Merge branch 'master' of git@github.com:snowkylin/TensorFlow-cn.git

Author: snowkylin

source/_static/code/en/basic/example/tensorflow_autograd.py

@@ -17,7 +17,7 @@
 y_pred = a * X_ + b
 loss = tf.constant(0.5) * tf.reduce_sum(tf.square(y_pred - y_))
 
-# 反向传播，利用TensorFlow的梯度下降优化器自动计算并更新变量（模型参数）的梯度
+# Back propagation，calculate and update gradient of varaibles(model parameters) with TensorFlow's GradientDescentOptimier 
 train_op = tf.train.GradientDescentOptimizer(learning_rate=learning_rate_).minimize(loss)
 
 num_epoch = 10000
@@ -26,4 +26,4 @@
     tf.global_variables_initializer().run()
     for e in range(num_epoch):
         sess.run(train_op, feed_dict={X_: X, y_: y, learning_rate_: learning_rate})
-    print(sess.run([a, b]))
+    print(sess.run([a, b]))

source/_static/code/en/basic/example/tensorflow_eager.py

@@ -18,15 +18,15 @@
 num_epoch = 10000
 learning_rate = 1e-3
 for e in range(num_epoch):
-    # 前向传播
+    # Forward propagation
     y_pred = a * X + b
     loss = 0.5 * tf.reduce_sum(tf.square(y_pred - y)) # loss = 0.5 * np.sum(np.square(a * X + b - y))
 
-    # 反向传播，手动计算变量（模型参数）的梯度
+    # Back propagation, calculate gradient of variables(model parameters) manually
     grad_a = tf.reduce_sum((y_pred - y) * X)
     grad_b = tf.reduce_sum(y_pred - y)
 
-    # 更新参数
+    # Update parameters
     a, b = a - learning_rate * grad_a, b - learning_rate * grad_b
 
-print(a, b)
+print(a, b)

source/_static/code/en/basic/example/tensorflow_manual_grad.py

@@ -8,7 +8,7 @@
 
 import tensorflow as tf
 
-# 定义数据流图
+# Define data flow gragh
 learning_rate_ = tf.placeholder(dtype=tf.float32)
 X_ = tf.placeholder(dtype=tf.float32, shape=[5])
 y_ = tf.placeholder(dtype=tf.float32, shape=[5])
@@ -18,26 +18,26 @@
 y_pred = a * X_ + b
 loss = tf.constant(0.5) * tf.reduce_sum(tf.square(y_pred - y_))
 
-# 反向传播，手动计算变量（模型参数）的梯度
+# Back propagation, calculate gradient of variables(model parameters) manually
 grad_a = tf.reduce_sum((y_pred - y_) * X_)
 grad_b = tf.reduce_sum(y_pred - y_)
 
-# 梯度下降法，手动更新参数
+# Gradient descent, update parameters manually
 new_a = a - learning_rate_ * grad_a
 new_b = b - learning_rate_ * grad_b
 update_a = tf.assign(a, new_a)
 update_b = tf.assign(b, new_b)
 
 train_op = [update_a, update_b] 
-# 数据流图定义到此结束
-# 注意，直到目前，我们都没有进行任何实质的数据计算，仅仅是定义了一个数据图
+# End of defining of data flow gragh
+# Attention, until now, we haven't do any actually data calculation, just defined a data flow gragh
 
 num_epoch = 10000
 learning_rate = 1e-3
 with tf.Session() as sess:
-    # 初始化变量a和b
+    # Initialize variables a and b
     tf.global_variables_initializer().run()
-    # 循环将数据送入上面建立的数据流图中进行计算和更新变量
+    # Put data in the data flow gragh created above to calculate and update variables
     for e in range(num_epoch):
         sess.run(train_op, feed_dict={X_: X, y_: y, learning_rate_: learning_rate})
-    print(sess.run([a, b]))
+    print(sess.run([a, b]))

source/_static/code/en/basic/graph/1plus1.py

@@ -1,10 +1,10 @@
 import tensorflow as tf
 
-# 定义一个“计算图”
-a = tf.constant(1)  # 定义一个常量Tensor（张量）
+# Defince a "Computation Graph"
+a = tf.constant(1)  # Defince a constant Tensor
 b = tf.constant(1)
-c = a + b  # 等价于 c = tf.add(a, b)，c是张量a和张量b通过Add这一Operation（操作）所形成的新张量
+c = a + b  # Equal to c = tf.add(a, b)，c is a new Tensor created by Tensor a and Tesor b's add Operation
 
-sess = tf.Session()     # 实例化一个Session（会话）
-c_ = sess.run(c)        # 通过Session的run()方法对计算图里的节点（张量）进行实际的计算
-print(c_)
+sess = tf.Session()     # Initailize a Session
+c_ = sess.run(c)        # Session的run() will do actually computation to the nodes (Tensor) in the Computation Graph
+print(c_)

source/_static/code/en/basic/graph/AmatmulB.py

@@ -1,9 +1,9 @@
 import tensorflow as tf
 
-A = tf.ones(shape=[2, 3])   # tf.ones(shape)定义了一个形状为shape的全1矩阵
+A = tf.ones(shape=[2, 3])   # tf.ones(shape) defines a all one matrix with shape
 B = tf.ones(shape=[3, 2])
 C = tf.matmul(A, B)
 
 sess = tf.Session()
 C_ = sess.run(C)
-print(C_)
+print(C_)

source/_static/code/en/basic/graph/aplusb.py

@@ -1,12 +1,12 @@
 import tensorflow as tf
 
-a = tf.placeholder(dtype=tf.int32)  # 定义一个占位符Tensor
+a = tf.placeholder(dtype=tf.int32)  # Define a placeholder Tensor
 b = tf.placeholder(dtype=tf.int32)
 c = a + b
 
-a_ = input("a = ")  # 从终端读入一个整数并放入变量a_
+a_ = input("a = ")  # Read an Integer from terminal and put it into a_
 b_ = input("b = ")
 
 sess = tf.Session()
-c_ = sess.run(c, feed_dict={a: a_, b: b_})  # feed_dict参数传入为了计算c所需要的张量的值
-print("a + b = %d" % c_)
+c_ = sess.run(c, feed_dict={a: a_, b: b_})  # feed_dict will input Tensors' value needed by computing c
+print("a + b = %d" % c_)

source/_static/code/en/basic/graph/variable.py

@@ -1,13 +1,13 @@
 import tensorflow as tf
 
 a = tf.get_variable(name='a', shape=[])
-initializer = tf.assign(a, 0)   # tf.assign(x, y)返回一个“将张量y的值赋给变量x”的操作
-a_plus_1 = a + 1    # 等价于 a + tf.constant(1)
+initializer = tf.assign(a, 0)   # tf.assign(x, y) will return a operation “assign Tensor y's value to Tensor x”
+a_plus_1 = a + 1    # Equal to a + tf.constant(1)
 plus_one_op = tf.assign(a, a_plus_1)
 
 sess = tf.Session()
 sess.run(initializer)
 for i in range(5):
-    sess.run(plus_one_op)                   # 对变量a执行加一操作
-    a_ = sess.run(a)                        # 获得变量a的值并存入a_
+    sess.run(plus_one_op)                   # Do plus one operation to a
+    a_ = sess.run(a)                        # Calculate a‘s value and put the result to a_
     print(a_)

source/_static/code/en/basic/graph/variable_with_initializer.py

@@ -1,11 +1,11 @@
 import tensorflow as tf
 
-a = tf.get_variable(name='a', shape=[], initializer=tf.zeros_initializer)   # 指定初始化器为全0初始化
+a = tf.get_variable(name='a', shape=[], initializer=tf.zeros_initializer)   # Made initializer as a all zero initializer
 a_plus_1 = a + 1
 plus_one_op = tf.assign(a, a_plus_1)
 
 sess = tf.Session()
-sess.run(tf.global_variables_initializer()) # 初始化所有变量
+sess.run(tf.global_variables_initializer()) # Initailize all the 
 for i in range(5):
     sess.run(plus_one_op)
     a_ = sess.run(a)

source/en/models.rst

@@ -113,7 +113,7 @@ The specific implementation is as follows, which is very similar to MLP except s
     Figure of the CNN structure
 
 .. literalinclude:: ../_static/code/en/model/cnn/cnn.py
-    :lines: 4-37
+    :lines: 4-38
 
 By substituting ``model = MLP()`` in the last chapter with ``model = CNN()``, we get the following output::
 

source/zh/models.rst

@@ -113,7 +113,7 @@ TensorFlow模型
     CNN结构图示
 
 .. literalinclude:: ../_static/code/zh/model/cnn/cnn.py
-    :lines: 4-37
+    :lines: 4-38
 
 将前节的 ``model = MLP()`` 更换成 ``model = CNN()`` ，输出如下::