OpenAI Gym step函数解包5变量仍报错：ValueError(预期5值仅得4个)

按照官方文档编写代码后仍触发过时错误，代码如下：

done = True
env.reset()
for step in range(5000):
    action = env.action_space.sample()
    obs, reward, terminated, truncated, info = env.step(action)
    done = terminated or truncated

    if done:
       env.reset()

env.close()

报错信息：

ValueError                                Traceback (most recent call last)
/UsersProjects/GAME_AI_RL/Super_Mario/main.ipynb Cell 8 line 5
3 for step in range(5000):
4     action = env.action_space.sample()
----> 5     obs, reward, terminated, truncated, info = env.step(action)
6     done = terminated or truncated
8     if done:

File ~/miniforge3/envs/mlp/lib/python3.8/site-packages/nes_py/wrappers/joypad_space.py:74, in JoypadSpace.step(self, action)
59 """
60 Take a step using the given action.
61
(...)
71
72 """
73 # take the step and record the output
---> 74 return self.env.step(self._action_map[action])

File ~/miniforge3/envs/mlp/lib/python3.8/site-packages/gym/wrappers/time_limit.py:50, in TimeLimit.step(self, action)
39 def step(self, action):
40     """Steps through the environment and if the number of steps elapsed exceeds <code>max_episode_steps</code> then truncate.
41
42     Args:
(...)
...
---> 50     observation, reward, terminated, truncated, info = self.env.step(action)
51     self._elapsed_steps += 1
53     if self._elapsed_steps >= self._max_episode_steps:

ValueError: not enough values to unpack (expected 5, got 4)

运行环境说明：MACOS M1，基于虚拟环境。最初使用4个变量解包，查询得知接口已更新为5变量后修改，但仍报错；已尝试重启内核、重新安装库、更换Python版本。

解决方案 #

核心问题是你使用的nes_py库（对应Super Mario环境）未适配Gym新API——Gym新API的step方法返回5个值（obs, reward, terminated, truncated, info），但该库底层仍返回旧API的4个值（obs, reward, done, info）。

可选方案： #

• 方案1：回退到Gym旧版本
  安装Gym 0.25.x及以前的版本，使用旧API的4变量解包逻辑：

  done = True
  env.reset()
  for step in range(5000):
      action = env.action_space.sample()
      obs, reward, done, info = env.step(action)
  
      if done:
          env.reset()
  
  env.close()
  

• 方案2：切换到Gymnasium+适配新API的环境包
  使用Gymnasium（Gym的官方继任者），同时安装支持新API的gym-super-mario-bros最新版本，代码保持你当前的5变量解包逻辑即可。

• 方案3：手动封装Wrapper转换API格式
  自己写一个Wrapper将旧API的返回值转换为新API格式，无需修改环境依赖：

  import gym
  from gym import Wrapper
  
  class OldToNewAPIStepWrapper(Wrapper):
      def step(self, action):
          obs, reward, done, info = self.env.step(action)
          # 旧API的done对应新API的terminated，truncated默认设为False（如需区分可自行扩展）
          terminated = done
          truncated = False
          return obs, reward, terminated, truncated, info
  
  # 创建环境后应用该Wrapper
  env = gym.make('SuperMarioBros-v0')
  env = OldToNewAPIStepWrapper(env)
  

内容的提问来源于stack exchange，提问作者Rishit Chugh