DiscreteReward

class coba.primitives.DiscreteReward

A reward function mapping actions to rewards.

Constructors

__init__(actions: Sequence[Action], rewards: Sequence[float], *, default: float = 0) → None

__init__(mapping: Mapping[Action, float], *, default: float = 0) → None

Instantiate a DiscreteReward.

Parameters:

• actions – The actions to define rewards for.

• rewards – The rewards for the given actions.

• mapping – A mapping of actions to rewards.

• default – The value to return for actions without mappings.

Methods

__call__(action: str | Number | Sequence | Mapping) → float

Get reward for action.

Parameters:

action – An action taken.

Returns:

Reward received for the action.