todo: implement q learning

Player.py

@@ -58,21 +58,8 @@ class ComputerPlayer(Player):
     '''
 
     def choose_action(self, positions: list, current_board: np.array = None, symbol: int = -1) -> tuple:
-        if np.random.uniform(0, 1) <= self.exp_rate:
-            # take random action
-            idx = np.random.choice(len(positions))
-            action = positions[idx]
-        else:
-            value_max = -999
-            for p in positions:
-                next_board = current_board.copy()
-                next_board[p] = symbol
-                next_board_hash = self.get_hash(next_board)
-                value = 0 if self.states_value.get(next_board_hash) is None else self.states_value.get(next_board_hash)
-                if value >= value_max:
-                    value_max = value
-                    action = p
-        return action
+        # TODO: Implement this
+        pass
 
     def add_state(self, state: np.array) -> None:
         self.states.append(state)
@@ -82,12 +69,8 @@ class ComputerPlayer(Player):
     '''
 
     def feed_reward(self, reward: float) -> None:
-        for st in reversed(self.states):
-            if self.states_value.get(st) is None:
-                self.states_value[st] = 0
-            # bellman equation
-            self.states_value[st] += self.lr * (self.decay_gamma * reward - self.states_value[st])
-            reward = self.states_value[st]
+        # TODO: Implement this
+        pass
 
     def reset(self) -> None:
         self.states = []