Check for Dirt vs wall collisions

environments/factory/renderer.py

@@ -1,4 +1,3 @@
-import numpy as np
 import pygame
 from pathlib import Path
 
@@ -66,5 +65,5 @@ class Renderer:
 if __name__ == '__main__':
     renderer = Renderer(fps=2, cell_size=40, assets=['wall', 'agent', 'dirt'])
     for i in range(15):
-        renderer.render({'agent': [(5, i)], 'wall': [(0, i), (i, 0)], 'dirt': [(3,3), (3,4)]})
+        renderer.render({'agent': [(5, i)], 'wall': [(0, i), (i, 0)], 'dirt': [(3, 3), (3, 4)]})
 

environments/factory/simple_factory_getting_dirty.py

@@ -1,14 +1,15 @@
-from collections import defaultdict, OrderedDict
+from collections import OrderedDict
+from dataclasses import dataclass
 from typing import List
 
 import numpy as np
-from attr import dataclass
 
 from environments.factory.base_factory import BaseFactory, AgentState
 from environments import helpers as h
 
 from environments.factory.renderer import Renderer
 
+
 DIRT_INDEX = -1
 @dataclass
 class DirtProperties:
@@ -20,6 +21,7 @@ class DirtProperties:
 class GettingDirty(BaseFactory):
 
     def _is_clean_up_action(self, action):
+        # Account for NoOP; remove -1 when activating NoOP
         return self.movement_actions + 1 - 1 == action
 
     def __init__(self, *args, dirt_properties: DirtProperties, **kwargs):
@@ -30,12 +32,12 @@ class GettingDirty(BaseFactory):
 
     def render(self):
         if not self.renderer:  # lazy init
-            h, w = self.state.shape[1:]
-            self.renderer = Renderer(w, h, view_radius=0)
-        self.renderer.render(  # todo: nur fuers prinzip, ist hardgecoded Dreck aktuell
-            OrderedDict(dirt=np.argwhere(self.state[DIRT_INDEX] > 0), # Ordered dict defines the drawing order! important
-                        wall=np.argwhere(self.state[0] > 0),
-                        agent=np.argwhere(self.state[1] > 0)
+            height, width = self.state.shape[1:]
+            self.renderer = Renderer(width, height, view_radius=0)
+        self.renderer.render(
+            OrderedDict(dirt=np.argwhere(self.state[DIRT_INDEX] > 0),
+                        wall=np.argwhere(self.state[h.LEVEL_IDX] > 0),
+                        agent=np.argwhere(self.state[h.AGENT_START_IDX] > 0)
                         )
         )
 
@@ -88,6 +90,12 @@ class GettingDirty(BaseFactory):
 
     def calculate_reward(self, agent_states: List[AgentState]) -> (int, dict):
         this_step_reward = 0
+
+        dirt_vs_level_collisions = np.argwhere(self.state[h.LEVEL_IDX, DIRT_INDEX].sum(0) == h.IS_FREE_CELL)
+        for dirt_vs_level_collision in dirt_vs_level_collisions:
+            print(f'Dirt was placed on Level at: {dirt_vs_level_collision.squeeze()}')
+            pass
+
         for agent_state in agent_states:
             collisions = agent_state.collisions
             print(f't = {self.steps}\tAgent {agent_state.i} has collisions with '
@@ -116,7 +124,8 @@ if __name__ == '__main__':
         state, r, done, _ = factory.reset()
         for action in random_actions:
             state, r, done, info = factory.step(action)
-            if render: factory.render()
+            if render:
+                factory.render()
         monitor_list.append(factory.monitor.to_pd_dataframe())
         print(f'Factory run {epoch} done, reward is:\n    {r}')