List of Wrappers

Gymnasium provides a number of commonly used wrappers listed below. More information can be found on the particular wrapper in the page on the wrapper type

Name	Description
AtariPreprocessing	Implements the common preprocessing techniques for Atari environments (excluding frame stacking).
Autoreset	The wrapped environment is automatically reset when an terminated or truncated state is reached.
ClipAction	Clips the action pass to step to be within the environment’s action_space.
ClipReward	Clips the rewards for an environment between an upper and lower bound.
DelayObservation	Adds a delay to the returned observation from the environment.
DtypeObservation	Modifies the dtype of an observation array to a specified dtype.
FilterObservation	Filters a Dict or Tuple observation spaces by a set of keys or indexes.
FlattenObservation	Flattens the environment’s observation space and each observation from reset and step functions.
FrameStackObservation	Stacks the observations from the last N time steps in a rolling manner.
GrayscaleObservation	Converts an image observation computed by reset and step from RGB to Grayscale.
HumanRendering	Allows human like rendering for environments that support “rgb_array” rendering.
JaxToNumpy	Wraps a Jax-based environment such that it can be interacted with NumPy arrays.
JaxToTorch	Wraps a Jax-based environment so that it can be interacted with PyTorch Tensors.
MaxAndSkipObservation	Skips the N-th frame (observation) and return the max values between the two last observations.
NormalizeObservation	Normalizes observations to be centered at the mean with unit variance.
NormalizeReward	Normalizes immediate rewards such that their exponential moving average has a fixed variance.
NumpyToTorch	Wraps a NumPy-based environment such that it can be interacted with PyTorch Tensors.
OrderEnforcing	Will produce an error if step or render is called before render.
PassiveEnvChecker	A passive environment checker wrapper that surrounds the step, reset and render functions to check they follows gymnasium’s API.
RecordEpisodeStatistics	This wrapper will keep track of cumulative rewards and episode lengths.
RecordVideo	Records videos of environment episodes using the environment’s render function.
RenderCollection	Collect rendered frames of an environment such render returns a list[RenderedFrame].
AddRenderObservation	Includes the rendered observations in the environment’s observations.
RescaleAction	Affinely (linearly) rescales a Box action space of the environment to within the range of [min_action, max_action].
RescaleObservation	Affinely (linearly) rescales a Box observation space of the environment to within the range of [min_obs, max_obs].
ReshapeObservation	Reshapes Array based observations to a specified shape.
ResizeObservation	Resizes image observations using OpenCV to a specified shape.
StickyAction	Adds a probability that the action is repeated for the same step function.
TimeAwareObservation	Augment the observation with the number of time steps taken within an episode.
TimeLimit	Limits the number of steps for an environment through truncating the environment if a maximum number of timesteps is exceeded.
TransformAction	Applies a function to the action before passing the modified value to the environment step function.
TransformObservation	Applies a function to the observation received from the environment’s reset and step that is passed back to the user.
TransformReward	Applies a function to the reward received from the environment’s step.

Vector only Wrappers

Name	Description
DictInfoToList	Converts infos of vectorized environments from dict to List[dict].
VectorizeTransformAction	Vectorizes a single-agent transform action wrapper for vector environments.
VectorizeTransformObservation	Vectorizes a single-agent transform observation wrapper for vector environments.
VectorizeTransformReward	Vectorizes a single-agent transform reward wrapper for vector environments.