Fixed sample

tic_tac_toe.py

@@ -6,7 +6,7 @@ np.set_printoptions(formatter={'float_kind': float_formatter})
 
 
 def softmax(values: np.array, eps=1.0E-6) -> np.array:
-    result = (values + eps) / (sum(values) + eps)
+    result = (values + eps) / (sum(values + eps))
     return result
 
 
@@ -174,7 +174,7 @@ class MachinePlayer(APlayer):
                 v1 = self.get_value(self.state)
                 d = max(0, v1-v0)
                 if d > 0:
-                    self.set_value(0.01*d)
+                    self.set_value(0.2*d)
                     if self.with_debug:
                         print(f"{self.mark}: Learned {d:0.3f}")