MeanZeroStdOneObservationNormalizationStrategy¶
-
class
maze.core.wrappers.observation_normalization.normalization_strategies.mean_zero_std_one.
MeanZeroStdOneObservationNormalizationStrategy
(observation_space: gym.spaces.Box, clip_range: Tuple[Union[float, int], Union[float, int]], axis: Optional[Union[int, Tuple[int], List[int]]])¶ Normalizes observations to have zero mean and standard deviation one.
The strategy first subtracts the observation mean followed by a division with the standard deviation. Depending on the original distribution of the input observations this yields a standard Normal.
-
estimate_stats
(observations: List[numpy.ndarray]) → Dict[str, Union[numpy.ndarray, float, int, Iterable[Union[float, int]]]]¶ (overrides
ObservationNormalizationStrategy
)- Implementation of
ObservationNormalizationStrategy
interface.
-
normalize_value
(value: numpy.ndarray) → numpy.ndarray¶ (overrides
ObservationNormalizationStrategy
)- Implementation of
ObservationNormalizationStrategy
interface.
-