Return of the hidden obj npc reward

thatguy11325 · Mar 22, 2024 · b25b3c7 · b25b3c7
1 parent 5470b7c
commit b25b3c7
Show file tree

Hide file tree

Showing 3 changed files with 7 additions and 9 deletions.
diff --git a/config.yaml b/config.yaml
@@ -148,6 +148,8 @@ rewards:
       stats_menu: 0.1
       bag_menu: 0.1
       taught_cut: 10.0
+      explore_npcs: 0.02
+      explore_hidden_objs: 0.02
 
 
 policies:

diff --git a/pokemonred_puffer/environment.py b/pokemonred_puffer/environment.py
@@ -255,6 +255,8 @@ def reset(self, seed: Optional[int] = None):
             self.explore_map = np.zeros(GLOBAL_MAP_SHAPE, dtype=np.float32)
             self.cut_explore_map = np.zeros(GLOBAL_MAP_SHAPE, dtype=np.float32)
             self.init_mem()
+            # We only init seen hidden objs once cause they can only be found once!
+            self.seen_hidden_objs = {}
             self.reset_count = 0
             with open(self.init_state_path, "rb") as f:
                 self.pyboy.load_state(f)
@@ -266,7 +268,6 @@ def reset(self, seed: Optional[int] = None):
                 self.read_m(i).bit_count()
                 for i in range(EVENT_FLAGS_START, EVENT_FLAGS_START + EVENTS_FLAGS_LENGTH)
             )
-
         else:
             self.reset_count += 1
 
@@ -311,14 +312,10 @@ def init_mem(self):
         self.seen_coords = {}
         # self.seen_global_coords = np.zeros(GLOBAL_MAP_SHAPE)
         self.seen_map_ids = np.zeros(256)
-
         self.seen_npcs = {}
 
-        self.seen_hidden_objs = {}
-
         self.cut_coords = {}
         self.cut_tiles = {}
-        self.cut_state = deque(maxlen=3)
 
         self.seen_start_menu = 0
         self.seen_pokemon_menu = 0
@@ -329,13 +326,9 @@ def init_mem(self):
     def reset_mem(self):
         self.seen_coords.update((k, 0) for k, _ in self.seen_coords.items())
         self.seen_map_ids *= 0
-
         self.seen_npcs.update((k, 0) for k, _ in self.seen_npcs.items())
 
-        self.seen_hidden_objs.update((k, 0) for k, _ in self.seen_hidden_objs.items())
-
         self.cut_coords.update((k, 0) for k, _ in self.cut_coords.items())
-        self.cut_state = deque(maxlen=3)
 
         self.seen_start_menu = 0
         self.seen_pokemon_menu = 0

diff --git a/pokemonred_puffer/rewards/baseline.py b/pokemonred_puffer/rewards/baseline.py
@@ -152,6 +152,9 @@ def get_game_state_reward(self):
             "hm_count": self.reward_config["hm_count"] * self.get_hm_count(),
             "badges": self.reward_config["badges"] * self.get_badges(),
             "exploration": self.reward_config["exploration"] * sum(self.seen_coords.values()),
+            "explore_npcs": self.reward_config["explore_npcs"] * sum(self.seen_npcs.values()),
+            "explore_hidden_objs": self.reward_config["explore_hidden_objs"]
+            * sum(self.seen_hidden_objs.values()),
             "cut_coords": self.reward_config["cut_coords"] * sum(self.cut_coords.values()),
             "cut_tiles": self.reward_config["cut_tiles"] * sum(self.cut_tiles),
             "start_menu": self.reward_config["start_menu"] * self.seen_start_menu,