gym-notices is a Python library that provides important advisories and warnings for the original Gym (OpenAI Gym) reinforcement learning toolkit. These notices may be displayed automatically upon importing Gym on internet-connected systems, informing users about major reproducibility issues, outdated versions, or the library's maintenance status. It is currently at version 0.1.0 and releases are infrequent, primarily focused on issuing critical warnings.
pip install gym-noticesVerified import paths — ran on the pinned version, not inferred.
Simply importing `gym_notices` or `gym` (if `gym_notices` is a dependency of `gym`) triggers the notice system. Warnings are displayed automatically if relevant conditions (like using an unsupported Gym version or Gym's deprecation) are detected and an internet connection is available to fetch the latest advisories.
Migrate your code from `gym` to `gymnasium` by installing `gymnasium` and changing `import gym` to `import gymnasium as gym`.
Understand that the library's primary function is passive notification. Ensure `gym_notices` is installed and imported alongside `gym` (or if `gym` depends on it) to receive advisories.
Ensure internet connectivity when running code that imports `gym` or `gym_notices` to receive timely warnings. Regularly check the `Farama-Foundation/gym-notices` GitHub repository for the latest information if working offline.
Avoid using `gym` versions 0.24.0 and 0.24.1. Upgrade to a more stable version of `gym` or, preferably, migrate to `gymnasium` for continued support and bug fixes.
pip install gym-notices
Explicitly specify the `dtype` when creating `gym.spaces.Box` objects, for example: `observation_space = gym.spaces.Box(low=0.0, high=1.0, shape=(n,), dtype=np.float32)`.
To suppress warnings, set the logging level for 'gym' to ERROR: `import gym
gym.logger.set_level(gym.logger.ERROR)` or using Python's standard logging: `import logging
logging.getLogger('gym').setLevel(logging.ERROR)`.No dependency data recorded yet.