题意：尝试导入 OpenAI Gym 环境时遇到错误

问题背景：

I am trying to run an OpenAI Gym environment:

我正在尝试运行一个 OpenAI Gym 环境：

env = gym.make("ALE/Breakout-v5", render_mode="rgb_array")

But I get the following error        但是我得到了以下错误信息：

Traceback (most recent call last):
  File "/Users/----/Documents/Spring 2023/----/----/yum.py", line 10, in <module>
    env = gym.make("ALE/Breakout-v5", render_mode="rgb_array")
  File "/Users/----/anaconda3/envs/----/lib/python3.10/site-packages/gymnasium/envs/registration.py", line 592, in make
    _check_version_exists(ns, name, version)
  File "/Users/----/anaconda3/envs/----/lib/python3.10/site-packages/gymnasium/envs/registration.py", line 218, in _check_version_exists
    _check_name_exists(ns, name)
  File "/Users/----/anaconda3/envs/----/lib/python3.10/site-packages/gymnasium/envs/registration.py", line 185, in _check_name_exists
    _check_namespace_exists(ns)
  File "/Users/----/anaconda3/envs/----/lib/python3.10/site-packages/gymnasium/envs/registration.py", line 180, in _check_namespace_exists
    raise error.NamespaceNotFound(f"Namespace {ns} not found. {suggestion_msg}")
gymnasium.error.NamespaceNotFound: Namespace ALE not found. Have you installed the proper package for ALE?

I've looked at all the similar errors that people have discussed online, but I cannot find any fixes that work for me.

我查看了网上讨论的所有类似错误，但找不到任何对我有效的解决方法。

问题解决：

Installing the gym as below worked in my environment. (Python 3.7)

如下安装 `gym` 在我的环境中有效。（Python 3.7）

pip install "gym[atari, accept-rom-license]"

if you are using gymnasium:

如果你在使用 `gymnasium`：

pip install "gymnasium[atari, accept-rom-license]"

import gym # or "import gymnasium as gym"

if __name__ == '__main__':
    env = gym.make("ALE/Breakout-v5")

A.L.E: Arcade Learning Environment (version 0.8.1+53f58b7)
[Powered by Stella]

If it still doesn't work try adding the following import

如果仍然无效，试试添加以下导入：

import ale_py
# if using gymnasium
import shimmy

import gym # or "import gymnasium as gym"

Remember to create a new empty environment before installation.

记得在安装前创建一个新的空环境。