NavToObjRewardMeasurementConfig class
Rearrangement Navigation task only. The reward for rearrangement navigation.
Special methods
- def __init__(self, type: str = 'NavToObjReward', should_reward_turn: bool = True, turn_reward_dist: float = 3.0, angle_dist_reward: float = 1.0, dist_reward: float = 1.0, constraint_violate_pen: float = 1.0, force_pen: float = 0.0001, max_force_pen: float = 0.01, force_end_pen: float = 1.0, count_coll_pen: float = -1.0, max_count_colls: int = -1, count_coll_end_pen: float = 1.0) -> None
- def __repr__(self)
Data
- angle_dist_reward: float = 1.0
- constraint_violate_pen: float = 1.0
- count_coll_end_pen: float = 1.0
- count_coll_pen: float = -1.0
- dist_reward: float = 1.0
- force_end_pen: float = 1.0
- force_pen: float = 0.0001
- max_count_colls: int = -1
- max_force_pen: float = 0.01
- should_reward_turn: bool = True
- turn_reward_dist: float = 3.0
- type: str = 'NavToObjReward'