sinergym.utils.wrappers.NormalizeObservation

class sinergym.utils.wrappers.NormalizeObservation(env: Env, automatic_update: bool = True, epsilon: float = 1e-08, mean: list | float64 | str = None, var: list | float64 | str = None)

__init__(env: Env, automatic_update: bool = True, epsilon: float = 1e-08, mean: list | float64 | str = None, var: list | float64 | str = None)

Initializes the NormalizationWrapper. Mean and var values can be None and being updated during interaction with environment.

Parameters:

env (Env) – The environment to apply the wrapper.
automatic_update (bool, optional) – Whether or not to update the mean and variance values automatically. Defaults to True.
epsilon (float, optional) – A stability parameter used when scaling the observations. Defaults to 1e-8.
mean (list, np.float64, str, optional) – The mean value used for normalization. It can be a mean.txt path too. Defaults to None.
var (list, np.float64, str, optional) – The variance value used for normalization. It can be a var.txt path too. Defaults to None.

Methods

`__init__`(env[, automatic_update, epsilon, ...])	Initializes the NormalizationWrapper.
`activate_update`()	Activates the automatic update of the normalization wrapper.
`class_name`()	Returns the class name of the wrapper.
`close`()	save normalization calibration and close the environment.
`deactivate_update`()	Deactivates the automatic update of the normalization wrapper.
`get_wrapper_attr`(name)	Gets an attribute from the wrapper and lower environments if name doesn't exist in this object.
`normalize`(obs)	Normalizes the observation using the running mean and variance of the observations.
`render`()	Uses the `render()` of the `env` that can be overwritten to change the returned data.
`reset`(**kwargs)	Resets the environment and normalizes the observation.
`set_mean`(mean)	Sets the mean value of the observations.
`set_var`(var)	Sets the variance value of the observations.
`step`(action)	Steps through the environment and normalizes the observation.
`wrapper_spec`(**kwargs)	Generates a WrapperSpec for the wrappers.

Attributes

`action_space`	Return the `Env` `action_space` unless overwritten then the wrapper `action_space` is used.
`logger`
`mean`	Returns the mean value of the observations.
`metadata`	Returns the `Env` `metadata`.
`np_random`	Returns the `Env` `np_random` attribute.
`observation_space`	Return the `Env` `observation_space` unless overwritten then the wrapper `observation_space` is used.
`render_mode`	Returns the `Env` `render_mode`.
`reward_range`	Return the `Env` `reward_range` unless overwritten then the wrapper `reward_range` is used.
`spec`	Returns the `Env` `spec` attribute with the WrapperSpec if the wrapper inherits from EzPickle.
`unwrapped`	Returns the base environment of the wrapper.
`var`	Returns the variance value of the observations.

activate_update(): Activates the automatic update of the normalization wrapper. After calling this method, the normalization wrapper will update its calibration automatically.

close(): save normalization calibration and close the environment.

deactivate_update(): Deactivates the automatic update of the normalization wrapper. After calling this method, the normalization wrapper will not update its calibration automatically.

logger = <Logger WRAPPER NormalizeObservation (INFO)>

property mean: float64: Returns the mean value of the observations.

normalize(obs): Normalizes the observation using the running mean and variance of the observations. If automatic_update is enabled, the running mean and variance will be updated too.

reset(**kwargs): Resets the environment and normalizes the observation.

set_mean(mean: list | float64 | str): Sets the mean value of the observations.

set_var(var: list | float64 | str): Sets the variance value of the observations.

step(action: int | ndarray) → Tuple[ndarray, float, bool, bool, Dict[str, Any]]: Steps through the environment and normalizes the observation.

property var: float64: Returns the variance value of the observations.