mimic_play

  • Description:

Real dataset of 14 long horizon manipulation tasks. A mix of human play data and single robot arm data performing the same tasks.

Split Examples
'train' 378
  • Feature structure:
FeaturesDict({
    'episode_metadata': FeaturesDict({
        'file_path': string,
    }),
    'steps': Dataset({
        'action': Tensor(shape=(7,), dtype=float32),
        'discount': Scalar(shape=(), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'language_embedding': Tensor(shape=(512,), dtype=float32),
        'language_instruction': string,
        'observation': FeaturesDict({
            'image': FeaturesDict({
                'front_image_1': Image(shape=(120, 120, 3), dtype=uint8),
                'front_image_2': Image(shape=(120, 120, 3), dtype=uint8),
            }),
            'state': FeaturesDict({
                'ee_pose': Tensor(shape=(7,), dtype=float32),
                'gripper_position': float32,
                'joint_positions': Tensor(shape=(7,), dtype=float32),
                'joint_velocities': Tensor(shape=(7,), dtype=float32),
            }),
            'wrist_image': FeaturesDict({
                'wrist_image': Image(shape=(120, 120, 3), dtype=uint8),
            }),
        }),
        'reward': Scalar(shape=(), dtype=float32),
    }),
})
  • Feature documentation:
Feature Class Shape Dtype Description
FeaturesDict
episode_metadata FeaturesDict
episode_metadata/file_path Tensor string
steps Dataset
steps/action Tensor (7,) float32
steps/discount Scalar float32
steps/is_first Tensor bool
steps/is_last Tensor bool
steps/is_terminal Tensor bool
steps/language_embedding Tensor (512,) float32
steps/language_instruction Tensor string
steps/observation FeaturesDict
steps/observation/image FeaturesDict
steps/observation/image/front_image_1 Image (120, 120, 3) uint8
steps/observation/image/front_image_2 Image (120, 120, 3) uint8
steps/observation/state FeaturesDict
steps/observation/state/ee_pose Tensor (7,) float32
steps/observation/state/gripper_position Tensor float32
steps/observation/state/joint_positions Tensor (7,) float32
steps/observation/state/joint_velocities Tensor (7,) float32
steps/observation/wrist_image FeaturesDict
steps/observation/wrist_image/wrist_image Image (120, 120, 3) uint8
steps/reward Scalar float32
  • Citation:
@article{wang2023mimicplay,title={Mimicplay: Long-horizon imitation learning by watching human play},author={Wang, Chen and Fan, Linxi and Sun, Jiankai and Zhang, Ruohan and Fei-Fei, Li and Xu, Danfei and Zhu, Yuke and Anandkumar, Anima},journal={arXiv preprint arXiv:2302.12422},year={2023} }