add learning after episode

a_player.py

@@ -21,10 +21,10 @@ class APlayer(object):
 		state.reshape(state.size)[move] = self.mark
 		return state
 
-	def new_game(self):
+	def new_game(self, state: np.array):
 		pass
 
-	def end_game(self, reward):
+	def end_game(self, state: np.array, value):
 		pass
 
 	def has_won(self, state: np.array):