shape=(44,), dtype=float32),
            'force_torque': Tensor(shape=(24,), dtype=float32),
            'imu': Tensor(shape=(6,), dtype=float32),
            'torso_upright': Tensor(shape=(1,), dtype=float32),
            'torso_velocity': Tensor(shape=(3,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(12,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/egocentric_state	Tensor	(44,)	float32
steps/observation/force_torque	Tensor	(24,)	float32
steps/observation/imu	Tensor	(6,)	float32
steps/observation/torso_upright	Tensor	(1,)	float32
steps/observation/torso_velocity	Tensor	(3,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/quadruped_walk_combined_challenge_none_100_percent

• Dataset size: 177.29 MiB

• Auto-cached (documentation): Only when shuffle_files=False (train)

• Splits:

Split	Examples
'train'	500

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(12,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'egocentric_state': Tensor(shape=(44,), dtype=float32),
            'force_torque': Tensor(shape=(24,), dtype=float32),
            'imu': Tensor(shape=(6,), dtype=float32),
            'torso_upright': Tensor(shape=(1,), dtype=float32),
            'torso_velocity': Tensor(shape=(3,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(12,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/egocentric_state	Tensor	(44,)	float32
steps/observation/force_torque	Tensor	(24,)	float32
steps/observation/imu	Tensor	(6,)	float32
steps/observation/torso_upright	Tensor	(1,)	float32
steps/observation/torso_velocity	Tensor	(3,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/walker_walk_combined_challenge_none_1_percent

• Dataset size: 6.27 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	50

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(6,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'height': Tensor(shape=(1,), dtype=float32),
            'orientations': Tensor(shape=(14,), dtype=float32),
            'velocity': Tensor(shape=(9,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(6,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/height	Tensor	(1,)	float32
steps/observation/orientations	Tensor	(14,)	float32
steps/observation/velocity	Tensor	(9,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/walker_walk_combined_challenge_none_5_percent

• Dataset size: 31.34 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	250

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(6,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'height': Tensor(shape=(1,), dtype=float32),
            'orientations': Tensor(shape=(14,), dtype=float32),
            'velocity': Tensor(shape=(9,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(6,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/height	Tensor	(1,)	float32
steps/observation/orientations	Tensor	(14,)	float32
steps/observation/velocity	Tensor	(9,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/walker_walk_combined_challenge_none_20_percent

• Dataset size: 125.37 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	1,000

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(6,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'height': Tensor(shape=(1,), dtype=float32),
            'orientations': Tensor(shape=(14,), dtype=float32),
            'velocity': Tensor(shape=(9,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(6,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/height	Tensor	(1,)	float32
steps/observation/orientations	Tensor	(14,)	float32
steps/observation/velocity	Tensor	(9,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/walker_walk_combined_challenge_none_40_percent

• Dataset size: 250.75 MiB

• Auto-cached (documentation): No

• Splits:

Split	Examples
'train'	2,000

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(6,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'height': Tensor(shape=(1,), dtype=float32),
            'orientations': Tensor(shape=(14,), dtype=float32),
            'velocity': Tensor(shape=(9,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(6,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/height	Tensor	(1,)	float32
steps/observation/orientations	Tensor	(14,)	float32
steps/observation/velocity	Tensor	(9,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/walker_walk_combined_challenge_none_100_percent

• Dataset size: 626.86 MiB

• Auto-cached (documentation): No

• Splits:

Split	Examples
'train'	5,000

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(6,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'height': Tensor(shape=(1,), dtype=float32),
            'orientations': Tensor(shape=(14,), dtype=float32),
            'velocity': Tensor(shape=(9,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(6,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/height	Tensor	(1,)	float32
steps/observation/orientations	Tensor	(14,)	float32
steps/observation/velocity	Tensor	(9,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/humanoid_walk_combined_challenge_none_1_percent

• Dataset size: 69.40 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	200

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(21,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'com_velocity': Tensor(shape=(3,), dtype=float32),
            'extremities': Tensor(shape=(12,), dtype=float32),
            'head_height': Tensor(shape=(1,), dtype=float32),
            'joint_angles': Tensor(shape=(21,), dtype=float32),
            'torso_vertical': Tensor(shape=(3,), dtype=float32),
            'velocity': Tensor(shape=(27,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(21,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/com_velocity	Tensor	(3,)	float32
steps/observation/extremities	Tensor	(12,)	float32
steps/observation/head_height	Tensor	(1,)	float32
steps/observation/joint_angles	Tensor	(21,)	float32
steps/observation/torso_vertical	Tensor	(3,)	float32
steps/observation/velocity	Tensor	(27,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/humanoid_walk_combined_challenge_none_5_percent

• Dataset size: 346.98 MiB

• Auto-cached (documentation): No

• Splits:

Split	Examples
'train'	1,000

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(21,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'com_velocity': Tensor(shape=(3,), dtype=float32),
            'extremities': Tensor(shape=(12,), dtype=float32),
            'head_height': Tensor(shape=(1,), dtype=float32),
            'joint_angles': Tensor(shape=(21,), dtype=float32),
            'torso_vertical': Tensor(shape=(3,), dtype=float32),
            'velocity': Tensor(shape=(27,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(21,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/com_velocity	Tensor	(3,)	float32
steps/observation/extremities	Tensor	(12,)	float32
steps/observation/head_height	Tensor	(1,)	float32
steps/observation/joint_angles	Tensor	(21,)	float32
steps/observation/torso_vertical	Tensor	(3,)	float32
steps/observation/velocity	Tensor	(27,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/humanoid_walk_combined_challenge_none_20_percent

• Dataset size: 1.36 GiB

• Auto-cached (documentation): No

• Splits:

Split	Examples
'train'	4,000

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(21,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'com_velocity': Tensor(shape=(3,), dtype=float32),
            'extremities': Tensor(shape=(12,), dtype=float32),
            'head_height': Tensor(shape=(1,), dtype=float32),
            'joint_angles': Tensor(shape=(21,), dtype=float32),
            'torso_vertical': Tensor(shape=(3,), dtype=float32),
            'velocity': Tensor(shape=(27,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(21,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/com_velocity	Tensor	(3,)	float32
steps/observation/extremities	Tensor	(12,)	float32
steps/observation/head_height	Tensor	(1,)	float32
steps/observation/joint_angles	Tensor	(21,)	float32
steps/observation/torso_vertical	Tensor	(3,)	float32
steps/observation/velocity	Tensor	(27,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/humanoid_walk_combined_challenge_none_40_percent

• Dataset size: 2.71 GiB

• Auto-cached (documentation): No

• Splits:

Split	Examples
'train'	8,000

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(21,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'com_velocity': Tensor(shape=(3,), dtype=float32),
            'extremities': Tensor(shape=(12,), dtype=float32),
            'head_height': Tensor(shape=(1,), dtype=float32),
            'joint_angles': Tensor(shape=(21,), dtype=float32),
            'torso_vertical': Tensor(shape=(3,), dtype=float32),
            'velocity': Tensor(shape=(27,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(21,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/com_velocity	Tensor	(3,)	float32
steps/observation/extremities	Tensor	(12,)	float32
steps/observation/head_height	Tensor	(1,)	float32
steps/observation/joint_angles	Tensor	(21,)	float32
steps/observation/torso_vertical	Tensor	(3,)	float32
steps/observation/velocity	Tensor	(27,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/humanoid_walk_combined_challenge_none_100_percent

• Dataset size: 6.78 GiB

• Auto-cached (documentation): No

• Splits:

Split	Examples
'train'	20,000

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(21,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'com_velocity': Tensor(shape=(3,), dtype=float32),
            'extremities': Tensor(shape=(12,), dtype=float32),
            'head_height': Tensor(shape=(1,), dtype=float32),
            'joint_angles': Tensor(shape=(21,), dtype=float32),
            'torso_vertical': Tensor(shape=(3,), dtype=float32),
            'velocity': Tensor(shape=(27,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(21,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/com_velocity	Tensor	(3,)	float32
steps/observation/extremities	Tensor	(12,)	float32
steps/observation/head_height	Tensor	(1,)	float32
steps/observation/joint_angles	Tensor	(21,)	float32
steps/observation/torso_vertical	Tensor	(3,)	float32
steps/observation/velocity	Tensor	(27,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/cartpole_swingup_combined_challenge_easy_1_percent

• Dataset size: 369.84 KiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	5

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(1,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'dummy-0': Tensor(shape=(1,), dtype=float32),
            'dummy-1': Tensor(shape=(1,), dtype=float32),
            'dummy-2': Tensor(shape=(1,), dtype=float32),
            'dummy-3': Tensor(shape=(1,), dtype=float32),
            'dummy-4': Tensor(shape=(1,), dtype=float32),
            'dummy-5': Tensor(shape=(1,), dtype=float32),
            'dummy-6': Tensor(shape=(1,), dtype=float32),
            'dummy-7': Tensor(shape=(1,), dtype=float32),
            'dummy-8': Tensor(shape=(1,), dtype=float32),
            'dummy-9': Tensor(shape=(1,), dtype=float32),
            'position': Tensor(shape=(3,), dtype=float32),
            'velocity': Tensor(shape=(2,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(1,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/dummy-0	Tensor	(1,)	float32
steps/observation/dummy-1	Tensor	(1,)	float32
steps/observation/dummy-2	Tensor	(1,)	float32
steps/observation/dummy-3	Tensor	(1,)	float32
steps/observation/dummy-4	Tensor	(1,)	float32
steps/observation/dummy-5	Tensor	(1,)	float32
steps/observation/dummy-6	Tensor	(1,)	float32
steps/observation/dummy-7	Tensor	(1,)	float32
steps/observation/dummy-8	Tensor	(1,)	float32
steps/observation/dummy-9	Tensor	(1,)	float32
steps/observation/position	Tensor	(3,)	float32
steps/observation/velocity	Tensor	(2,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/cartpole_swingup_combined_challenge_easy_5_percent

• Dataset size: 1.81 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	25

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(1,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'dummy-0': Tensor(shape=(1,), dtype=float32),
            'dummy-1': Tensor(shape=(1,), dtype=float32),
            'dummy-2': Tensor(shape=(1,), dtype=float32),
            'dummy-3': Tensor(shape=(1,), dtype=float32),
            'dummy-4': Tensor(shape=(1,), dtype=float32),
            'dummy-5': Tensor(shape=(1,), dtype=float32),
            'dummy-6': Tensor(shape=(1,), dtype=float32),
            'dummy-7': Tensor(shape=(1,), dtype=float32),
            'dummy-8': Tensor(shape=(1,), dtype=float32),
            'dummy-9': Tensor(shape=(1,), dtype=float32),
            'position': Tensor(shape=(3,), dtype=float32),
            'velocity': Tensor(shape=(2,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(1,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/dummy-0	Tensor	(1,)	float32
steps/observation/dummy-1	Tensor	(1,)	float32
steps/observation/dummy-2	Tensor	(1,)	float32
steps/observation/dummy-3	Tensor	(1,)	float32
steps/observation/dummy-4	Tensor	(1,)	float32
steps/observation/dummy-5	Tensor	(1,)	float32
steps/observation/dummy-6	Tensor	(1,)	float32
steps/observation/dummy-7	Tensor	(1,)	float32
steps/observation/dummy-8	Tensor	(1,)	float32
steps/observation/dummy-9	Tensor	(1,)	float32
steps/observation/position	Tensor	(3,)	float32
steps/observation/velocity	Tensor	(2,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/cartpole_swingup_combined_challenge_easy_20_percent

• Dataset size: 7.22 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	100

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(1,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'dummy-0': Tensor(shape=(1,), dtype=float32),
            'dummy-1': Tensor(shape=(1,), dtype=float32),
            'dummy-2': Tensor(shape=(1,), dtype=float32),
            'dummy-3': Tensor(shape=(1,), dtype=float32),
            'dummy-4': Tensor(shape=(1,), dtype=float32),
            'dummy-5': Tensor(shape=(1,), dtype=float32),
            'dummy-6': Tensor(shape=(1,), dtype=float32),
            'dummy-7': Tensor(shape=(1,), dtype=float32),
            'dummy-8': Tensor(shape=(1,), dtype=float32),
            'dummy-9': Tensor(shape=(1,), dtype=float32),
            'position': Tensor(shape=(3,), dtype=float32),
            'velocity': Tensor(shape=(2,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(1,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/dummy-0	Tensor	(1,)	float32
steps/observation/dummy-1	Tensor	(1,)	float32
steps/observation/dummy-2	Tensor	(1,)	float32
steps/observation/dummy-3	Tensor	(1,)	float32
steps/observation/dummy-4	Tensor	(1,)	float32
steps/observation/dummy-5	Tensor	(1,)	float32
steps/observation/dummy-6	Tensor	(1,)	float32
steps/observation/dummy-7	Tensor	(1,)	float32
steps/observation/dummy-8	Tensor	(1,)	float32
steps/observation/dummy-9	Tensor	(1,)	float32
steps/observation/position	Tensor	(3,)	float32
steps/observation/velocity	Tensor	(2,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/cartpole_swingup_combined_challenge_easy_40_percent

• Dataset size: 14.45 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	200

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(1,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'dummy-0': Tensor(shape=(1,), dtype=float32),
            'dummy-1': Tensor(shape=(1,), dtype=float32),
            'dummy-2': Tensor(shape=(1,), dtype=float32),
            'dummy-3': Tensor(shape=(1,), dtype=float32),
            'dummy-4': Tensor(shape=(1,), dtype=float32),
            'dummy-5': Tensor(shape=(1,), dtype=float32),
            'dummy-6': Tensor(shape=(1,), dtype=float32),
            'dummy-7': Tensor(shape=(1,), dtype=float32),
            'dummy-8': Tensor(shape=(1,), dtype=float32),
            'dummy-9': Tensor(shape=(1,), dtype=float32),
            'position': Tensor(shape=(3,), dtype=float32),
            'velocity': Tensor(shape=(2,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(1,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/dummy-0	Tensor	(1,)	float32
steps/observation/dummy-1	Tensor	(1,)	float32
steps/observation/dummy-2	Tensor	(1,)	float32
steps/observation/dummy-3	Tensor	(1,)	float32
steps/observation/dummy-4	Tensor	(1,)	float32
steps/observation/dummy-5	Tensor	(1,)	float32
steps/observation/dummy-6	Tensor	(1,)	float32
steps/observation/dummy-7	Tensor	(1,)	float32
steps/observation/dummy-8	Tensor	(1,)	float32
steps/observation/dummy-9	Tensor	(1,)	float32
steps/observation/position	Tensor	(3,)	float32
steps/observation/velocity	Tensor	(2,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/cartpole_swingup_combined_challenge_easy_100_percent

• Dataset size: 36.12 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	500

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(1,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'dummy-0': Tensor(shape=(1,), dtype=float32),
            'dummy-1': Tensor(shape=(1,), dtype=float32),
            'dummy-2': Tensor(shape=(1,), dtype=float32),
            'dummy-3': Tensor(shape=(1,), dtype=float32),
            'dummy-4': Tensor(shape=(1,), dtype=float32),
            'dummy-5': Tensor(shape=(1,), dtype=float32),
            'dummy-6': Tensor(shape=(1,), dtype=float32),
            'dummy-7': Tensor(shape=(1,), dtype=float32),
            'dummy-8': Tensor(shape=(1,), dtype=float32),
            'dummy-9': Tensor(shape=(1,), dtype=float32),
            'position': Tensor(shape=(3,), dtype=float32),
            'velocity': Tensor(shape=(2,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(1,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/dummy-0	Tensor	(1,)	float32
steps/observation/dummy-1	Tensor	(1,)	float32
steps/observation/dummy-2	Tensor	(1,)	float32
steps/observation/dummy-3	Tensor	(1,)	float32
steps/observation/dummy-4	Tensor	(1,)	float32
steps/observation/dummy-5	Tensor	(1,)	float32
steps/observation/dummy-6	Tensor	(1,)	float32
steps/observation/dummy-7	Tensor	(1,)	float32
steps/observation/dummy-8	Tensor	(1,)	float32
steps/observation/dummy-9	Tensor	(1,)	float32
steps/observation/position	Tensor	(3,)	float32
steps/observation/velocity	Tensor	(2,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/quadruped_walk_combined_challenge_easy_1_percent

• Dataset size: 1.97 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	5

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(12,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'dummy-0': Tensor(shape=(1,), dtype=float32),
            'dummy-1': Tensor(shape=(1,), dtype=float32),
            'dummy-2': Tensor(shape=(1,), dtype=float32),
            'dummy-3': Tensor(shape=(1,), dtype=float32),
            'dummy-4': Tensor(shape=(1,), dtype=float32),
            'dummy-5': Tensor(shape=(1,), dtype=float32),
            'dummy-6': Tensor(shape=(1,), dtype=float32),
            'dummy-7': Tensor(shape=(1,), dtype=float32),
            'dummy-8': Tensor(shape=(1,), dtype=float32),
            'dummy-9': Tensor(shape=(1,), dtype=float32),
            'egocentric_state': Tensor(shape=(44,), dtype=float32),
            'force_torque': Tensor(shape=(24,), dtype=float32),
            'imu': Tensor(shape=(6,), dtype=float32),
            'torso_upright': Tensor(shape=(1,), dtype=float32),
            'torso_velocity': Tensor(shape=(3,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(12,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/dummy-0	Tensor	(1,)	float32
steps/observation/dummy-1	Tensor	(1,)	float32
steps/observation/dummy-2	Tensor	(1,)	float32
steps/observation/dummy-3	Tensor	(1,)	float32
steps/observation/dummy-4	Tensor	(1,)	float32
steps/observation/dummy-5	Tensor	(1,)	float32
steps/observation/dummy-6	Tensor	(1,)	float32
steps/observation/dummy-7	Tensor	(1,)	float32
steps/observation/dummy-8	Tensor	(1,)	float32
steps/observation/dummy-9	Tensor	(1,)	float32
steps/observation/egocentric_state	Tensor	(44,)	float32
steps/observation/force_torque	Tensor	(24,)	float32
steps/observation/imu	Tensor	(6,)	float32
steps/observation/torso_upright	Tensor	(1,)	float32
steps/observation/torso_velocity	Tensor	(3,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/quadruped_walk_combined_challenge_easy_5_percent

• Dataset size: 9.83 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	25

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(12,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'dummy-0': Tensor(shape=(1,), dtype=float32),
            'dummy-1': Tensor(shape=(1,), dtype=float32),
            'dummy-2': Tensor(shape=(1,), dtype=float32),
            'dummy-3': Tensor(shape=(1,), dtype=float32),
            'dummy-4': Tensor(shape=(1,), dtype=float32),
            'dummy-5': Tensor(shape=(1,), dtype=float32),
            'dummy-6': Tensor(shape=(1,), dtype=float32),
            'dummy-7': Tensor(shape=(1,), dtype=float32),
            'dummy-8': Tensor(shape=(1,), dtype=float32),
            'dummy-9': Tensor(shape=(1,), dtype=float32),
            'egocentric_state': Tensor(shape=(44,), dtype=float32),
            'force_torque': Tensor(shape=(24,), dtype=float32),
            'imu': Tensor(shape=(6,), dtype=float32),
            'torso_upright': Tensor(shape=(1,), dtype=float32),
            'torso_velocity': Tensor(shape=(3,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(12,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/dummy-0	Tensor	(1,)	float32
steps/observation/dummy-1	Tensor	(1,)	float32
steps/observation/dummy-2	Tensor	(1,)	float32
steps/observation/dummy-3	Tensor	(1,)	float32
steps/observation/dummy-4	Tensor	(1,)	float32
steps/observation/dummy-5	Tensor	(1,)	float32
steps/observation/dummy-6	Tensor	(1,)	float32
steps/observation/dummy-7	Tensor	(1,)	float32
steps/observation/dummy-8	Tensor	(1,)	float32
steps/observation/dummy-9	Tensor	(1,)	float32
steps/observation/egocentric_state	Tensor	(44,)	float32
steps/observation/force_torque	Tensor	(24,)	float32
steps/observation/imu	Tensor	(6,)	float32
steps/observation/torso_upright	Tensor	(1,)	float32
steps/observation/torso_velocity	Tensor	(3,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/quadruped_walk_combined_challenge_easy_20_percent

• Dataset size: 39.31 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	100

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(12,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'dummy-0': Tensor(shape=(1,), dtype=float32),
            'dummy-1': Tensor(shape=(1,), dtype=float32),
            'dummy-2': Tensor(shape=(1,), dtype=float32),
            'dummy-3': Tensor(shape=(1,), dtype=float32),
            'dummy-4': Tensor(shape=(1,), dtype=float32),
            'dummy-5': Tensor(shape=(1,), dtype=float32),
            'dummy-6': Tensor(shape=(1,), dtype=float32),
            'dummy-7': Tensor(shape=(1,), dtype=float32),
            'dummy-8': Tensor(shape=(1,), dtype=float32),
            'dummy-9': Tensor(shape=(1,), dtype=float32),
            'egocentric_state': Tensor(shape=(44,), dtype=float32),
            'force_torque': Tensor(shape=(24,), dtype=float32),
            'imu': Tensor(shape=(6,), dtype=float32),
            'torso_upright': Tensor(shape=(1,), dtype=float32),
            'torso_velocity': Tensor(shape=(3,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(12,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/dummy-0	Tensor	(1,)	float32
steps/observation/dummy-1	Tensor	(1,)	float32
steps/observation/dummy-2	Tensor	(1,)	float32
steps/observation/dummy-3	Tensor	(1,)	float32
steps/observation/dummy-4	Tensor	(1,)	float32
steps/observation/dummy-5	Tensor	(1,)	float32
steps/observation/dummy-6	Tensor	(1,)	float32
steps/observation/dummy-7	Tensor	(1,)	float32
steps/observation/dummy-8	Tensor	(1,)	float32
steps/observation/dummy-9	Tensor	(1,)	float32
steps/observation/egocentric_state	Tensor	(44,)	float32
steps/observation/force_torque	Tensor	(24,)	float32
steps/observation/imu	Tensor	(6,)	float32
steps/observation/torso_upright	Tensor	(1,)	float32
steps/observation/torso_velocity	Tensor	(3,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/quadruped_walk_combined_challenge_easy_40_percent

• Dataset size: 78.63 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	200

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(12,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'dummy-0': Tensor(shape=(1,), dtype=float32),
            'dummy-1': Tensor(shape=(1,), dtype=float32),
            'dummy-2': Tensor(shape=(1,), dtype=float32),
            'dummy-3': Tensor(shape=(1,), dtype=float32),
            'dummy-4': Tensor(shape=(1,), dtype=float32),
            'dummy-5': Tensor(shape=(1,), dtype=float32),
            'dummy-6': Tensor(shape=(1,), dtype=float32),
            'dummy-7': Tensor(shape=(1,), dtype=float32),
            'dummy-8': Tensor(shape=(1,), dtype=float32),
            'dummy-9': Tensor(shape=(1,), dtype=float32),
            'egocentric_state': Tensor(shape=(44,), dtype=float32),
            'force_torque': Tensor(shape=(24,), dtype=float32),
            'imu': Tensor(shape=(6,), dtype=float32),
            'torso_upright': Tensor(shape=(1,), dtype=float32),
            'torso_velocity': Tensor(shape=(3,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(12,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/dummy-0	Tensor	(1,)	float32
steps/observation/dummy-1	Tensor	(1,)	float32
steps/observation/dummy-2	Tensor	(1,)	float32
steps/observation/dummy-3	Tensor	(1,)	float32
steps/observation/dummy-4	Tensor	(1,)	float32
steps/observation/dummy-5	Tensor	(1,)	float32
steps/observation/dummy-6	Tensor	(1,)	float32
steps/observation/dummy-7	Tensor	(1,)	float32
steps/observation/dummy-8	Tensor	(1,)	float32
steps/observation/dummy-9	Tensor	(1,)	float32
steps/observation/egocentric_state	Tensor	(44,)	float32
steps/observation/force_torque	Tensor	(24,)	float32
steps/observation/imu	Tensor	(6,)	float32
steps/observation/torso_upright	Tensor	(1,)	float32
steps/observation/torso_velocity	Tensor	(3,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/quadruped_walk_combined_challenge_easy_100_percent

• Dataset size: 196.57 MiB

• Auto-cached (documentation): Only when shuffle_files=False (train)

• Splits:

Split	Examples
'train'	500

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(12,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'dummy-0': Tensor(shape=(1,), dtype=float32),
            'dummy-1': Tensor(shape=(1,), dtype=float32),
            'dummy-2': Tensor(shape=(1,), dtype=float32),
            'dummy-3': Tensor(shape=(1,), dtype=float32),
            'dummy-4': Tensor(shape=(1,), dtype=float32),
            'dummy-5': Tensor(shape=(1,), dtype=float32),
            'dummy-6': Tensor(shape=(1,), dtype=float32),
            'dummy-7': Tensor(shape=(1,), dtype=float32),
            'dummy-8': Tensor(shape=(1,), dtype=float32),
            'dummy-9': Tensor(shape=(1,), dtype=float32),
            'egocentric_state': Tensor(shape=(44,), dtype=float32),
            'force_torque': Tensor(shape=(24,), dtype=float32),
            'imu': Tensor(shape=(6,), dtype=float32),
            'torso_upright': Tensor(shape=(1,), dtype=float32),
            'torso_velocity': Tensor(shape=(3,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(12,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/dummy-0	Tensor	(1,)	float32
steps/observation/dummy-1	Tensor	(1,)	float32
steps/observation/dummy-2	Tensor	(1,)	float32
steps/observation/dummy-3	Tensor	(1,)	float32
steps/observation/dummy-4	Tensor	(1,)	float32
steps/observation/dummy-5	Tensor	(1,)	float32
steps/observation/dummy-6	Tensor	(1,)	float32
steps/observation/dummy-7	Tensor	(1,)	float32
steps/observation/dummy-8	Tensor	(1,)	float32
steps/observation/dummy-9	Tensor	(1,)	float32
steps/observation/egocentric_state	Tensor	(44,)	float32
steps/observation/force_torque	Tensor	(24,)	float32
steps/observation/imu	Tensor	(6,)	float32
steps/observation/torso_upright	Tensor	(1,)	float32
steps/observation/torso_velocity	Tensor	(3,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/walker_walk_combined_challenge_easy_1_percent

• Dataset size: 8.20 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	50

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(6,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'dummy-0': Tensor(shape=(1,), dtype=float32),
            'dummy-1': Tensor(shape=(1,), dtype=float32),
            'dummy-2': Tensor(shape=(1,), dtype=float32),
            'dummy-3': Tensor(shape=(1,), dtype=float32),
            'dummy-4': Tensor(shape=(1,), dtype=float32),
            'dummy-5': Tensor(shape=(1,), dtype=float32),
            'dummy-6': Tensor(shape=(1,), dtype=float32),
            'dummy-7': Tensor(shape=(1,), dtype=float32),
            'dummy-8': Tensor(shape=(1,), dtype=float32),
            'dummy-9': Tensor(shape=(1,), dtype=float32),
            'height': Tensor(shape=(1,), dtype=float32),
            'orientations': Tensor(shape=(14,), dtype=float32),
            'velocity': Tensor(shape=(9,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(6,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/dummy-0	Tensor	(1,)	float32
steps/observation/dummy-1	Tensor	(1,)	float32
steps/observation/dummy-2	Tensor	(1,)	float32
steps/observation/dummy-3	Tensor	(1,)	float32
steps/observation/dummy-4	Tensor	(1,)	float32
steps/observation/dummy-5	Tensor	(1,)	float32
steps/observation/dummy-6	Tensor	(1,)	float32
steps/observation/dummy-7	Tensor	(1,)	float32
steps/observation/dummy-8	Tensor	(1,)	float32
steps/observation/dummy-9	Tensor	(1,)	float32
steps/observation/height	Tensor	(1,)	float32
steps/observation/orientations	Tensor	(14,)	float32
steps/observation/velocity	Tensor	(9,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/walker_walk_combined_challenge_easy_5_percent

• Dataset size: 40.98 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	250

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(6,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'dummy-0': Tensor(shape=(1,), dtype=float32),
            'dummy-1': Tensor(shape=(1,), dtype=float32),
            'dummy-2': Tensor(shape=(1,), dtype=float32),
            'dummy-3': Tensor(shape=(1,), dtype=float32),
            'dummy-4': Tensor(shape=(1,), dtype=float32),
            'dummy-5': Tensor(shape=(1,), dtype=float32),
            'dummy-6': Tensor(shape=(1,), dtype=float32),
            'dummy-7': Tensor(shape=(1,), dtype=float32),
            'dummy-8': Tensor(shape=(1,), dtype=float32),
            'dummy-9': Tensor(shape=(1,), dtype=float32),
            'height': Tensor(shape=(1,), dtype=float32),
            'orientations': Tensor(shape=(14,), dtype=float32),
            'velocity': Tensor(shape=(9,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(6,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/dummy-0	Tensor	(1,)	float32
steps/observation/dummy-1	Tensor	(1,)	float32
steps/observation/dummy-2	Tensor	(1,)	float32
steps/observation/dummy-3	Tensor	(1,)	float32
steps/observation/dummy-4	Tensor	(1,)	float32
steps/observation/dummy-5	Tensor	(1,)	float32
steps/observation/dummy-6	Tensor	(1,)	float32
steps/observation/dummy-7	Tensor	(1,)	float32
steps/observation/dummy-8	Tensor	(1,)	float32
steps/observation/dummy-9	Tensor	(1,)	float32
steps/observation/height	Tensor	(1,)	float32
steps/observation/orientations	Tensor	(14,)	float32
steps/observation/velocity	Tensor	(9,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/walker_walk_combined_challenge_easy_20_percent

• Dataset size: 163.93 MiB

• Auto-cached (documentation): Only when shuffle_files=False (train)

• Splits:

Split	Examples
'train'	1,000

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(6,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'dummy-0': Tensor(shape=(1,), dtype=float32),
            'dummy-1': Tensor(shape=(1,), dtype=float32),
            'dummy-2': Tensor(shape=(1,), dtype=float32),
            'dummy-3': Tensor(shape=(1,), dtype=float32),
            'dummy-4': Tensor(shape=(1,), dtype=float32),
            'dummy-5': Tensor(shape=(1,), dtype=float32),
            'dummy-6': Tensor(shape=(1,), dtype=float32),
            'dummy-7': Tensor(shape=(1,), dtype=float32),
            'dummy-8': Tensor(shape=(1,), dtype=float32),
            'dummy-9': Tensor(shape=(1,), dtype=float32),
            'height': Tensor(shape=(1,), dtype=float32),
            'orientations': Tensor(shape=(14,), dtype=float32),
            'velocity': Tensor(shape=(9,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(6,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/dummy-0	Tensor	(1,)	float32
steps/observation/dummy-1	Tensor	(1,)	float32
steps/observation/dummy-2	Tensor	(1,)	float32
steps/observation/dummy-3	Tensor	(1,)	float32
steps/observation/dummy-4	Tensor	(1,)	float32
steps/observation/dummy-5	Tensor	(1,)	float32
steps/observation/dummy-6	Tensor	(1,)	float32
steps/observation/dummy-7	Tensor	(1,)	float32
steps/observation/dummy-8	Tensor	(1,)	float32
steps/observation/dummy-9	Tensor	(1,)	float32
steps/observation/height	Tensor	(1,)	float32
steps/observation/orientations	Tensor	(14,)	float32
steps/observation/velocity	Tensor	(9,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/walker_walk_combined_challenge_easy_40_percent

• Dataset size: 327.86 MiB

• Auto-cached (documentation): No

• Splits:

Split	Examples
'train'	2,000

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(6,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'dummy-0': Tensor(shape=(1,), dtype=float32),
            'dummy-1': Tensor(shape=(1,), dtype=float32),
            'dummy-2': Tensor(shape=(1,), dtype=float32),
            'dummy-3': Tensor(shape=(1,), dtype=float32),
            'dummy-4': Tensor(shape=(1,), dtype=float32),
            'dummy-5': Tensor(shape=(1,), dtype=float32),
            'dummy-6': Tensor(shape=(1,), dtype=float32),
            'dummy-7': Tensor(shape=(1,), dtype=float32),
            'dummy-8': Tensor(shape=(1,), dtype=float32),
            'dummy-9': Tensor(shape=(1,), dtype=float32),
            'height': Tensor(shape=(1,), dtype=float32),
            'orientations': Tensor(shape=(14,), dtype=float32),
            'velocity': Tensor(shape=(9,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(6,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/dummy-0	Tensor	(1,)	float32
steps/observation/dummy-1	Tensor	(1,)	float32
steps/observation/dummy-2	Tensor	(1,)	float32
steps/observation/dummy-3	Tensor	(1,)	float32
steps/observation/dummy-4	Tensor	(1,)	float32
steps/observation/dummy-5	Tensor	(1,)	float32
steps/observation/dummy-6	Tensor	(1,)	float32
steps/observation/dummy-7	Tensor	(1,)	float32
steps/observation/dummy-8	Tensor	(1,)	float32
steps/observation/dummy-9	Tensor	(1,)	float32
steps/observation/height	Tensor	(1,)	float32
steps/observation/orientations	Tensor	(14,)	float32
steps/observation/velocity	Tensor	(9,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/walker_walk_combined_challenge_easy_100_percent

• Dataset size: 819.65 MiB

• Auto-cached (documentation): No

• Splits:

Split	Examples
'train'	5,000

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(6,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'dummy-0': Tensor(shape=(1,), dtype=float32),
            'dummy-1': Tensor(shape=(1,), dtype=float32),
            'dummy-2': Tensor(shape=(1,), dtype=float32),
            'dummy-3': Tensor(shape=(1,), dtype=float32),
            'dummy-4': Tensor(shape=(1,), dtype=float32),
            'dummy-5': Tensor(shape=(1,), dtype=float32),
            'dummy-6': Tensor(shape=(1,), dtype=float32),
            'dummy-7': Tensor(shape=(1,), dtype=float32),
            'dummy-8': Tensor(shape=(1,), dtype=float32),
            'dummy-9': Tensor(shape=(1,), dtype=float32),
            'height': Tensor(shape=(1,), dtype=float32),
            'orientations': Tensor(shape=(14,), dtype=float32),
            'velocity': Tensor(shape=(9,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(6,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/dummy-0	Tensor	(1,)	float32
steps/observation/dummy-1	Tensor	(1,)	float32
steps/observation/dummy-2	Tensor	(1,)	float32
steps/observation/dummy-3	Tensor	(1,)	float32
steps/observation/dummy-4	Tensor	(1,)	float32
steps/observation/dummy-5	Tensor	(1,)	float32
steps/observation/dummy-6	Tensor	(1,)	float32
steps/observation/dummy-7	Tensor	(1,)	float32
steps/observation/dummy-8	Tensor	(1,)	float32
steps/observation/dummy-9	Tensor	(1,)	float32
steps/observation/height	Tensor	(1,)	float32
steps/observation/orientations	Tensor	(14,)	float32
steps/observation/velocity	Tensor	(9,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/humanoid_walk_combined_challenge_easy_1_percent

• Dataset size: 77.11 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'train'	200

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(21,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'com_velocity': Tensor(shape=(3,), dtype=float32),
            'dummy-0': Tensor(shape=(1,), dtype=float32),
            'dummy-1': Tensor(shape=(1,), dtype=float32),
            'dummy-2': Tensor(shape=(1,), dtype=float32),
            'dummy-3': Tensor(shape=(1,), dtype=float32),
            'dummy-4': Tensor(shape=(1,), dtype=float32),
            'dummy-5': Tensor(shape=(1,), dtype=float32),
            'dummy-6': Tensor(shape=(1,), dtype=float32),
            'dummy-7': Tensor(shape=(1,), dtype=float32),
            'dummy-8': Tensor(shape=(1,), dtype=float32),
            'dummy-9': Tensor(shape=(1,), dtype=float32),
            'extremities': Tensor(shape=(12,), dtype=float32),
            'head_height': Tensor(shape=(1,), dtype=float32),
            'joint_angles': Tensor(shape=(21,), dtype=float32),
            'torso_vertical': Tensor(shape=(3,), dtype=float32),
            'velocity': Tensor(shape=(27,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(21,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/com_velocity	Tensor	(3,)	float32
steps/observation/dummy-0	Tensor	(1,)	float32
steps/observation/dummy-1	Tensor	(1,)	float32
steps/observation/dummy-2	Tensor	(1,)	float32
steps/observation/dummy-3	Tensor	(1,)	float32
steps/observation/dummy-4	Tensor	(1,)	float32
steps/observation/dummy-5	Tensor	(1,)	float32
steps/observation/dummy-6	Tensor	(1,)	float32
steps/observation/dummy-7	Tensor	(1,)	float32
steps/observation/dummy-8	Tensor	(1,)	float32
steps/observation/dummy-9	Tensor	(1,)	float32
steps/observation/extremities	Tensor	(12,)	float32
steps/observation/head_height	Tensor	(1,)	float32
steps/observation/joint_angles	Tensor	(21,)	float32
steps/observation/torso_vertical	Tensor	(3,)	float32
steps/observation/velocity	Tensor	(27,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/humanoid_walk_combined_challenge_easy_5_percent

• Dataset size: 385.54 MiB

• Auto-cached (documentation): No

• Splits:

Split	Examples
'train'	1,000

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(21,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'com_velocity': Tensor(shape=(3,), dtype=float32),
            'dummy-0': Tensor(shape=(1,), dtype=float32),
            'dummy-1': Tensor(shape=(1,), dtype=float32),
            'dummy-2': Tensor(shape=(1,), dtype=float32),
            'dummy-3': Tensor(shape=(1,), dtype=float32),
            'dummy-4': Tensor(shape=(1,), dtype=float32),
            'dummy-5': Tensor(shape=(1,), dtype=float32),
            'dummy-6': Tensor(shape=(1,), dtype=float32),
            'dummy-7': Tensor(shape=(1,), dtype=float32),
            'dummy-8': Tensor(shape=(1,), dtype=float32),
            'dummy-9': Tensor(shape=(1,), dtype=float32),
            'extremities': Tensor(shape=(12,), dtype=float32),
            'head_height': Tensor(shape=(1,), dtype=float32),
            'joint_angles': Tensor(shape=(21,), dtype=float32),
            'torso_vertical': Tensor(shape=(3,), dtype=float32),
            'velocity': Tensor(shape=(27,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(21,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/com_velocity	Tensor	(3,)	float32
steps/observation/dummy-0	Tensor	(1,)	float32
steps/observation/dummy-1	Tensor	(1,)	float32
steps/observation/dummy-2	Tensor	(1,)	float32
steps/observation/dummy-3	Tensor	(1,)	float32
steps/observation/dummy-4	Tensor	(1,)	float32
steps/observation/dummy-5	Tensor	(1,)	float32
steps/observation/dummy-6	Tensor	(1,)	float32
steps/observation/dummy-7	Tensor	(1,)	float32
steps/observation/dummy-8	Tensor	(1,)	float32
steps/observation/dummy-9	Tensor	(1,)	float32
steps/observation/extremities	Tensor	(12,)	float32
steps/observation/head_height	Tensor	(1,)	float32
steps/observation/joint_angles	Tensor	(21,)	float32
steps/observation/torso_vertical	Tensor	(3,)	float32
steps/observation/velocity	Tensor	(27,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/humanoid_walk_combined_challenge_easy_20_percent

• Dataset size: 1.51 GiB

• Auto-cached (documentation): No

• Splits:

Split	Examples
'train'	4,000

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(21,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'com_velocity': Tensor(shape=(3,), dtype=float32),
            'dummy-0': Tensor(shape=(1,), dtype=float32),
            'dummy-1': Tensor(shape=(1,), dtype=float32),
            'dummy-2': Tensor(shape=(1,), dtype=float32),
            'dummy-3': Tensor(shape=(1,), dtype=float32),
            'dummy-4': Tensor(shape=(1,), dtype=float32),
            'dummy-5': Tensor(shape=(1,), dtype=float32),
            'dummy-6': Tensor(shape=(1,), dtype=float32),
            'dummy-7': Tensor(shape=(1,), dtype=float32),
            'dummy-8': Tensor(shape=(1,), dtype=float32),
            'dummy-9': Tensor(shape=(1,), dtype=float32),
            'extremities': Tensor(shape=(12,), dtype=float32),
            'head_height': Tensor(shape=(1,), dtype=float32),
            'joint_angles': Tensor(shape=(21,), dtype=float32),
            'torso_vertical': Tensor(shape=(3,), dtype=float32),
            'velocity': Tensor(shape=(27,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(21,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/com_velocity	Tensor	(3,)	float32
steps/observation/dummy-0	Tensor	(1,)	float32
steps/observation/dummy-1	Tensor	(1,)	float32
steps/observation/dummy-2	Tensor	(1,)	float32
steps/observation/dummy-3	Tensor	(1,)	float32
steps/observation/dummy-4	Tensor	(1,)	float32
steps/observation/dummy-5	Tensor	(1,)	float32
steps/observation/dummy-6	Tensor	(1,)	float32
steps/observation/dummy-7	Tensor	(1,)	float32
steps/observation/dummy-8	Tensor	(1,)	float32
steps/observation/dummy-9	Tensor	(1,)	float32
steps/observation/extremities	Tensor	(12,)	float32
steps/observation/head_height	Tensor	(1,)	float32
steps/observation/joint_angles	Tensor	(21,)	float32
steps/observation/torso_vertical	Tensor	(3,)	float32
steps/observation/velocity	Tensor	(27,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/humanoid_walk_combined_challenge_easy_40_percent

• Dataset size: 3.01 GiB

• Auto-cached (documentation): No

• Splits:

Split	Examples
'train'	8,000

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(21,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'com_velocity': Tensor(shape=(3,), dtype=float32),
            'dummy-0': Tensor(shape=(1,), dtype=float32),
            'dummy-1': Tensor(shape=(1,), dtype=float32),
            'dummy-2': Tensor(shape=(1,), dtype=float32),
            'dummy-3': Tensor(shape=(1,), dtype=float32),
            'dummy-4': Tensor(shape=(1,), dtype=float32),
            'dummy-5': Tensor(shape=(1,), dtype=float32),
            'dummy-6': Tensor(shape=(1,), dtype=float32),
            'dummy-7': Tensor(shape=(1,), dtype=float32),
            'dummy-8': Tensor(shape=(1,), dtype=float32),
            'dummy-9': Tensor(shape=(1,), dtype=float32),
            'extremities': Tensor(shape=(12,), dtype=float32),
            'head_height': Tensor(shape=(1,), dtype=float32),
            'joint_angles': Tensor(shape=(21,), dtype=float32),
            'torso_vertical': Tensor(shape=(3,), dtype=float32),
            'velocity': Tensor(shape=(27,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(21,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/com_velocity	Tensor	(3,)	float32
steps/observation/dummy-0	Tensor	(1,)	float32
steps/observation/dummy-1	Tensor	(1,)	float32
steps/observation/dummy-2	Tensor	(1,)	float32
steps/observation/dummy-3	Tensor	(1,)	float32
steps/observation/dummy-4	Tensor	(1,)	float32
steps/observation/dummy-5	Tensor	(1,)	float32
steps/observation/dummy-6	Tensor	(1,)	float32
steps/observation/dummy-7	Tensor	(1,)	float32
steps/observation/dummy-8	Tensor	(1,)	float32
steps/observation/dummy-9	Tensor	(1,)	float32
steps/observation/extremities	Tensor	(12,)	float32
steps/observation/head_height	Tensor	(1,)	float32
steps/observation/joint_angles	Tensor	(21,)	float32
steps/observation/torso_vertical	Tensor	(3,)	float32
steps/observation/velocity	Tensor	(27,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):

rlu_rwrl/humanoid_walk_combined_challenge_easy_100_percent

• Dataset size: 7.53 GiB

• Auto-cached (documentation): No

• Splits:

Split	Examples
'train'	20,000

• Feature structure:

FeaturesDict({
    'episode_return': float32,
    'steps': Dataset({
        'action': Tensor(shape=(21,), dtype=float32),
        'discount': Tensor(shape=(1,), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'observation': FeaturesDict({
            'com_velocity': Tensor(shape=(3,), dtype=float32),
            'dummy-0': Tensor(shape=(1,), dtype=float32),
            'dummy-1': Tensor(shape=(1,), dtype=float32),
            'dummy-2': Tensor(shape=(1,), dtype=float32),
            'dummy-3': Tensor(shape=(1,), dtype=float32),
            'dummy-4': Tensor(shape=(1,), dtype=float32),
            'dummy-5': Tensor(shape=(1,), dtype=float32),
            'dummy-6': Tensor(shape=(1,), dtype=float32),
            'dummy-7': Tensor(shape=(1,), dtype=float32),
            'dummy-8': Tensor(shape=(1,), dtype=float32),
            'dummy-9': Tensor(shape=(1,), dtype=float32),
            'extremities': Tensor(shape=(12,), dtype=float32),
            'head_height': Tensor(shape=(1,), dtype=float32),
            'joint_angles': Tensor(shape=(21,), dtype=float32),
            'torso_vertical': Tensor(shape=(3,), dtype=float32),
            'velocity': Tensor(shape=(27,), dtype=float32),
        }),
        'reward': Tensor(shape=(1,), dtype=float32),
    }),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
episode_return	Tensor		float32
steps	Dataset
steps/action	Tensor	(21,)	float32
steps/discount	Tensor	(1,)	float32
steps/is_first	Tensor		bool
steps/is_last	Tensor		bool
steps/is_terminal	Tensor		bool
steps/observation	FeaturesDict
steps/observation/com_velocity	Tensor	(3,)	float32
steps/observation/dummy-0	Tensor	(1,)	float32
steps/observation/dummy-1	Tensor	(1,)	float32
steps/observation/dummy-2	Tensor	(1,)	float32
steps/observation/dummy-3	Tensor	(1,)	float32
steps/observation/dummy-4	Tensor	(1,)	float32
steps/observation/dummy-5	Tensor	(1,)	float32
steps/observation/dummy-6	Tensor	(1,)	float32
steps/observation/dummy-7	Tensor	(1,)	float32
steps/observation/dummy-8	Tensor	(1,)	float32
steps/observation/dummy-9	Tensor	(1,)	float32
steps/observation/extremities	Tensor	(12,)	float32
steps/observation/head_height	Tensor	(1,)	float32
steps/observation/joint_angles	Tensor	(21,)	float32
steps/observation/torso_vertical	Tensor	(3,)	float32
steps/observation/velocity	Tensor	(27,)	float32
steps/reward	Tensor	(1,)	float32

• Examples (tfds.as_dataframe):