fix PaddleRL policy_gradient bug

PaddleRL/policy_gradient/brain.py

@@ -22,6 +22,8 @@ class PolicyGradient:
         self.place = fluid.CPUPlace()
         self.exe = fluid.Executor(self.place)
 
+        self.all_act_prob = None
+
     def build_net(self):
 
         obs = fluid.layers.data(
@@ -31,10 +33,10 @@ class PolicyGradient:
         # fc1
         fc1 = fluid.layers.fc(input=obs, size=10, act="tanh")  # tanh activation
         # fc2
-        all_act_prob = fluid.layers.fc(input=fc1,
+        self.all_act_prob = fluid.layers.fc(input=fc1,
                                        size=self.n_actions,
                                        act="softmax")
-        self.inferece_program = fluid.defaul_main_program().clone()
+        self.inferece_program = fluid.default_main_program().clone()
         # to maximize total reward (log_p * R) is to minimize -(log_p * R)
         neg_log_prob = fluid.layers.cross_entropy(
             input=self.all_act_prob,

PaddleRL/policy_gradient/env.py

@@ -43,9 +43,9 @@ class Env():
 
     def draw(self, new_line=False):
         if new_line:
-            print ""
+            print("")
         else:
-            print "\r",
+            print("\r")
         for i in range(self.stage_len):
             if i == self.position:
                 sys.stdout.write("O")