Minor changes

qlearning.py

@@ -128,14 +128,16 @@ class QTable:
 
 
 # Perform learning
+width,height=10,10
 perf=0
 last_state=None
 last_action=None
-game=Snake(length=1,fps=300,startat=(random.randint(0,29),random.randint(0,29)))
+game=Snake(length=1,fps=500,startat=(random.randint(0,width-1),random.randint(0,height-1)),grid_width=width,grid_height=height)
 qtable=QTable("qtable.txt")
 while True:
     result=0
     stuck=0
+    stuck_tolerance=1
     state=qtable.get_state(game)
     while result >= 0: 
         action=qtable.get_action(state)
@@ -143,7 +145,7 @@ while True:
         new_state=qtable.get_state(game)
 
         # Agent is stuck
-        if stuck>=(game.grid_width*game.grid_height)/2:
+        if stuck>=(game.grid_width*game.grid_height)/stuck_tolerance:
             game.new_game()
             break
 
@@ -153,7 +155,7 @@ while True:
             reward=-10
             stuck=0
         elif result==1:
-            reward=1
+            reward=50
             stuck=0
 
         # Apply learning    

snake.py

@@ -18,6 +18,7 @@ class Snake:
         self.fps=fps
         self.startat=startat
         self.last_score=-1
+        self.score=0
         # Setup pygame
         pygame.init()
         self.font=pygame.font.SysFont(pygame.font.get_default_font(), int(self.margin/2))
@@ -35,6 +36,7 @@ class Snake:
         self.snake=[self.startat]*self.default_length
         self.direction=3 # Like clock (12=up, 3=right, 6=bottom, 9=left)
         self.new_apple()
+        self.last_score=self.score
         self.score=0
         self.attempt+=1
 
@@ -134,6 +136,10 @@ class Snake:
         Play using wall clock directions (12=up, 3=right, 6=down and 9=left)
         """
         # Play
+        if  (self.direction==12 and direction!=6) or\
+            (self.direction==6 and direction!=12) or\
+            (self.direction==3 and direction!=9) or\
+            (self.direction==9 and direction!=3):
             self.direction=direction
         self.move()
         # Return code
@@ -144,7 +150,6 @@ class Snake:
             self.score+=1
             code=1
         elif self.has_loose():
-            self.last_score=self.score
             self.new_game()
             code=-1
         # Refresh screen