Have a question? Connect with the community at the TensorFlow Forum Visit Forum

d4rl_mujoco_hopper

  • Description:

D4RL is an open-source benchmark for offline reinforcement learning. It provides standardized environments and datasets for training and benchmarking algorithms.

@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 (default config)

  • Download size: 51.56 MiB

  • Dataset size: 63.12 MiB

  • Auto-cached (documentation): Yes

  • Splits:

Split Examples
'train' 1,029
  • Features:
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=tf.float32),
        'discount': tf.float32,
        'is_first': tf.bool,
        'is_terminal': tf.bool,
        'observation': Tensor(shape=(11,), dtype=tf.float32),
        'reward': tf.float32,
    }),
})

d4rl_mujoco_hopper/v0-medium

  • Download size: 51.74 MiB

  • Dataset size: 63.64 MiB

  • Auto-cached (documentation): Yes

  • Splits:

Split Examples
'train' 3,064
  • Features:
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=tf.float32),
        'discount': tf.float32,
        'is_first': tf.bool,
        'is_terminal': tf.bool,
        'observation': Tensor(shape=(11,), dtype=tf.float32),
        'reward': tf.float32,
    }),
})

d4rl_mujoco_hopper/v0-medium-expert

  • Download size: 62.01 MiB

  • Dataset size: 76.05 MiB

  • Auto-cached (documentation): Yes

  • Splits:

Split Examples
'train' 2,277
  • Features:
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=tf.float32),
        'discount': tf.float32,
        'is_first': tf.bool,
        'is_terminal': tf.bool,
        'observation': Tensor(shape=(11,), dtype=tf.float32),
        'reward': tf.float32,
    }),
})

d4rl_mujoco_hopper/v0-mixed

  • Download size: 10.48 MiB

  • Dataset size: 12.93 MiB

  • Auto-cached (documentation): Yes

  • Splits:

Split Examples
'train' 1,250
  • Features:
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=tf.float32),
        'discount': tf.float32,
        'is_first': tf.bool,
        'is_terminal': tf.bool,
        'observation': Tensor(shape=(11,), dtype=tf.float32),
        'reward': tf.float32,
    }),
})

d4rl_mujoco_hopper/v0-random

  • Download size: 51.83 MiB

  • Dataset size: 64.90 MiB

  • Auto-cached (documentation): Yes

  • Splits:

Split Examples
'train' 8,793
  • Features:
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=tf.float32),
        'discount': tf.float32,
        'is_first': tf.bool,
        'is_terminal': tf.bool,
        'observation': Tensor(shape=(11,), dtype=tf.float32),
        'reward': tf.float32,
    }),
})

d4rl_mujoco_hopper/v1-expert

  • Download size: 93.19 MiB

  • Dataset size: 607.03 MiB

  • Auto-cached (documentation): No

  • Splits:

Split Examples
'train' 1,836
  • Features:
FeaturesDict({
    'algorithm': tf.string,
    'iteration': tf.int32,
    'policy': FeaturesDict({
        'fc0': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=tf.float32),
            'weight': Tensor(shape=(256, 11), dtype=tf.float32),
        }),
        'fc1': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=tf.float32),
            'weight': Tensor(shape=(256, 256), dtype=tf.float32),
        }),
        'last_fc': FeaturesDict({
            'bias': Tensor(shape=(3,), dtype=tf.float32),
            'weight': Tensor(shape=(3, 256), dtype=tf.float32),
        }),
        'last_fc_log_std': FeaturesDict({
            'bias': Tensor(shape=(3,), dtype=tf.float32),
            'weight': Tensor(shape=(3, 256), dtype=tf.float32),
        }),
        'nonlinearity': tf.string,
        'output_distribution': tf.string,
    }),
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=tf.float32),
        'discount': tf.float32,
        'infos': FeaturesDict({
            'action_log_probs': tf.float32,
            'qpos': Tensor(shape=(6,), dtype=tf.float32),
            'qvel': Tensor(shape=(6,), dtype=tf.float32),
        }),
        'is_first': tf.bool,
        'is_terminal': tf.bool,
        'observation': Tensor(shape=(11,), dtype=tf.float32),
        'reward': tf.float32,
    }),
})

d4rl_mujoco_hopper/v1-medium

  • Download size: 92.03 MiB

  • Dataset size: 1.77 GiB

  • Auto-cached (documentation): No

  • Splits:

Split Examples
'train' 6,328
  • Features:
FeaturesDict({
    'algorithm': tf.string,
    'iteration': tf.int32,
    'policy': FeaturesDict({
        'fc0': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=tf.float32),
            'weight': Tensor(shape=(256, 11), dtype=tf.float32),
        }),
        'fc1': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=tf.float32),
            'weight': Tensor(shape=(256, 256), dtype=tf.float32),
        }),
        'last_fc': FeaturesDict({
            'bias': Tensor(shape=(3,), dtype=tf.float32),
            'weight': Tensor(shape=(3, 256), dtype=tf.float32),
        }),
        'last_fc_log_std': FeaturesDict({
            'bias': Tensor(shape=(3,), dtype=tf.float32),
            'weight': Tensor(shape=(3, 256), dtype=tf.float32),
        }),
        'nonlinearity': tf.string,
        'output_distribution': tf.string,
    }),
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=tf.float32),
        'discount': tf.float32,
        'infos': FeaturesDict({
            'action_log_probs': tf.float32,
            'qpos': Tensor(shape=(6,), dtype=tf.float32),
            'qvel': Tensor(shape=(6,), dtype=tf.float32),
        }),
        'is_first': tf.bool,
        'is_terminal': tf.bool,
        'observation': Tensor(shape=(11,), dtype=tf.float32),
        'reward': tf.float32,
    }),
})

d4rl_mujoco_hopper/v1-medium-expert

  • Download size: 184.59 MiB

  • Dataset size: 228.11 MiB

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

  • Splits:

Split Examples
'train' 8,163
  • Features:
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=tf.float32),
        'discount': tf.float32,
        'infos': FeaturesDict({
            'action_log_probs': tf.float32,
            'qpos': Tensor(shape=(6,), dtype=tf.float32),
            'qvel': Tensor(shape=(6,), dtype=tf.float32),
        }),
        'is_first': tf.bool,
        'is_terminal': tf.bool,
        'observation': Tensor(shape=(11,), dtype=tf.float32),
        'reward': tf.float32,
    }),
})

d4rl_mujoco_hopper/v1-medium-replay

  • Download size: 55.65 MiB

  • Dataset size: 34.43 MiB

  • Auto-cached (documentation): Yes

  • Splits:

Split Examples
'train' 1,151
  • Features:
FeaturesDict({
    'algorithm': tf.string,
    'iteration': tf.int32,
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=tf.float64),
        'discount': tf.float64,
        'infos': FeaturesDict({
            'action_log_probs': tf.float64,
            'qpos': Tensor(shape=(6,), dtype=tf.float64),
            'qvel': Tensor(shape=(6,), dtype=tf.float64),
        }),
        'is_first': tf.bool,
        'is_terminal': tf.bool,
        'observation': Tensor(shape=(11,), dtype=tf.float64),
        'reward': tf.float64,
    }),
})

d4rl_mujoco_hopper/v1-full-replay

  • Download size: 183.32 MiB

  • Dataset size: 113.63 MiB

  • Auto-cached (documentation): Yes

  • Splits:

Split Examples
'train' 2,907
  • Features:
FeaturesDict({
    'algorithm': tf.string,
    'iteration': tf.int32,
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=tf.float64),
        'discount': tf.float64,
        'infos': FeaturesDict({
            'action_log_probs': tf.float64,
            'qpos': Tensor(shape=(6,), dtype=tf.float64),
            'qvel': Tensor(shape=(6,), dtype=tf.float64),
        }),
        'is_first': tf.bool,
        'is_terminal': tf.bool,
        'observation': Tensor(shape=(11,), dtype=tf.float64),
        'reward': tf.float64,
    }),
})

d4rl_mujoco_hopper/v1-random

  • Download size: 91.11 MiB

  • Dataset size: 128.74 MiB

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

  • Splits:

Split Examples
'train' 45,265
  • Features:
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=tf.float32),
        'discount': tf.float32,
        'infos': FeaturesDict({
            'action_log_probs': tf.float32,
            'qpos': Tensor(shape=(6,), dtype=tf.float32),
            'qvel': Tensor(shape=(6,), dtype=tf.float32),
        }),
        'is_first': tf.bool,
        'is_terminal': tf.bool,
        'observation': Tensor(shape=(11,), dtype=tf.float32),
        'reward': tf.float32,
    }),
})

d4rl_mujoco_hopper/v2-expert

  • Download size: 134.46 MiB

  • Dataset size: 389.31 MiB

  • Auto-cached (documentation): No

  • Splits:

Split Examples
'train' 1,028
  • Features:
FeaturesDict({
    'algorithm': tf.string,
    'iteration': tf.int32,
    'policy': FeaturesDict({
        'fc0': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=tf.float32),
            'weight': Tensor(shape=(256, 11), dtype=tf.float32),
        }),
        'fc1': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=tf.float32),
            'weight': Tensor(shape=(256, 256), dtype=tf.float32),
        }),
        'last_fc': FeaturesDict({
            'bias': Tensor(shape=(3,), dtype=tf.float32),
            'weight': Tensor(shape=(3, 256), dtype=tf.float32),
        }),
        'last_fc_log_std': FeaturesDict({
            'bias': Tensor(shape=(3,), dtype=tf.float32),
            'weight': Tensor(shape=(3, 256), dtype=tf.float32),
        }),
        'nonlinearity': tf.string,
        'output_distribution': tf.string,
    }),
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=tf.float32),
        'discount': tf.float32,
        'infos': FeaturesDict({
            'action_log_probs': tf.float32,
            'qpos': Tensor(shape=(6,), dtype=tf.float32),
            'qvel': Tensor(shape=(6,), dtype=tf.float32),
        }),
        'is_first': tf.bool,
        'is_terminal': tf.bool,
        'observation': Tensor(shape=(11,), dtype=tf.float32),
        'reward': tf.float32,
    }),
})

d4rl_mujoco_hopper/v2-full-replay

  • Download size: 182.80 MiB

  • Dataset size: 113.88 MiB

  • Auto-cached (documentation): Yes

  • Splits:

Split Examples
'train' 3,515
  • Features:
FeaturesDict({
    'algorithm': tf.string,
    'iteration': tf.int32,
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=tf.float64),
        'discount': tf.float64,
        'infos': FeaturesDict({
            'action_log_probs': tf.float64,
            'qpos': Tensor(shape=(6,), dtype=tf.float64),
            'qvel': Tensor(shape=(6,), dtype=tf.float64),
        }),
        'is_first': tf.bool,
        'is_terminal': tf.bool,
        'observation': Tensor(shape=(11,), dtype=tf.float64),
        'reward': tf.float64,
    }),
})

d4rl_mujoco_hopper/v2-medium

  • Download size: 134.93 MiB

  • Dataset size: 701.56 MiB

  • Auto-cached (documentation): No

  • Splits:

Split Examples
'train' 2,187
  • Features:
FeaturesDict({
    'algorithm': tf.string,
    'iteration': tf.int32,
    'policy': FeaturesDict({
        'fc0': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=tf.float32),
            'weight': Tensor(shape=(256, 11), dtype=tf.float32),
        }),
        'fc1': FeaturesDict({
            'bias': Tensor(shape=(256,), dtype=tf.float32),
            'weight': Tensor(shape=(256, 256), dtype=tf.float32),
        }),
        'last_fc': FeaturesDict({
            'bias': Tensor(shape=(3,), dtype=tf.float32),
            'weight': Tensor(shape=(3, 256), dtype=tf.float32),
        }),
        'last_fc_log_std': FeaturesDict({
            'bias': Tensor(shape=(3,), dtype=tf.float32),
            'weight': Tensor(shape=(3, 256), dtype=tf.float32),
        }),
        'nonlinearity': tf.string,
        'output_distribution': tf.string,
    }),
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=tf.float32),
        'discount': tf.float32,
        'infos': FeaturesDict({
            'action_log_probs': tf.float32,
            'qpos': Tensor(shape=(6,), dtype=tf.float32),
            'qvel': Tensor(shape=(6,), dtype=tf.float32),
        }),
        'is_first': tf.bool,
        'is_terminal': tf.bool,
        'observation': Tensor(shape=(11,), dtype=tf.float32),
        'reward': tf.float32,
    }),
})

d4rl_mujoco_hopper/v2-medium-expert

  • Download size: 268.78 MiB

  • Dataset size: 226.18 MiB

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

  • Splits:

Split Examples
'train' 3,214
  • Features:
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=tf.float32),
        'discount': tf.float32,
        'infos': FeaturesDict({
            'action_log_probs': tf.float32,
            'qpos': Tensor(shape=(6,), dtype=tf.float32),
            'qvel': Tensor(shape=(6,), dtype=tf.float32),
        }),
        'is_first': tf.bool,
        'is_terminal': tf.bool,
        'observation': Tensor(shape=(11,), dtype=tf.float32),
        'reward': tf.float32,
    }),
})

d4rl_mujoco_hopper/v2-medium-replay

  • Download size: 73.67 MiB

  • Dataset size: 46.03 MiB

  • Auto-cached (documentation): Yes

  • Splits:

Split Examples
'train' 2,041
  • Features:
FeaturesDict({
    'algorithm': tf.string,
    'iteration': tf.int32,
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=tf.float64),
        'discount': tf.float64,
        'infos': FeaturesDict({
            'action_log_probs': tf.float64,
            'qpos': Tensor(shape=(6,), dtype=tf.float64),
            'qvel': Tensor(shape=(6,), dtype=tf.float64),
        }),
        'is_first': tf.bool,
        'is_terminal': tf.bool,
        'observation': Tensor(shape=(11,), dtype=tf.float64),
        'reward': tf.float64,
    }),
})

d4rl_mujoco_hopper/v2-random

  • Download size: 132.99 MiB

  • Dataset size: 128.73 MiB

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

  • Splits:

Split Examples
'train' 45,240
  • Features:
FeaturesDict({
    'steps': Dataset({
        'action': Tensor(shape=(3,), dtype=tf.float32),
        'discount': tf.float32,
        'infos': FeaturesDict({
            'action_log_probs': tf.float32,
            'qpos': Tensor(shape=(6,), dtype=tf.float32),
            'qvel': Tensor(shape=(6,), dtype=tf.float32),
        }),
        'is_first': tf.bool,
        'is_terminal': tf.bool,
        'observation': Tensor(shape=(11,), dtype=tf.float32),
        'reward': tf.float32,
    }),
})