base

Source code: armscan_env/envs/base.py

---

class ArrayFromDictObservation(dict_observation: DictObservation[TStateAction])

compute_observation(state: TStateAction) → ndarray

property observation_space: Box

class ArrayObservation

abstract property observation_space: Box

class ConcatenatedArrayObservation(array_observations: list[ArrayObservation[TStateAction]])

compute_observation(state: TStateAction) → ndarray

static concatenate_boxes(boxes: list[Box]) → Box

property observation_space: Box

class DictObservation

merged_with(other: Self) → MergedDictObservation[TStateAction]

abstract property observation_space: Dict

to_array_observation() → ArrayObservation[TStateAction]

class DummyArrayObservation

compute_observation(state: Any) → ndarray

property observation_space: Space[TObs]

class DummyDictObservation

compute_observation(state: Any) → dict[str, ndarray]

property observation_space: Dict

exception EnvPreconditionError

class EnvRollout(*, observations: list[~TObs] = <factory>, rewards: list[float] = <factory>, actions: list[typing.Optional[~TAction]] = <factory>, infos: list[dict[str, typing.Any]] = <factory>, terminated: list[bool] = <factory>, truncated: list[bool] = <factory>)

actions: list[TAction | None]

append_reset(observation: TObs, info: dict[str, Any], reward: float = 0, terminated: bool = False, truncated: bool = False) → None

append_step(action: TAction, observation: TObs, reward: float, terminated: bool, truncated: bool, info: dict[str, Any]) → None

infos: list[dict[str, Any]]

observations: list[TObs]

rewards: list[float]

terminated: list[bool]

truncated: list[bool]

class EnvStatus(*, episode_len: int, state_action: TStateAction | None, observation: TObs | None, reward: float | None, is_terminated: bool, is_truncated: bool, is_closed: bool, info: dict[str, Any])

episode_len: int

info: dict[str, Any]

is_closed: bool

is_terminated: bool

is_truncated: bool

observation: TObs | None

reward: float | None

state_action: TStateAction | None

class MergedDictObservation(dict_observations: list[DictObservation[TStateAction]])

compute_observation(state: TStateAction) → dict[str, ndarray]

property dict_observations: list[DictObservation[TStateAction]]

property observation_space: Dict

class ModularEnv(reward_metric: RewardMetric[TStateAction], observation: Observation[TStateAction, TObs], termination_criterion: TerminationCriterion | None = None, max_episode_len: int | None = None)

abstract property action_space: Space[TAction]

close() → None

After the user has finished using the environment, close contains the code necessary to "clean up" the environment.

This is critical for closing rendering windows, database or HTTP connections. Calling close on an already closed environment has no effect and won't raise an error.

compute_cur_observation() → TObs

compute_cur_reward() → float

abstract compute_next_state(action: TAction) → TStateAction

property cur_episode_len: int

property cur_observation: TObs | None

property cur_reward: float | None

property cur_state_action: TStateAction | None

get_cur_env_status() → EnvStatus[TStateAction, TObs]

get_info_dict() → dict[str, Any]

property is_closed: bool

property is_terminated: bool

property is_truncated: bool

property observation_space: Space[TObs]

reset(seed: int | None = None, **kwargs: Any) → tuple[TObs, dict[str, Any]]

Resets the environment to an initial internal state, returning an initial observation and info.

This method generates a new starting state often with some randomness to ensure that the agent explores the state space and learns a generalised policy about the environment. This randomness can be controlled with the seed parameter otherwise if the environment already has a random number generator and reset() is called with seed=None, the RNG is not reset.

Therefore, reset() should (in the typical use case) be called with a seed right after initialization and then never again.

For Custom environments, the first line of reset() should be super().reset(seed=seed) which implements the seeding correctly.

Changed in version v0.25: The return_info parameter was removed and now info is expected to be returned.

Args:

seed (optional int): The seed that is used to initialize the environment's PRNG (np_random) and

the read-only attribute np_random_seed. If the environment does not already have a PRNG and seed=None (the default option) is passed, a seed will be chosen from some source of entropy (e.g. timestamp or /dev/urandom). However, if the environment already has a PRNG and seed=None is passed, the PRNG will not be reset and the env's np_random_seed will not be altered. If you pass an integer, the PRNG will be reset even if it already exists. Usually, you want to pass an integer right after the environment has been initialized and then never again. Please refer to the minimal example above to see this paradigm in action.

options (optional dict): Additional information to specify how the environment is reset (optional,

depending on the specific environment)

Returns:

observation (ObsType): Observation of the initial state. This will be an element of observation_space

(typically a numpy array) and is analogous to the observation returned by step().

info (dictionary): This dictionary contains auxiliary information complementing observation. It should be analogous to

the info returned by step().

abstract sample_initial_state() → TStateAction

should_terminate() → bool

should_truncate() → bool

step(action: TAction) → tuple[TObs, float, bool, bool, dict[str, Any]]

Step through the environment to navigate to the next state.

class NeverTerminate

should_terminate(env: Any) → bool

class Observation

abstract compute_observation(state: TStateAction) → TObs

abstract property observation_space: Space[TObs]

class RewardMetric

abstract compute_reward(state: TStateAction) → float

abstract property range: tuple[float, float]

class StateAction(*, normalized_action_arr: Any)

normalized_action_arr: Any

class TerminationCriterion

abstract should_terminate(env: TEnv) → bool