gymからgymnasiumに変更

nes_py/nes_env.py

@@ -4,9 +4,9 @@
 import itertools
 import os
 import sys
-import gym
-from gym.spaces import Box
-from gym.spaces import Discrete
+import gymnasium as gym
+from gymnasium.spaces import Box
+from gymnasium.spaces import Discrete
 import numpy as np
 from ._rom import ROM
 from ._image_viewer import ImageViewer
@@ -86,6 +86,8 @@ class NESEnv(gym.Env):
     # relevant meta-data about the environment
     metadata = {
         'render.modes': ['rgb_array', 'human'],
+        'render_modes': ['rgb_array', 'human'],
+        'render_fps': 50,
         'video.frames_per_second': 60
     }
 
@@ -270,7 +272,7 @@ def reset(self, seed=None, options=None, return_info=None):
         # set the done flag to false
         self.done = False
         # return the screen from the emulator
-        return self.screen
+        return self.screen, {}
 
     def _did_reset(self):
         """Handle any RAM hacking after a reset occurs."""
@@ -312,7 +314,8 @@ def step(self, action):
         elif reward > self.reward_range[1]:
             reward = self.reward_range[1]
         # return the screen from the emulator and other relevant data
-        return self.screen, reward, self.done, info
+        truncated = False
+        return self.screen, reward, self.done, truncated, info
 
     def _get_reward(self):
         """Return the reward after a step occurs."""

nes_py/tests/test_multiple_makes.py

@@ -26,7 +26,7 @@ def play(steps):
         if done:
             _ = env.reset()
         action = env.action_space.sample()
-        _, _, done, _ = env.step(action)
+        _, _, done, _, _ = env.step(action)
     # close the environment
     env.close()
 
@@ -84,4 +84,4 @@ def test(self):
                 if dones[idx]:
                     _ = envs[idx].reset()
                 action = envs[idx].action_space.sample()
-                _, _, dones[idx], _ = envs[idx].step(action)
+                _, _, dones[idx], _, _ = envs[idx].step(action)

nes_py/tests/test_nes_env.py

@@ -1,6 +1,6 @@
 """Test cases for the NESEnv class."""
 from unittest import TestCase
-import gym
+import gymnasium as gym
 import numpy as np
 from .rom_file_abs_path import rom_file_abs_path
 from nes_py.nes_env import NESEnv
@@ -79,17 +79,17 @@ def test(self):
         for _ in range(500):
             if done:
                 # reset the environment and check the output value
-                state = env.reset()
+                state, _ = env.reset()
                 self.assertIsInstance(state, np.ndarray)
             # sample a random action and check it
             action = env.action_space.sample()
-            self.assertIsInstance(action, int)
+            self.assertIsInstance(action, np.int64)
             # take a step and check the outputs
             output = env.step(action)
             self.assertIsInstance(output, tuple)
-            self.assertEqual(4, len(output))
+            self.assertEqual(5, len(output))
             # check each output
-            state, reward, done, info = output
+            state, reward, done, _, info = output
             self.assertIsInstance(state, np.ndarray)
             self.assertIsInstance(reward, float)
             self.assertIsInstance(done, bool)
@@ -108,9 +108,9 @@ def test(self):
 
         for _ in range(250):
             if done:
-                state = env.reset()
+                state, _ = env.reset()
                 done = False
-            state, _, done, _ = env.step(0)
+            state, _, done, _, _ = env.step(0)
 
         backup = state.copy()
 
@@ -120,7 +120,7 @@ def test(self):
             if done:
                 state = env.reset()
                 done = False
-            state, _, done, _ = env.step(0)
+            state, _, done, _, _ = env.step(0)
 
         self.assertFalse(np.array_equal(backup, state))
         env._restore()

nes_py/wrappers/joypad_space.py

@@ -1,7 +1,7 @@
 """An environment wrapper to convert binary to discrete action space."""
-import gym
-from gym import Env
-from gym import Wrapper
+import gymnasium as gym
+from gymnasium import Env
+from gymnasium import Wrapper
 
 
 class JoypadSpace(Wrapper):
@@ -73,9 +73,9 @@ def step(self, action):
         # take the step and record the output
         return self.env.step(self._action_map[action])
 
-    def reset(self):
+    def reset(self, **kwargs):
         """Reset the environment and return the initial observation."""
-        return self.env.reset()
+        return self.env.reset(**kwargs)
 
     def get_keys_to_action(self):
         """Return the dictionary of keyboard keys to actions."""

requirements.txt

@@ -1,4 +1,4 @@
-gym>=0.17.2
+gymnasium~=0.29.1
 numpy>=1.18.5
 pyglet<=1.5.21,>=1.4.0
 tqdm>=4.48.2

setup.py

@@ -69,7 +69,7 @@
     ext_modules=[LIB_NES_ENV],
     zip_safe=False,
     install_requires=[
-        'gym>=0.17.2',
+        'gymnasium>=0.29.1,<0.30.0',
         'numpy>=1.18.5',
         'pyglet<=1.5.21,>=1.4.0',
         'tqdm>=4.48.2',