implement mark suggestions (test, docs, sets)

gym/core.py

@@ -28,6 +28,8 @@
 
 
 class _EnvDecorator(type):  # TODO: remove with gym 1.0
+    """Metaclass used for adding deprecation warning to the mode kwarg in the render method."""
+
     def __new__(cls, name, bases, attr):
         if "render" in attr.keys():
             attr["render"] = _EnvDecorator._deprecate_mode(attr["render"])
@@ -190,8 +192,10 @@ def render(self, mode="human") -> Optional[Union[RenderFrame, List[RenderFrame]]
         - None (default): no render is computed.
         - human: render return None.
           The environment is continuously rendered in the current display or terminal. Usually for human consumption.
-        - rgb_array: Return a list of frames. Each frame is a numpy.ndarray with shape (x, y, 3),
-          representing RGB values for an x-by-y pixel image.
+        - single_rgb_array: return a single frame representing the current state of the environment.
+          A frame is a numpy.ndarray with shape (x, y, 3) representing RGB values for an x-by-y pixel image.
+        - rgb_array: return a list of frames representing the states of the environment since the last reset.
+          Each frame is a numpy.ndarray with shape (x, y, 3), as with single_rgb_array.
         - ansi: Return a list of strings (str) or StringIO.StringIO containing a
           terminal-style text representation for each time step.
           The text can include newlines and ANSI escape sequences (e.g. for colors).

gym/envs/box2d/bipedal_walker.py

@@ -671,17 +671,17 @@ def _render(self, mode: str = "human"):
         self.lidar_render = (self.lidar_render + 1) % 100
         i = self.lidar_render
         if i < 2 * len(self.lidar):
-            l = (
+            single_lidar = (
                 self.lidar[i]
                 if i < len(self.lidar)
                 else self.lidar[len(self.lidar) - i - 1]
             )
-            if hasattr(l, "p1") and hasattr(l, "p2"):
+            if hasattr(single_lidar, "p1") and hasattr(single_lidar, "p2"):
                 pygame.draw.line(
                     self.surf,
                     color=(255, 0, 0),
-                    start_pos=(l.p1[0] * SCALE, l.p1[1] * SCALE),
-                    end_pos=(l.p2[0] * SCALE, l.p2[1] * SCALE),
+                    start_pos=(single_lidar.p1[0] * SCALE, single_lidar.p1[1] * SCALE),
+                    end_pos=(single_lidar.p2[0] * SCALE, single_lidar.p2[1] * SCALE),
                     width=1,
                 )
 
@@ -742,7 +742,7 @@ def _render(self, mode: str = "human"):
             pygame.event.pump()
             self.clock.tick(self.metadata["render_fps"])
             pygame.display.flip()
-        elif mode in ["rgb_array", "single_rgb_array"]:
+        elif mode in {"rgb_array", "single_rgb_array"}:
             return np.transpose(
                 np.array(pygame.surfarray.pixels3d(self.surf)), axes=(1, 0, 2)
             )

gym/envs/box2d/car_racing.py

@@ -566,9 +566,9 @@ def _render(self, mode: str = "human"):
             self.screen.blit(self.surf, (0, 0))
             pygame.display.flip()
 
-        if mode in ["rgb_array", "single_rgb_array"]:
+        if mode in {"rgb_array", "single_rgb_array"}:
             return self._create_image_array(self.surf, (VIDEO_W, VIDEO_H))
-        elif mode in ["state_pixels", "single_state_pixels"]:
+        elif mode in {"state_pixels", "single_state_pixels"}:
             return self._create_image_array(self.surf, (STATE_W, STATE_H))
         else:
             return self.isopen

gym/envs/box2d/lunar_lander.py

@@ -689,7 +689,7 @@ def _render(self, mode="human"):
             pygame.event.pump()
             self.clock.tick(self.metadata["render_fps"])
             pygame.display.flip()
-        elif mode in ["rgb_array", "single_rgb_array"]:
+        elif mode in {"rgb_array", "single_rgb_array"}:
             return np.transpose(
                 np.array(pygame.surfarray.pixels3d(self.surf)), axes=(1, 0, 2)
             )

gym/envs/classic_control/acrobot.py

@@ -304,7 +304,7 @@ def _render(self, mode="human"):
                 self.screen = pygame.display.set_mode(
                     (self.SCREEN_DIM, self.SCREEN_DIM)
                 )
-            else:  # mode in ["rgb_array", "single_rgb_array"]
+            else:  # mode in {"rgb_array", "single_rgb_array"}
                 self.screen = pygame.Surface((self.SCREEN_DIM, self.SCREEN_DIM))
         if self.clock is None:
             self.clock = pygame.time.Clock()
@@ -365,7 +365,7 @@ def _render(self, mode="human"):
             self.clock.tick(self.metadata["render_fps"])
             pygame.display.flip()
 
-        elif mode in ["rgb_array", "single_rgb_array"]:
+        elif mode in {"rgb_array", "single_rgb_array"}:
             return np.transpose(
                 np.array(pygame.surfarray.pixels3d(self.screen)), axes=(1, 0, 2)
             )

gym/envs/classic_control/cartpole.py

@@ -227,7 +227,7 @@ def _render(self, mode="human"):
                 self.screen = pygame.display.set_mode(
                     (self.screen_width, self.screen_height)
                 )
-            else:  # mode in ["rgb_array", "single_rgb_array"]
+            else:  # mode in {"rgb_array", "single_rgb_array"}
                 self.screen = pygame.Surface((self.screen_width, self.screen_height))
         if self.clock is None:
             self.clock = pygame.time.Clock()
@@ -295,7 +295,7 @@ def _render(self, mode="human"):
             self.clock.tick(self.metadata["render_fps"])
             pygame.display.flip()
 
-        elif mode in ["rgb_array", "single_rgb_array"]:
+        elif mode in {"rgb_array", "single_rgb_array"}:
             return np.transpose(
                 np.array(pygame.surfarray.pixels3d(self.screen)), axes=(1, 0, 2)
             )

gym/envs/classic_control/continuous_mountain_car.py

@@ -216,7 +216,7 @@ def _render(self, mode="human"):
                 self.screen = pygame.display.set_mode(
                     (self.screen_width, self.screen_height)
                 )
-            else:  # mode in ["rgb_array", "single_rgb_array"]
+            else:  # mode in {"rgb_array", "single_rgb_array"}
                 self.screen = pygame.Surface((self.screen_width, self.screen_height))
         if self.clock is None:
             self.clock = pygame.time.Clock()
@@ -290,7 +290,7 @@ def _render(self, mode="human"):
             self.clock.tick(self.metadata["render_fps"])
             pygame.display.flip()
 
-        elif mode in ["rgb_array", "single_rgb_array"]:
+        elif mode in {"rgb_array", "single_rgb_array"}:
             return np.transpose(
                 np.array(pygame.surfarray.pixels3d(self.screen)), axes=(1, 0, 2)
             )

gym/envs/classic_control/mountain_car.py

@@ -189,7 +189,7 @@ def _render(self, mode="human"):
                 self.screen = pygame.display.set_mode(
                     (self.screen_width, self.screen_height)
                 )
-            else:  # mode in ["rgb_array", "single_rgb_array"]
+            else:  # mode in {"rgb_array", "single_rgb_array"}
                 self.screen = pygame.Surface((self.screen_width, self.screen_height))
         if self.clock is None:
             self.clock = pygame.time.Clock()
@@ -263,7 +263,7 @@ def _render(self, mode="human"):
             self.clock.tick(self.metadata["render_fps"])
             pygame.display.flip()
 
-        elif mode in ["rgb_array", "single_rgb_array"]:
+        elif mode in {"rgb_array", "single_rgb_array"}:
             return np.transpose(
                 np.array(pygame.surfarray.pixels3d(self.screen)), axes=(1, 0, 2)
             )

gym/envs/classic_control/pendulum.py

@@ -181,7 +181,7 @@ def _render(self, mode="human"):
                 self.screen = pygame.display.set_mode(
                     (self.screen_dim, self.screen_dim)
                 )
-            else:  # mode in ["rgb_array", "single_rgb_array"]
+            else:  # mode in {"rgb_array", "single_rgb_array"}
                 self.screen = pygame.Surface((self.screen_dim, self.screen_dim))
         if self.clock is None:
             self.clock = pygame.time.Clock()

gym/envs/toy_text/frozen_lake.py

@@ -270,7 +270,7 @@ def _render(self, mode="human"):
         assert mode in self.metadata["render_modes"]
         if mode == "ansi":
             return self._render_text()
-        elif mode in ["human", "rgb_array", "single_rgb_array"]:
+        elif mode in {"human", "rgb_array", "single_rgb_array"}:
             return self._render_gui(mode)
 
     def _render_gui(self, mode):
@@ -287,7 +287,7 @@ def _render_gui(self, mode):
             pygame.display.set_caption("Frozen Lake")
             if mode == "human":
                 self.window_surface = pygame.display.set_mode(self.window_size)
-            elif mode in ["rgb_array", "single_rgb_array"]:
+            elif mode in {"rgb_array", "single_rgb_array"}:
                 self.window_surface = pygame.Surface(self.window_size)
         if self.clock is None:
             self.clock = pygame.time.Clock()
@@ -378,7 +378,7 @@ def _render_gui(self, mode):
             pygame.event.pump()
             pygame.display.update()
             self.clock.tick(self.metadata["render_fps"])
-        elif mode in ["rgb_array", "single_rgb_array"]:
+        elif mode in {"rgb_array", "single_rgb_array"}:
             return np.transpose(
                 np.array(pygame.surfarray.pixels3d(self.window_surface)), axes=(1, 0, 2)
             )

gym/envs/toy_text/taxi.py

@@ -252,7 +252,7 @@ def _render(self, mode):
         assert mode in self.metadata["render_modes"]
         if mode == "ansi":
             return self._render_text()
-        elif mode in ["human", "rgb_array", "single_rgb_array"]:
+        elif mode in {"human", "rgb_array", "single_rgb_array"}:
             return self._render_gui(mode)
 
     def _render_gui(self, mode):
@@ -268,7 +268,7 @@ def _render_gui(self, mode):
             pygame.display.set_caption("Taxi")
             if mode == "human":
                 self.window = pygame.display.set_mode(WINDOW_SIZE)
-            elif mode in ["rgb_array", "single_rgb_array"]:
+            elif mode in {"rgb_array", "single_rgb_array"}:
                 self.window = pygame.Surface(WINDOW_SIZE)
         if self.clock is None:
             self.clock = pygame.time.Clock()
@@ -376,7 +376,7 @@ def _render_gui(self, mode):
         if mode == "human":
             pygame.display.update()
             self.clock.tick(self.metadata["render_fps"])
-        elif mode in ["rgb_array", "single_rgb_array"]:
+        elif mode in {"rgb_array", "single_rgb_array"}:
             return np.transpose(
                 np.array(pygame.surfarray.pixels3d(self.window)), axes=(1, 0, 2)
             )

tests/envs/test_envs.py

@@ -72,6 +72,21 @@ def test_reset_info(spec):
     env.close()
 
 
+@pytest.mark.parametrize(
+    "spec", spec_list_no_mujoco_py, ids=[spec.id for spec in spec_list_no_mujoco_py]
+)
+def test_render_modes(spec):
+    env = spec.make()
+
+    for mode in env.metadata.get("render_modes", []):
+        if mode != "human":
+            new_env = spec.make(render_mode=mode)
+
+            new_env.reset()
+            new_env.step(new_env.action_space.sample())
+            new_env.render()
+
+
 def test_env_render_result_is_immutable():
     environs = [
         envs.make("Taxi-v3", render_mode="ansi"),