miniworld-maze 1.3.2__tar.gz → 1.4.0__tar.gz

{miniworld_maze-1.3.2 → miniworld_maze-1.4.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: miniworld-maze
-Version: 1.3.2
+Version: 1.4.0
 Summary: Multi-room maze environments from the DrStrategy paper. Provides NineRooms-v0, SpiralNineRooms-v0, and TwentyFiveRooms-v0 gymnasium environments.
 Keywords: reinforcement-learning,environment,gymnasium,multi-room-maze,drstrategy,maze-navigation,partial-observability,3d-environments
 Author: Tim Joseph

{miniworld_maze-1.3.2 → miniworld_maze-1.4.0}/examples/benchmark_rendering.py

@@ -61,8 +61,8 @@ def benchmark_rendering_performance(
         # Create environment with specific observation size
         variant_mapping = {
             "NineRooms": "NineRooms-v0",
-            "SpiralNineRooms": "SpiralNineRooms-v0", 
-            "TwentyFiveRooms": "TwentyFiveRooms-v0"
+            "SpiralNineRooms": "SpiralNineRooms-v0",
+            "TwentyFiveRooms": "TwentyFiveRooms-v0",
         }
         env = gym.make(variant_mapping[variant], obs_width=size, obs_height=size)
 

{miniworld_maze-1.3.2 → miniworld_maze-1.4.0}/examples/generate_observations.py

@@ -29,7 +29,7 @@ def generate_observations(variant, output_dir=None, high_res_full_views=False):
     variant_mapping = {
         "NineRooms": "NineRooms-v0",
         "SpiralNineRooms": "SpiralNineRooms-v0",
-        "TwentyFiveRooms": "TwentyFiveRooms-v0"
+        "TwentyFiveRooms": "TwentyFiveRooms-v0",
     }
     env = gym.make(variant_mapping[variant], obs_width=64, obs_height=64)
 

{miniworld_maze-1.3.2 → miniworld_maze-1.4.0}/examples/live_control_opencv.py

@@ -121,7 +121,7 @@ class OpenCVLiveController:
         variant_mapping = {
             "NineRooms": "NineRooms-v0",
             "SpiralNineRooms": "SpiralNineRooms-v0",
-            "TwentyFiveRooms": "TwentyFiveRooms-v0"
+            "TwentyFiveRooms": "TwentyFiveRooms-v0",
         }
         self.env = gym.make(
             variant_mapping[variant],
@@ -215,7 +215,10 @@ class OpenCVLiveController:
         Returns:
             BGR image array for OpenCV or None if failed
         """
-        if self.current_info is not None and ObservationLevel.FIRST_PERSON in self.current_info:
+        if (
+            self.current_info is not None
+            and ObservationLevel.FIRST_PERSON in self.current_info
+        ):
             # Get first person observation from info
             obs_hwc = self.current_info[ObservationLevel.FIRST_PERSON]
 
@@ -253,8 +256,12 @@ class OpenCVLiveController:
             Map image with agent and goal markers
         """
         # Get positions from info dictionary - these are guaranteed to be available
-        agent_pos_2d = self.current_info["agent_position"]  # [x, z] coordinates as ndarray
-        goal_pos_2d = self.current_info["goal_position"]  # [x, z] coordinates as ndarray
+        agent_pos_2d = self.current_info[
+            "agent_position"
+        ]  # [x, z] coordinates as ndarray
+        goal_pos_2d = self.current_info[
+            "goal_position"
+        ]  # [x, z] coordinates as ndarray
 
         # Get environment bounds using utility function
         env_min_bounds, env_max_bounds = get_environment_bounds(self.map_env)
@@ -590,7 +597,7 @@ class OpenCVLiveController:
             ]
             if action < len(action_names):
                 print(
-                    f"🎯 {action_names[action]} | Reward: {reward:.2f} | "
+                    f"🎯 {action_names[action]} | Reward: {reward:.2f} | Success {self.current_info['success']} | "
                     f"Step: {self.step_count}"
                 )
 

{miniworld_maze-1.3.2 → miniworld_maze-1.4.0}/examples/save_topdown_views.py

@@ -21,151 +21,183 @@ import miniworld_maze  # noqa: F401
 def save_single_topdown_view(env: gym.Env) -> str:
     """Generate a single top-down full view and save it as PNG."""
     print("🎯 Generating single top-down full view...")
-    
+
     # Reset environment to ensure consistent state
     _, info = env.reset(seed=42)
-    
+
     # Get top-down full view from info dict
     obs_image = info[ObservationLevel.TOP_DOWN_FULL]
-    
+
     # Save the observation as PNG
     filename = "single_topdown_view.png"
     Image.fromarray(obs_image).save(filename)
     print(f"   ✅ Saved single top-down full view to: {filename}")
-    
+
     return filename
 
 
 def save_steps_comparison(env: gym.Env) -> List[str]:
     """Run 5 steps and save observation/desired_goal side by side."""
     print("🚶 Running 5 steps and saving observation/desired_goal comparisons...")
-    
+
     # Reset environment
     obs, _ = env.reset(seed=42)
-    
+
     saved_files = []
-    
+
     # Run 5 steps
     for step in range(5):
         # Take a random action
         action = env.action_space.sample()
         obs, reward, terminated, truncated, _ = env.step(action)
-        
+
         # Get observation and desired_goal
         observation = obs["observation"]
         desired_goal = obs["desired_goal"]
-        
+
         # Create side-by-side comparison
         # observation (left) and desired_goal (right)
         combined_width = observation.shape[1] + desired_goal.shape[1]
         combined_height = max(observation.shape[0], desired_goal.shape[0])
-        
+
         # Create blank canvas
-        combined_image = np.ones((combined_height, combined_width, 3), dtype=np.uint8) * 255
-        
+        combined_image = (
+            np.ones((combined_height, combined_width, 3), dtype=np.uint8) * 255
+        )
+
         # Place observation on the left
-        combined_image[:observation.shape[0], :observation.shape[1]] = observation
-        
+        combined_image[: observation.shape[0], : observation.shape[1]] = observation
+
         # Place desired_goal on the right
-        combined_image[:desired_goal.shape[0], observation.shape[1]:] = desired_goal
-        
+        combined_image[: desired_goal.shape[0], observation.shape[1] :] = desired_goal
+
         # Save the combined image
         filename = f"step_{step + 1}_obs_vs_desired.png"
         Image.fromarray(combined_image).save(filename)
         saved_files.append(filename)
-        
+
         print(f"   Step {step + 1}: reward={reward:.3f}, saved to {filename}")
-        
+
         # Reset if episode ended
         if terminated or truncated:
             obs, _ = env.reset()
-    
+
     return saved_files
 
 
 def collect_trajectory_and_visualize(env: gym.Env) -> str:
     """Execute 1000 random steps and visualize agent trajectory on top-down view."""
     print("🔍 Collecting agent trajectory over 1000 random steps...")
-    
+
     # Reset environment
     obs, info = env.reset(seed=42)
-    
+
     # Get scene extent from environment using the new public method
     env_unwrapped = env.unwrapped
     scene_extent = env_unwrapped.get_extent()
     scene_min_x, scene_max_x, scene_min_z, scene_max_z = scene_extent
-    
-    print(f"   Environment bounds: x=[{scene_min_x:.1f}, {scene_max_x:.1f}], z=[{scene_min_z:.1f}, {scene_max_z:.1f}]")
-    
+
+    print(
+        f"   Environment bounds: x=[{scene_min_x:.1f}, {scene_max_x:.1f}], z=[{scene_min_z:.1f}, {scene_max_z:.1f}]"
+    )
+
     # Collect trajectory positions
     trajectory_positions = []
-    
+
     # Execute 1000 random steps
     for step in range(1000):
         # Take random action first to get info dict with position
         action = env.action_space.sample()
         _, _, terminated, truncated, info = env.step(action)
-        
+
         # Get current agent position from info dict
         agent_pos = info["pos"]
         trajectory_positions.append((agent_pos[0], agent_pos[1]))  # x, z coordinates
-        
+
         # Reset if episode ended
         if terminated or truncated:
             obs, info = env.reset()
-        
+
         if (step + 1) % 200 == 0:
             print(f"   Completed {step + 1}/1000 steps")
-    
+
     print(f"   ✅ Collected {len(trajectory_positions)} position points")
-    
+
     # Get top-down view for background
     env.reset(seed=42)
     background_image = env.unwrapped.render_top_view(POMDP=False)
-    
+
     # Create matplotlib figure
     fig, ax = plt.subplots(1, 1, figsize=(12, 12))
-    
+
     # Display background image using actual scene bounds
     # Use 'upper' origin to match how single_topdown_view.png is saved (standard image coordinates)
-    ax.imshow(background_image, extent=[scene_extent[0], scene_extent[1], scene_extent[3], scene_extent[2]], origin='upper')
-    
+    ax.imshow(
+        background_image,
+        extent=[scene_extent[0], scene_extent[1], scene_extent[3], scene_extent[2]],
+        origin="upper",
+    )
+
     # Extract x and z coordinates (no normalization needed now)
     x_coords = [pos[0] for pos in trajectory_positions]
     z_coords = [pos[1] for pos in trajectory_positions]
-    
+
     # Plot trajectory using real world coordinates
-    ax.plot(x_coords, z_coords, 'r-', linewidth=2, alpha=0.7, label='Agent Trajectory')
-    
+    ax.plot(x_coords, z_coords, "r-", linewidth=2, alpha=0.7, label="Agent Trajectory")
+
     # Mark start and end points
     if trajectory_positions:
-        ax.scatter(x_coords[0], z_coords[0], c='green', s=100, marker='o', 
-                  label='Start', zorder=5, edgecolors='black', linewidths=2)
-        ax.scatter(x_coords[-1], z_coords[-1], c='red', s=100, marker='s', 
-                  label='End', zorder=5, edgecolors='black', linewidths=2)
-    
+        ax.scatter(
+            x_coords[0],
+            z_coords[0],
+            c="green",
+            s=100,
+            marker="o",
+            label="Start",
+            zorder=5,
+            edgecolors="black",
+            linewidths=2,
+        )
+        ax.scatter(
+            x_coords[-1],
+            z_coords[-1],
+            c="red",
+            s=100,
+            marker="s",
+            label="End",
+            zorder=5,
+            edgecolors="black",
+            linewidths=2,
+        )
+
     # Add colormap for trajectory progression (plasma colormap for better visibility)
     colors = plt.cm.plasma(np.linspace(0, 1, len(trajectory_positions)))
     for i in range(len(trajectory_positions) - 1):
-        ax.plot([x_coords[i], x_coords[i+1]], 
-               [z_coords[i], z_coords[i+1]], 
-               color=colors[i], linewidth=2, alpha=0.8)
-    
+        ax.plot(
+            [x_coords[i], x_coords[i + 1]],
+            [z_coords[i], z_coords[i + 1]],
+            color=colors[i],
+            linewidth=2,
+            alpha=0.8,
+        )
+
     # Formatting using scene bounds (flipped Z to match image orientation)
     ax.set_xlim(scene_min_x, scene_max_x)
     ax.set_ylim(scene_max_z, scene_min_z)
-    ax.set_title('Agent Trajectory over 1000 Random Steps', fontsize=16, fontweight='bold')
-    ax.set_xlabel('X Position (world coordinates)', fontsize=12)
-    ax.set_ylabel('Z Position (world coordinates)', fontsize=12)
-    ax.legend(loc='upper right')
-    
+    ax.set_title(
+        "Agent Trajectory over 1000 Random Steps", fontsize=16, fontweight="bold"
+    )
+    ax.set_xlabel("X Position (world coordinates)", fontsize=12)
+    ax.set_ylabel("Z Position (world coordinates)", fontsize=12)
+    ax.legend(loc="upper right")
+
     # Save the visualization
     filename = "agent_trajectory_visualization.png"
-    plt.savefig(filename, dpi=300, bbox_inches='tight')
+    plt.savefig(filename, dpi=300, bbox_inches="tight")
     plt.close()
-    
+
     print(f"   ✅ Saved trajectory visualization to: {filename}")
-    
+
     return filename
 
 
@@ -173,7 +205,7 @@ def main() -> None:
     """Main function demonstrating both features."""
     print("🖼️  Top-Down View Generator")
     print("=" * 50)
-    
+
     # Create a single environment instance with info_obs for top-down full view
     env = gym.make(
         "NineRooms-v0",
@@ -182,21 +214,21 @@ def main() -> None:
         obs_height=256,
         info_obs=[ObservationLevel.TOP_DOWN_FULL],
     )
-    
+
     try:
         # 1. Generate single top-down view
         single_file = save_single_topdown_view(env)
-        
+
         print("\n" + "=" * 50)
-        
+
         # 2. Run 5 steps and save comparisons
         step_files = save_steps_comparison(env)
-        
+
         print("\n" + "=" * 50)
-        
+
         # 3. Collect trajectory and visualize
         trajectory_file = collect_trajectory_and_visualize(env)
-        
+
         print("\n" + "=" * 50)
         print("🎉 Generation complete!")
         print("📁 Files created:")
@@ -204,12 +236,12 @@ def main() -> None:
         for file in step_files:
             print(f"   • {file} (observation vs desired_goal)")
         print(f"   • {trajectory_file} (agent trajectory visualization)")
-        
+
         print("\n💡 Tip: Open the PNG files to see the different views!")
-        
+
     finally:
         env.close()
 
 
 if __name__ == "__main__":
-    main()
+    main()

{miniworld_maze-1.3.2 → miniworld_maze-1.4.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "uv_build"
 
 [project]
 name = "miniworld-maze"
-version = "1.3.2"
+version = "1.4.0"
 description = "Multi-room maze environments from the DrStrategy paper. Provides NineRooms-v0, SpiralNineRooms-v0, and TwentyFiveRooms-v0 gymnasium environments."
 readme = "README.md"
 license = {file = "LICENSE"}

{miniworld_maze-1.3.2 → miniworld_maze-1.4.0}/src/miniworld_maze/__init__.py

@@ -25,7 +25,7 @@ if "PYGLET_HEADLESS" not in os.environ:
         "Automatically set PYGLET_HEADLESS=1 for headless rendering. "
         "Set PYGLET_HEADLESS=0 before importing miniworld_maze to override this behavior.",
         UserWarning,
-        stacklevel=2
+        stacklevel=2,
     )
 
 from .core import ObservationLevel
@@ -33,10 +33,6 @@ from .environments.nine_rooms import NineRooms
 from .environments.spiral_nine_rooms import SpiralNineRooms
 from .environments.twenty_five_rooms import TwentyFiveRooms
 
-# Import factory to register environments
-from .environments import factory  # noqa: F401
-
-__version__ = "1.1.0"
 __all__ = [
     "NineRooms",
     "SpiralNineRooms",

{miniworld_maze-1.3.2 → miniworld_maze-1.4.0}/src/miniworld_maze/core/miniworld_gymnasium/unified_env.py

@@ -9,29 +9,71 @@ import gymnasium as gym
 import numpy as np
 import pyglet
 from gymnasium import spaces
-from pyglet.gl import *
-
-from ..observation_types import ObservationLevel
-from .entities import *
-from .math import *
-from .objmesh import *
-from .occlusion_queries import OcclusionQueryManager
-from .opengl import *
-from .params import DEFAULT_PARAMS
-from .random import RandGen
-from .room import Room
+from pyglet.gl import (
+    glEnable,
+    glDeleteLists,
+    glNewList,
+    glLightfv,
+    glShadeModel,
+    glColorMaterial,
+    glCallList,
+    glClearColor,
+    glClearDepth,
+    glClear,
+    glMatrixMode,
+    glLoadIdentity,
+    glOrtho,
+    glLoadMatrixf,
+    glDisable,
+    glBindFramebuffer,
+    glFlush,
+    glEndList,
+    gluPerspective,
+    gluLookAt,
+    GL_DEPTH_TEST,
+    GL_CULL_FACE,
+    GL_COMPILE,
+    GL_LIGHT0,
+    GL_POSITION,
+    GL_AMBIENT,
+    GL_DIFFUSE,
+    GL_SMOOTH,
+    GL_FRONT_AND_BACK,
+    GL_AMBIENT_AND_DIFFUSE,
+    GL_LIGHTING,
+    GL_COLOR_MATERIAL,
+    GL_COLOR_BUFFER_BIT,
+    GL_DEPTH_BUFFER_BIT,
+    GL_PROJECTION,
+    GL_MODELVIEW,
+    GL_TEXTURE_2D,
+    GL_FRAMEBUFFER,
+    GLubyte,
+    GLfloat,
+)
+
+from miniworld_maze.core.observation_types import ObservationLevel
+from miniworld_maze.core.miniworld_gymnasium.entities import Entity, Agent
+from miniworld_maze.core.miniworld_gymnasium.math import Y_VEC, intersect_circle_segs
+from miniworld_maze.core.miniworld_gymnasium.occlusion_queries import (
+    OcclusionQueryManager,
+)
+from miniworld_maze.core.miniworld_gymnasium.opengl import Texture, FrameBuffer, drawBox
+from miniworld_maze.core.miniworld_gymnasium.params import DEFAULT_PARAMS
+from miniworld_maze.core.miniworld_gymnasium.random import RandGen
+from miniworld_maze.core.miniworld_gymnasium.room import Room
 
 # Optional architectural improvements
 try:
-    from .entity_manager import EntityManager
-    from .rendering_engine import RenderingEngine
+    from miniworld_maze.core.miniworld_gymnasium.entity_manager import EntityManager
+    from miniworld_maze.core.miniworld_gymnasium.rendering_engine import RenderingEngine
 
     ARCHITECTURAL_IMPROVEMENTS_AVAILABLE = True
 except ImportError:
     RenderingEngine = None
     EntityManager = None
     ARCHITECTURAL_IMPROVEMENTS_AVAILABLE = False
-from ..constants import (
+from miniworld_maze.core.constants import (
     CARRY_POSITION_OFFSET,
     DEFAULT_DISPLAY_WIDTH,
     DEFAULT_WINDOW_HEIGHT,
@@ -47,6 +89,7 @@ from ..constants import (
     ORTHOGRAPHIC_DEPTH_RANGE,
     PICKUP_RADIUS_MULTIPLIER,
     PICKUP_REACH_MULTIPLIER,
+    POMDP_VIEW_RADIUS,
     PORTAL_CONNECTION_TOLERANCE,
     TEXT_LABEL_WIDTH,
     TEXT_MARGIN_X,
@@ -363,8 +406,6 @@ class UnifiedMiniWorldEnv(gym.Env):
             render_agent: Whether to render the agent in the observation.
                          If None, uses default behavior based on observation level.
         """
-        # Import ObservationLevel here to avoid circular imports
-        from ..observation_types import ObservationLevel
 
         if observation_level == ObservationLevel.TOP_DOWN_PARTIAL:
             if self.agent_mode == "empty":
@@ -1039,9 +1080,6 @@ class UnifiedMiniWorldEnv(gym.Env):
     def _calculate_scene_extents(self, POMDP):
         """Calculate scene extents for rendering."""
         if POMDP:
-            # Import constants for POMDP view radius
-            from ..constants import POMDP_VIEW_RADIUS
-
             agent_x, _, agent_z = self.agent.pos
             min_x = agent_x - POMDP_VIEW_RADIUS
             max_x = agent_x + POMDP_VIEW_RADIUS
@@ -1287,9 +1325,6 @@ class UnifiedMiniWorldEnv(gym.Env):
         if view == "agent":
             img = self.render_obs(self.vis_fb)
         else:
-            # Import ObservationLevel here to avoid circular imports
-            from ..observation_types import ObservationLevel
-
             if self.obs_level == ObservationLevel.TOP_DOWN_PARTIAL:
                 img = self.render_top_view(self.vis_fb, POMDP=True)
             else:

miniworld_maze-1.3.2/src/miniworld_maze/environments/factory.py → miniworld_maze-1.4.0/src/miniworld_maze/environments/__init__.py

@@ -1,8 +1,20 @@
-"""Gymnasium environment registrations for Nine Rooms environment variants."""
+"""Nine Rooms environment implementations."""
+
+from miniworld_maze.environments.base_grid_rooms import GridRoomsEnvironment
+from miniworld_maze.environments.nine_rooms import NineRooms
+from miniworld_maze.environments.spiral_nine_rooms import SpiralNineRooms
+from miniworld_maze.environments.twenty_five_rooms import TwentyFiveRooms
 
 from gymnasium.envs.registration import register
-from ..core import ObservationLevel
-from ..core.constants import FACTORY_DOOR_SIZE, FACTORY_ROOM_SIZE
+from miniworld_maze.core import ObservationLevel
+from miniworld_maze.core.constants import FACTORY_DOOR_SIZE, FACTORY_ROOM_SIZE
+
+__all__ = [
+    "GridRoomsEnvironment",
+    "NineRooms",
+    "SpiralNineRooms",
+    "TwentyFiveRooms",
+]
 
 # Register environment variants with factory defaults matching the original wrapper
 register(
@@ -18,7 +30,7 @@ register(
 )
 
 register(
-    id="SpiralNineRooms-v0", 
+    id="SpiralNineRooms-v0",
     entry_point="miniworld_maze.environments.spiral_nine_rooms:SpiralNineRooms",
     max_episode_steps=1000,
     kwargs={
@@ -31,7 +43,7 @@ register(
 
 register(
     id="TwentyFiveRooms-v0",
-    entry_point="miniworld_maze.environments.twenty_five_rooms:TwentyFiveRooms", 
+    entry_point="miniworld_maze.environments.twenty_five_rooms:TwentyFiveRooms",
     max_episode_steps=1000,
     kwargs={
         "room_size": FACTORY_ROOM_SIZE,

{miniworld_maze-1.3.2 → miniworld_maze-1.4.0}/src/miniworld_maze/environments/base_grid_rooms.py

@@ -242,9 +242,8 @@ class GridRoomsEnvironment(UnifiedMiniWorldEnv):
         agent_pos = self.agent.pos
         info["agent_position"] = np.array([agent_pos[0], agent_pos[2]])  # x, z
 
-        if hasattr(self, "_current_goal_position"):
-            goal_pos = self._current_goal_position
-            info["goal_position"] = np.array([goal_pos[0], goal_pos[2]])  # x, z
+        goal_pos = self._current_goal_position
+        info["goal_position"] = np.array([goal_pos[0], goal_pos[2]])  # x, z
 
         # Return observation as dict
         obs_dict = self._build_observation_dict(obs)
@@ -265,6 +264,8 @@ class GridRoomsEnvironment(UnifiedMiniWorldEnv):
         # Call parent reset
         obs, info = super().reset(seed=seed, options=options, pos=pos)
 
+        info["success"] = 0.0
+
         # Generate goal
         self.desired_goal = self._get_goal()
 
@@ -272,9 +273,8 @@ class GridRoomsEnvironment(UnifiedMiniWorldEnv):
         agent_pos = self.agent.pos
         info["agent_position"] = np.array([agent_pos[0], agent_pos[2]])  # x, z
 
-        if hasattr(self, "_current_goal_position"):
-            goal_pos = self._current_goal_position
-            info["goal_position"] = np.array([goal_pos[0], goal_pos[2]])  # x, z
+        goal_pos = self._current_goal_position
+        info["goal_position"] = np.array([goal_pos[0], goal_pos[2]])  # x, z
 
         # Return observation as dict with desired_goal and achieved_goal
         obs_dict = self._build_observation_dict(obs)
@@ -381,22 +381,24 @@ class GridRoomsEnvironment(UnifiedMiniWorldEnv):
                     goal_positions.append([[center_x, 0.0, center_z]])
                 else:
                     # Two goals per room: center-left and center-right
-                    goal_positions.append([
-                        [center_x - 1.0, 0.0, center_z],  # left goal
-                        [center_x + 1.0, 0.0, center_z],  # right goal
-                    ])
+                    goal_positions.append(
+                        [
+                            [center_x - 1.0, 0.0, center_z],  # left goal
+                            [center_x + 1.0, 0.0, center_z],  # right goal
+                        ]
+                    )
         return goal_positions
 
     def get_extent(self, padding: float = 1.0) -> Tuple[float, float, float, float]:
         """
         Get the scene extent for use with matplotlib imshow.
-        
-        Returns the scene bounds with padding in the format expected by 
+
+        Returns the scene bounds with padding in the format expected by
         matplotlib's imshow(extent=...) parameter: (left, right, bottom, top).
-        
+
         Args:
             padding: Padding to add around environment bounds (default: 1.0)
-            
+
         Returns:
             Tuple[float, float, float, float]: (min_x, max_x, min_z, max_z) with padding
         """
@@ -404,7 +406,7 @@ class GridRoomsEnvironment(UnifiedMiniWorldEnv):
             self.min_x - padding,
             self.max_x + padding,
             self.min_z - padding,
-            self.max_z + padding
+            self.max_z + padding,
         )
 
     def _build_observation_dict(self, obs: np.ndarray) -> dict:

miniworld_maze-1.3.2/src/miniworld_maze/environments/__init__.py

@@ -1,13 +0,0 @@
-"""Nine Rooms environment implementations."""
-
-from .base_grid_rooms import GridRoomsEnvironment
-from .nine_rooms import NineRooms
-from .spiral_nine_rooms import SpiralNineRooms
-from .twenty_five_rooms import TwentyFiveRooms
-
-__all__ = [
-    "GridRoomsEnvironment",
-    "NineRooms",
-    "SpiralNineRooms",
-    "TwentyFiveRooms",
-]

miniworld_maze-1.3.2/src/miniworld_maze/tools/__init__.py

@@ -1,3 +0,0 @@
-"""Tools for Nine Rooms environments."""
-
-__all__ = []