src.plangym.utils

src.plangym.utils#

Generic utilities for working with environments.

Attributes#

USE_PIL

Classes#

GrayScaleObservation

Convert the image observation from RGB to gray scale.

Functions#

get_display([visible, size])

Start a virtual display.

remove_time_limit_from_spec(spec)

Remove the maximum time limit of an environment spec.

remove_time_limit(gym_env)

Remove the maximum time limit of the provided environment.

process_frame_pil(frame[, width, height, mode])

Resize an RGB frame to a specified shape and mode.

process_frame_opencv(frame[, width, height, mode])

Resize an RGB frame to a specified shape and mode.

process_frame(frame[, width, height, mode])

Resize an RGB frame to a specified shape and mode.

Module Contents#

src.plangym.utils.USE_PIL = True#
src.plangym.utils.get_display(visible=False, size=(400, 400), **kwargs)[source]#

Start a virtual display.

src.plangym.utils.remove_time_limit_from_spec(spec)[source]#

Remove the maximum time limit of an environment spec.

src.plangym.utils.remove_time_limit(gym_env)[source]#

Remove the maximum time limit of the provided environment.

Parameters:

gym_env (gymnasium.Env)

Return type:

gymnasium.Env

src.plangym.utils.process_frame_pil(frame, width=None, height=None, mode='RGB')[source]#

Resize an RGB frame to a specified shape and mode.

Use PIL to resize an RGB frame to a specified height and width or changing it to a different mode.

Parameters:

  • frame (numpy.ndarray) – Target numpy array representing the image that will be resized.

  • width (int | None) – Width of the resized image.

  • height (int | None) – Height of the resized image.

  • mode (str) – Passed to Image.convert.

Returns:

The resized frame that matches the provided width and height.

Return type:

numpy.ndarray

src.plangym.utils.process_frame_opencv(frame, width=None, height=None, mode='RGB')[source]#

Resize an RGB frame to a specified shape and mode.

Use OpenCV to resize an RGB frame to a specified height and width or changing it to a different mode.

Parameters:

  • frame (numpy.ndarray) – Target numpy array representing the image that will be resized.

  • width (int | None) – Width of the resized image.

  • height (int | None) – Height of the resized image.

  • mode (str) – Passed to cv2.cvtColor.

Returns:

The resized frame that matches the provided width and height.

Return type:

numpy.ndarray

src.plangym.utils.process_frame(frame, width=None, height=None, mode='RGB')[source]#

Resize an RGB frame to a specified shape and mode.

Use either PIL or OpenCV to resize an RGB frame to a specified height and width or changing it to a different mode.

Parameters:

  • frame (numpy.ndarray) – Target numpy array representing the image that will be resized.

  • width (int | None) – Width of the resized image.

  • height (int | None) – Height of the resized image.

  • mode (str) – Passed to either Image.convert or cv2.cvtColor.

Returns:

The resized frame that matches the provided width and height.

Return type:

numpy.ndarray

class src.plangym.utils.GrayScaleObservation(env, keep_dim=False)[source]#

Bases: gymnasium.ObservationWrapper, gymnasium.utils.RecordConstructorArgs

Convert the image observation from RGB to gray scale.

Example

>>> import gymnasium as gym
>>> from gymnasium.wrappers import GrayScaleObservation
>>> env = gym.make("CarRacing-v2")
>>> env.observation_space
Box(0, 255, (96, 96, 3), uint8)
>>> env = GrayScaleObservation(gym.make("CarRacing-v2"))
>>> env.observation_space
Box(0, 255, (96, 96), uint8)
>>> env = GrayScaleObservation(gym.make("CarRacing-v2"), keep_dim=True)
>>> env.observation_space
Box(0, 255, (96, 96, 1), uint8)
Parameters:

  • env (gymnasium.Env)

  • keep_dim (bool)

keep_dim#
obs_shape#
observation(observation)[source]#

Convert the colour observation to greyscale.

Parameters:

observation – Color observations

Returns:

Grayscale observations

Contents