uhhhh

nim.py

@@ -56,11 +56,17 @@ class NimAI():
         float: The Q-value associated with the (state, action) pair. 
                Returns 0 if the pair is not yet in the Q-table.
     """
-        print(self.q)
+        print(self.q, state, action)
+        try:
+            return self.q[(tuple(state), action)]
+        except:
+            return 0
 
     def update_q_value(self, state, action, old_q, reward, future_q):
         """
     Update the Q-value for a state-action pair using the Q-learning formula.
+
+    Q(s, a) ← Q(s, a) + α * (Belohnung + γ * max_a' Q(s', a') - Q(s, a))
     
     Parameters:
         state (list): The current game state.
@@ -69,10 +75,11 @@ class NimAI():
         reward (float): The reward received after taking the action.
         future_q (float): The maximum Q-value for the next state.
     """
-        raise NotImplementedError
+        self.q[tuple(state), action] = old_q + self.alpha * (reward + self.epsilon * future_q - old_q)
+        return 0
     
     def best_future_reward(self, state):
-            """
+        """
     Determine the highest Q-value among all possible actions in a given state.
     
     Parameters:
@@ -82,7 +89,15 @@ class NimAI():
         float: The highest Q-value among available actions. 
                Returns 0 if no actions are available.
     """
-            raise NotImplementedError
+        # actions = []
+        # for q in self.q.key:
+        #     if q[0] == state:
+        #         actions.append(q[1])
+        actions = tuple([key[1] for key in self.q.keys() if key[0] == state])
+        try:
+            return max([q for q in self.q[tuple(state), actions]])
+        except:
+            return 0
 
     def choose_action(self, state, epsilon=True):
         """
@@ -95,7 +110,15 @@ class NimAI():
     Returns:
         tuple: The chosen action from the available actions.
     """
-        raise NotImplementedError
+        if epsilon:
+            return random.choice(tuple(Nim.available_actions(state)))
+            # keys = [key[1] for key in self.q.key if key[0] == state]
+            # for key in keys:
+        else:
+            try:
+                return max([key[1] for key in self.q.keys() if key[0] == state])
+            except:
+                return (0,0)
 
 def train(n):
     player = NimAI()