diff --git a/adversarial/fluid_mnist.py b/adversarial/fluid_mnist.py
index d46defda55ee5ffeea529905f8049a3818a75276..031928e994cfe26e04a4779acfd9ec783ed22e6c 100644
--- a/adversarial/fluid_mnist.py
+++ b/adversarial/fluid_mnist.py
@@ -4,6 +4,7 @@ CNN on mnist data using fluid api of paddlepaddle
 import paddle.v2 as paddle
 import paddle.v2.fluid as fluid
 
+
 def mnist_cnn_model(img):
     """
     Mnist cnn model
@@ -31,10 +32,7 @@ def mnist_cnn_model(img):
         pool_stride=2,
         act='relu')
 
-    logits = fluid.layers.fc(
-        input=conv_pool_2,
-        size=10,
-        act='softmax')
+    logits = fluid.layers.fc(input=conv_pool_2, size=10, act='softmax')
     return logits
 
 
@@ -73,17 +71,19 @@ def main():
                                 feed=feeder.feed(data),
                                 fetch_list=[avg_cost] + accuracy.metrics)
             pass_acc = accuracy.eval(exe)
-            print("pass_id=" + str(pass_id) + " acc=" + str(acc) + " pass_acc=" +
-                  str(pass_acc))
+            print("pass_id=" + str(pass_id) + " acc=" + str(acc) + " pass_acc="
+                  + str(pass_acc))
             # print loss, acc
             if loss < LOSS_THRESHOLD and pass_acc > ACC_THRESHOLD:
                 # if avg cost less than 10.0 and accuracy is larger than 0.9, we think our code is good.
                 break
+
 #                exit(0)
 
         pass_acc = accuracy.eval(exe)
         print("pass_id=" + str(pass_id) + " pass_acc=" + str(pass_acc))
-    fluid.io.save_params(exe, dirname='./mnist', main_program=fluid.default_main_program())
+    fluid.io.save_params(
+        exe, dirname='./mnist', main_program=fluid.default_main_program())
     print('train mnist done')
     exit(1)
 
diff --git a/adversarial/mnist_tutorial_fgsm.py b/adversarial/mnist_tutorial_fgsm.py
index 665062afd0e7a3013029269bd9ca5571b2346274..8b29346b8cd7f643771640afc4f783f7544cd071 100644
--- a/adversarial/mnist_tutorial_fgsm.py
+++ b/adversarial/mnist_tutorial_fgsm.py
@@ -9,6 +9,7 @@ import numpy as np
 from advbox.models.paddle import PaddleModel
 from advbox.attacks.gradientsign import GradientSignAttack
 
+
 def cnn_model(img):
     """
     Mnist cnn model
@@ -19,25 +20,22 @@ def cnn_model(img):
     """
     #conv1 = fluid.nets.conv2d()
     conv_pool_1 = fluid.nets.simple_img_conv_pool(
-            input=img,
-            num_filters=20,
-            filter_size=5,
-            pool_size=2,
-            pool_stride=2,
-            act='relu')
+        input=img,
+        num_filters=20,
+        filter_size=5,
+        pool_size=2,
+        pool_stride=2,
+        act='relu')
 
     conv_pool_2 = fluid.nets.simple_img_conv_pool(
-            input=conv_pool_1,
-            num_filters=50,
-            filter_size=5,
-            pool_size=2,
-            pool_stride=2,
-            act='relu')
+        input=conv_pool_1,
+        num_filters=50,
+        filter_size=5,
+        pool_size=2,
+        pool_stride=2,
+        act='relu')
 
-    logits = fluid.layers.fc(
-            input=conv_pool_2,
-            size=10,
-            act='softmax')
+    logits = fluid.layers.fc(input=conv_pool_2, size=10, act='softmax')
     return logits
 
 
@@ -65,22 +63,16 @@ def main():
             paddle.dataset.mnist.train(), buf_size=500),
         batch_size=BATCH_SIZE)
     feeder = fluid.DataFeeder(
-        feed_list=[IMG_NAME, LABEL_NAME], 
-        place=place, 
-        program=fluid.default_main_program()
-    )
+        feed_list=[IMG_NAME, LABEL_NAME],
+        place=place,
+        program=fluid.default_main_program())
 
-    fluid.io.load_params(exe, "./mnist/", main_program=fluid.default_main_program())
+    fluid.io.load_params(
+        exe, "./mnist/", main_program=fluid.default_main_program())
 
     # advbox demo
-    m = PaddleModel(
-        fluid.default_main_program(), 
-        IMG_NAME,
-        LABEL_NAME, 
-        logits.name, 
-        avg_cost.name, 
-        (-1, 1)
-    )
+    m = PaddleModel(fluid.default_main_program(), IMG_NAME, LABEL_NAME,
+                    logits.name, avg_cost.name, (-1, 1))
     att = GradientSignAttack(m)
     for data in train_reader():
         # fgsm attack
@@ -89,6 +81,7 @@ def main():
         plt.show()
         #np.save('adv_img', adv_img)
         break
-    
+
+
 if __name__ == '__main__':
     main()