update LeNet-5

02.recognize_digits/README.en.md

@@ -172,7 +172,7 @@ def convolutional_neural_network(img):
         num_channel=1,
         pool_size=2,
         pool_stride=2,
-        act=paddle.activation.Tanh())
+        act=paddle.activation.Relu())
 
     conv_pool_2 = paddle.networks.simple_img_conv_pool(
         input=conv_pool_1,
@@ -181,13 +181,9 @@ def convolutional_neural_network(img):
         num_channel=20,
         pool_size=2,
         pool_stride=2,
-        act=paddle.activation.Tanh())
+        act=paddle.activation.Relu())
 
-    fc1 = paddle.layer.fc(input=conv_pool_2,
-                          size=128,
-                          act=paddle.activation.Tanh())
-
-    predict = paddle.layer.fc(input=fc1,
+    predict = paddle.layer.fc(input=conv_pool_2,
                               size=10,
                               act=paddle.activation.Softmax())
     return predict
@@ -203,9 +199,9 @@ images = paddle.layer.data(
 label = paddle.layer.data(
     name='label', type=paddle.data_type.integer_value(10))
 
-predict = softmax_regression(images)
-#predict = multilayer_perceptron(images) # uncomment for MLP
-#predict = convolutional_neural_network(images) # uncomment for LeNet5
+# predict = softmax_regression(images)
+# predict = multilayer_perceptron(images) # uncomment for MLP
+predict = convolutional_neural_network(images) # uncomment for LeNet5
 
 cost = paddle.layer.classification_cost(input=predict, label=label)
 ```

02.recognize_digits/README.md

@@ -173,7 +173,7 @@ def convolutional_neural_network(img):
         num_channel=1,
         pool_size=2,
         pool_stride=2,
-        act=paddle.activation.Tanh())
+        act=paddle.activation.Relu())
     # 第二个卷积-池化层
     conv_pool_2 = paddle.networks.simple_img_conv_pool(
         input=conv_pool_1,
@@ -182,13 +182,9 @@ def convolutional_neural_network(img):
         num_channel=20,
         pool_size=2,
         pool_stride=2,
-        act=paddle.activation.Tanh())
-    # 全连接层
-    fc1 = paddle.layer.fc(input=conv_pool_2,
-                          size=128,
-                          act=paddle.activation.Tanh())
+        act=paddle.activation.Relu())
     # 以softmax为激活函数的全连接输出层，输出层的大小必须为数字的个数10
-    predict = paddle.layer.fc(input=fc1,
+    predict = paddle.layer.fc(input=conv_pool_2,
                               size=10,
                               act=paddle.activation.Softmax())
     return predict
@@ -205,9 +201,9 @@ images = paddle.layer.data(
 label = paddle.layer.data(
     name='label', type=paddle.data_type.integer_value(10))
 
-predict = softmax_regression(images) # Softmax回归
-#predict = multilayer_perceptron(images) #多层感知器
-#predict = convolutional_neural_network(images) #LeNet5卷积神经网络
+# predict = softmax_regression(images) # Softmax回归
+# predict = multilayer_perceptron(images) #多层感知器
+predict = convolutional_neural_network(images) #LeNet5卷积神经网络
 
 cost = paddle.layer.classification_cost(input=predict, label=label)
 ```

02.recognize_digits/index.en.html

@@ -214,7 +214,7 @@ def convolutional_neural_network(img):
         num_channel=1,
         pool_size=2,
         pool_stride=2,
-        act=paddle.activation.Tanh())
+        act=paddle.activation.Relu())
 
     conv_pool_2 = paddle.networks.simple_img_conv_pool(
         input=conv_pool_1,
@@ -223,13 +223,9 @@ def convolutional_neural_network(img):
         num_channel=20,
         pool_size=2,
         pool_stride=2,
-        act=paddle.activation.Tanh())
+        act=paddle.activation.Relu())
 
-    fc1 = paddle.layer.fc(input=conv_pool_2,
-                          size=128,
-                          act=paddle.activation.Tanh())
-
-    predict = paddle.layer.fc(input=fc1,
+    predict = paddle.layer.fc(input=conv_pool_2,
                               size=10,
                               act=paddle.activation.Softmax())
     return predict
@@ -245,9 +241,9 @@ images = paddle.layer.data(
 label = paddle.layer.data(
     name='label', type=paddle.data_type.integer_value(10))
 
-predict = softmax_regression(images)
-#predict = multilayer_perceptron(images) # uncomment for MLP
-#predict = convolutional_neural_network(images) # uncomment for LeNet5
+# predict = softmax_regression(images)
+# predict = multilayer_perceptron(images) # uncomment for MLP
+predict = convolutional_neural_network(images) # uncomment for LeNet5
 
 cost = paddle.layer.classification_cost(input=predict, label=label)
 ```

02.recognize_digits/index.html

@@ -215,7 +215,7 @@ def convolutional_neural_network(img):
         num_channel=1,
         pool_size=2,
         pool_stride=2,
-        act=paddle.activation.Tanh())
+        act=paddle.activation.Relu())
     # 第二个卷积-池化层
     conv_pool_2 = paddle.networks.simple_img_conv_pool(
         input=conv_pool_1,
@@ -224,13 +224,9 @@ def convolutional_neural_network(img):
         num_channel=20,
         pool_size=2,
         pool_stride=2,
-        act=paddle.activation.Tanh())
-    # 全连接层
-    fc1 = paddle.layer.fc(input=conv_pool_2,
-                          size=128,
-                          act=paddle.activation.Tanh())
+        act=paddle.activation.Relu())
     # 以softmax为激活函数的全连接输出层，输出层的大小必须为数字的个数10
-    predict = paddle.layer.fc(input=fc1,
+    predict = paddle.layer.fc(input=conv_pool_2,
                               size=10,
                               act=paddle.activation.Softmax())
     return predict
@@ -247,9 +243,9 @@ images = paddle.layer.data(
 label = paddle.layer.data(
     name='label', type=paddle.data_type.integer_value(10))
 
-predict = softmax_regression(images) # Softmax回归
-#predict = multilayer_perceptron(images) #多层感知器
-#predict = convolutional_neural_network(images) #LeNet5卷积神经网络
+# predict = softmax_regression(images) # Softmax回归
+# predict = multilayer_perceptron(images) #多层感知器
+predict = convolutional_neural_network(images) #LeNet5卷积神经网络
 
 cost = paddle.layer.classification_cost(input=predict, label=label)
 ```

02.recognize_digits/train.py

@@ -29,7 +29,7 @@ def convolutional_neural_network(img):
         num_channel=1,
         pool_size=2,
         pool_stride=2,
-        act=paddle.activation.Tanh())
+        act=paddle.activation.Relu())
     # second conv layer
     conv_pool_2 = paddle.networks.simple_img_conv_pool(
         input=conv_pool_1,
@@ -38,14 +38,10 @@ def convolutional_neural_network(img):
         num_channel=20,
         pool_size=2,
         pool_stride=2,
-        act=paddle.activation.Tanh())
-    # The first fully-connected layer
-    fc1 = paddle.layer.fc(
-        input=conv_pool_2, size=128, act=paddle.activation.Tanh())
-    # The softmax layer, note that the hidden size should be 10,
-    # which is the number of unique digits
+        act=paddle.activation.Relu())
+    # fully-connected layer
     predict = paddle.layer.fc(
-        input=fc1, size=10, act=paddle.activation.Softmax())
+        input=conv_pool_2, size=10, act=paddle.activation.Softmax())
     return predict
 
 
@@ -58,9 +54,9 @@ label = paddle.layer.data(name='label', type=paddle.data_type.integer_value(10))
 
 # Here we can build the prediction network in different ways. Please
 # choose one by uncomment corresponding line.
-predict = softmax_regression(images)
-#predict = multilayer_perceptron(images)
-#predict = convolutional_neural_network(images)
+# predict = softmax_regression(images)
+# predict = multilayer_perceptron(images)
+predict = convolutional_neural_network(images)
 
 cost = paddle.layer.classification_cost(input=predict, label=label)