rl4lms.envs.common package

Submodules

rl4lms.envs.common.action_space module

class rl4lms.envs.common.action_space.ActionSpace(actions: List[str])[source]

Bases: Discrete

__init__(actions: List[str])[source]

action_to_ix(action: str) → int[source]

ix_to_action(ix: int) → str[source]

size() → int[source]

rl4lms.envs.common.base_env module

rl4lms.envs.common.observation module

class rl4lms.envs.common.observation.BaseObservation[source]

Bases: object

Placeholder for observation data class

__init__() → None

class rl4lms.envs.common.observation.BaseObservationFeaturizer[source]

Bases: ABC

abstract featurize(observation: BaseObservation) → <MagicMock id='139934540113632'>[source]

get_observation_dim() → int[source]: Returns the observation dim

rl4lms.envs.common.reward module

class rl4lms.envs.common.reward.RewardFunction[source]

Bases: ABC

abstract classmethod __call__(observation: BaseObservation, action: str, targets: List[str]) → float[source]

[summary]

Parameters:

observation (Observation) – current observation at t
action (str) – current action at t
targets (List[str]) – targets of the current sample

Returns:

a scalar reward

Module contents